systematic literature review on opinion mining of big data for … · 2018-03-18 · systems....

6 http://www.webology.org/2017/v14n2/a156.pdf

Webology, Volume 14, Number 2, December, 2017

Home Table of Contents Titles & Subject Index Authors Index

Systematic Literature Review on Opinion Mining of Big Data for

Government Intelligence

Akshi Kumar

Department of Computer Science & Engineering, Delhi Technological University, Delhi-110042,

India. E-mail: [email protected]

Abhilasha Sharma

Department of Computer Science & Engineering, Delhi Technological University, Delhi-110042,

India. E-mail: [email protected]

Received August 21, 2017; Accepted December 25, 2017

Abstract

With the advent of new technology paradigm, SMAC (Social media, Mobile, Analytics and

Cloud) the information network generates an infinite ocean of data spreading faster and larger

than earlier. A high quality information extracted from this massive volume of data, named as big

data, urges the development of an efficient and effective decision support system and powerful

strategic tools in the area of government intelligence. This pool of information can be explored

for the benefit of an organization/system and to better understand its stakeholder needs by

collecting, mining opinions about every point or subject of interest. Digitally intelligent and

smart governance has been identified as a dynamic field with new studies being reported at

various research avenues. The need to review, analyze and evaluate research studies across

literature is thus fostered motivating us to identify existing trends, research gaps and potential

directions of future work within this domain. This paper intends to provide a systematic literature

review within the promising area of opinion mining and its application to the area of government

intelligence.

Keywords

Opinion mining; Big data; Digital governance; Systematic literature review

http://www.webology.org/index.html

http://www.webology.org/2017/v14n2/toc.html

http://www.webology.org/titleindex.html

http://www.webology.org/authorindex.html


Introduction

Optimal processing and analytics of big data enables to better decision making and strategic

movement in relevant area. Different real time industry verticals are expanding their existing

data sets with data available from social web, i.e. big data to get a better understanding of their

customers' behavior and preferences. And this can be achieved by the means of opinion mining

which aims to extract people opinion from web. Different sources of big data provide a massive

amount of valuable information which can be examine and analyze for extracting sentiments.

The major sources of big data are: black box data, transportation data, stock exchange data,

social media data, power grid data, search engine data, healthcare data, etc. (Arputhamary &

Jayapriya, 2016). Opinion mining on big data has emerged as a notable direction of research with

scientific trials and promising applications being explored substantially. It has turned out as an

exciting new trend with a gamut of practical applications that range from Business and Trades;

Recommender Systems; Expert Finding; Politics; Government Sector; Healthcare Industry;

amongst others.

The terminology "opinion mining" was first observed in 2003 in a research paper by Dave et al.

By 2006 it started to be involved in web applications and various other domain like marketing

for product and service reviews, entertainment, politics, recommendations, business etc. which in

turn broaden its application spectra. With the rapid increase of people participation and

communication over social web where they express their ideas, opinion over a public forum;

opinion mining extracts the public mindset and the findings are being used in various domains.

As per literature so far, various application areas of opinion mining can be broadly classified as:

business intelligence (BI), government intelligence (GI), information security and analysis (ISA),

market intelligence (MI), sub component technology (SCT) and smart society services (SSS)

whereas the techniques used are machine learning (ML), lexicon based (LB), hybrid and concept

based. But no study has been conducted on reviewing of opinion mining implementation in

various application areas till now. However, few surveys are there which covers the existing

opinion mining techniques, opinion mining task classification, opinion mining evolution, opinion

mining and big data etc., but none of them emphasizes over its application areas and correlates

them with its techniques.

Specifically, to improve and optimize decisions and performance of government, the thrust area

is to explore and understand opinion mining, its applications and techniques using big data in

government intelligence. The idea is to envision “digital governance” by virtue of social web

adoption (social media data; source of big data), where government can take advantage of social

platforms involving huge user participation. The motivation is clearly based on the “Digital India

Initiative” recently proposed and projected by the Government of India. The initiative motivates

to embed the concept of S-governance which includes Social and Sentiment governance (Kumar

& Sharma, 2016) combines to form Smart governance in order to transform the nation into


digitally empowered knowledge economy and includes projects that aim to ensure that

government services are available to citizens electronically/digitally and people get benefit of the

latest information and communication technology (Makeinindia, n.d.; Digital India, 2014).

Hence, the requirement is to go through a state-of-art literature survey and review the significant

research on the subject of opinion mining of big data for digital governance. In this review, we

explicate how the voluminous amount of data i.e. big data is utilized for analyzing public

posts/comments/tweets/blogs/discussions by the application of opinion mining in government

intelligence.

Induction of Internet and communication technologies transforms the conventional governance

in to digital governance and its existence is visible in the year 2010 (Janowski, 2015). Therefore,

the proposed survey is an exhaustive study of the published articles of opinion mining

applications from 2011-2017 with the usage of various datasets (big data) in the field of digital

governance and its applications. The comparative analysis of different techniques of opinion

mining in its various application areas has been done on the basis of original studies. This paper

conducts a systematic literature survey which outlines the application areas of opinion mining,

discusses the introduction of opinion mining in digital governance and summarizes opinion

mining based digital governance progress graph so far.

The process for carrying out a systematic literature review (SLR) is given by Kitchenham et al.

in 2009. However, in this SLR we reshape the core architecture of the phases and represent them

in a more structured and sequential manner. This makes the SLR more systematic, well classified

and organized, for a better understanding and course of action. The rest of the paper is organized

as follows: Section 2 elaborates the three terms, i.e., opinion mining, big data and digital

governance individually and their influence on each other when coupled together. Section 3

describes the purpose of required phases to be performed during the conduct of a Systematic

Literature Review (SLR). Section 4 details the Review Planning phase (structured in a new

template) for the proposed SLR which contains the motivation and aim of the research, to gather

and analyze the relevant primary studies of research, its quality assessment/verification and

documentation. The next phase Review Conduct is elaborated in Section 5 which contains

searching strategy, literature review of selected studies in visual and tabulated form. Section 6

deals with the Review Reporting phase which document the results and discussion of complete

review. Finally, the paper winds up with a conclusion and future works in the last section.

Opinion Mining of Big Data for Digital Governance

With the progression of Web 2.0 and its plug-ins such as social networks, micro blogs, mobile

technologies, audio, video, web searches, discussion forums, posts, comments and several other

social media, citizens can vocalize their opinion and experiences in a more effective manner.


Digital empowerment of citizens plays a vital role in providing people centered applications by

transforming the conventional IT paradigm into next generation paradigm of SMAC based

systems. Social media and Mobile generates a voluminous amount of data i.e. big data, which

can be exploited by government for the extraction of public opinion to improvise and enhance

the process of decision making using techniques of opinion mining. Hence, opinion mining of

big data in digital governance plays a significant role for the success of government programs

and initiatives. The three basic terminologies namely, opinion mining, big data and digital

governance are briefly discussed as follows:

Opinion mining (OM), deals with opinion based natural language processing which includes

genre distinctions, emotion and mood recognition, ranking, relevance computations, perspectives

in text, text source identification and opinion oriented summarization (Kumar & Teeja, 2012).

Opinion Mining is "a field of study that tends to use the natural language processing techniques

to extract, capture or identify the attitude (otherwise sentiment) of a person with respect to a

particular subject. It is the automated mining of attitudes, opinions, and emotions from text,

speech, and database sources through NLP" (Kumar & Sharma, 2016).

Sentiment analysis (also called opinion mining, review mining or appraisal extraction, attitude

analysis) is the "task of detecting, extracting and classifying opinions, sentiments and attitudes

concerning different topics, as expressed in textual input" (Ravi & Ravi, 2015).

Big Data, a huge amount of data which is full of conversations, social networking sites, online

discussion forums, web based audio/video presentations, micro blogs, commercial transactions,

web logs, comments, pictures, documents, etc. It is popping up from numerous sources. Due to

its complex structure/nature, traditional database architectures are unable to collect/capture,

contemplate and conceptualize it for further processing. For productive and proper use of big

data assets it is necessary to understand its key attributes. Seven (Vs) characteristics (Mcnulty,

2014) of big data are volume, variety, velocity, veracity, variability, visualization and value.


Figure 1. V- Model of People Participation in Governance (The Digital Governance Initiative, n.d.)

Digital Governance

Governance is all about the action, manner or power of governing to ensure the environment of

an organization is stable, transparent, responsive and accountable and working in best interest of

the public. It includes mechanism of policy establishment, formulation of decision making

process and implementation, authority balancing and performance monitoring in order to manage

public affair.

At the same time, "good governance is to establish, consensus amongst the stakeholders of a

country with the goal to improve quality of life enjoyed by all citizens" (Kumar & Sharma,

2016). Good governance follows the rule of law and aims for efficient processes and effective

outcomes. The key attributes of good governance provided by the commission of human rights

(United Nations Human Rights, Office of the High Commissioner, n.d.) in a resolution 2000/64

are Transparency, Responsibility, Accountability, Participation and Responsiveness. Good

governance is about taking decisions based on information set or knowledge set. "Digitization of

this information is required for its easy access available on network which open arms to all

individuals of every community or village for use of this information - paving the way for Digital

Governance" (The Digital Governance Initiative, n.d.).

Representative Mode Individual /

Collective

In-situ Domain Ex-situ

Passive /

Reactive Approach

Pro-active

/Interactive

Indirect / Delayed Direct/Immediate Impact

Conventional

Governance Model

People

Participation

Digital

Governance Model


The intent of digital governance is to ensure that common citizens should associated with

decision-making processes as it affects them directly or indirectly, which in turn improves their

conditions and the quality of lives (Nath, 2003). This new facet of governance will assure that

citizens are active contributor in deciding the kind of services they want as well as attentive

consumer of services offered to them. And this is what required for a powerful democratic

country.

The perspective and participation of people changes with the transformation of governance

model from conventional to digital is represented in Figure 1. This comparison states that in

contrast to conventional governance, digital governance allows a more closer contact of public

and decision-makers in the government and results in greater access and check over governance

mechanism in order to constitute a more responsive, transparent, responsible, accountable and

efficient governance" (The Digital Governance Initiative, n.d.).

Opinion Mining of Big Data for Digital Governance results in a more stabilized and smart

governance which analyze data available on social web and extract public sentiments for the

process of decision making. The evolution of web based digital governance is depicted in figure

2 which also exhibits the parallel mapping of opinion mining, big data and digital governance

with the versions of web.

Web 1.0 is the web of cognition and primarily read-only, Web 2.0 is the web of people

corresponds to communication and Web 3.0 has become the web of data which stands for co-

operation (Kumar & Sharma, 2015). With the emergence of Web 3.0, the concept of smart and

socio-sentimental governance is readily apparent, which transforms the data into decision. With

the evolution of web, the social media model for digital governance is also emerging to utilize

the high volume of unstructured data (Kumar & Joshi, 2017). Although, there are no fixed model

for digital governance. Rather, they are customized by investigating and research, conducive to

cater the needs of a nation or society. Some generic models being practiced are: critical flow

model, comparative analysis model, broadcasting model, e-advocacy model and interactive

service model (Nath, 2003).

The idea is to look for models that enrich the big data analytic solution architecture, tools and

technologies for creating awareness and opportunities that make big data and its analytics in

government domain. Amongst all, e-advocacy and lobbying is most frequently used model as it

motivates the personal and people participation in decision making process of organizations and

government officials. Some of the projects based on e-advocacy model are: Greenpeace Cyber-

Activist Community, Independent Media Centre, Panchayats, Drop the Debt Campaign, PRS

Legislative Research, IGC Internet etc. (The Digital Governance Initiative, n.d.), (Nath, 2003).

"Advocacy is an activity by an individual or group which aims to influence decisions within

political, economic, and social systems and institutions. Advocacy can include many activities

https://en.wikipedia.org/wiki/Advocacy_group

https://en.wikipedia.org/wiki/Decision_making


that a person or organization undertakes including media campaigns, public speaking,

commissioning and publishing research or conducting exit poll or the filing of an amicus brief."

(Wikipedia contributors, 2017).

The purpose of advocacy is to notify information and communicate facts to politicians and to

their deputies or officers about emotional and mental constitution of public mind and its

pertinence in governance. Legislators work for public welfare and they want views, outlook,

facts, and insights to figure out realism. Now here public participation is critical and their

reaction, inferences, viewpoint shared on social media could be mined (by the use of opinion

mining) in order to provide the information/news required by officials/lawmakers. Therefore,

inducement of opinion mining in advocacy will go a long way and continue a long innings in the

scope of advocacy.

Figure 2. Evolution of Web Based Digital Governance

Systematic Review Phases

The Systematic Literature Review(SLR) comprise of three phases represented in figure 3 and its

outline is planned, executed and organized as discussed by Kitchenham and Charters

(Kitchenham et al., 2009). However, the core architecture of the phases is restructured to make

them more classified and understandable.

Review Planning, being the first phase of SLR is crucial to identify and capture the existing and

relevant research work done in the area. The information gathered is further examined and

evaluated for the development of SLRS (Systematic Literature Review Specification) document.

The second phase is Review Conduct, which provides a collective and comprehensive review of

existing literature. Review Reporting is the last phase of SLR in which various findings,

observations and results are inspected and reported.

https://en.wikipedia.org/wiki/Media_campaign

https://en.wikipedia.org/wiki/Public_speaking

https://en.wikipedia.org/wiki/Exit_poll

https://en.wikipedia.org/wiki/Amicus_curiae


Review Planning

Review planning is the most important and fundamental stage of SLR. It leads

o To explore and understand the motivation behind the survey; and

o To make out the need of review.

Next step is to define research objectives with the help of questions as per the aim of SLR. A

well formed/designed question can help in taking key decisions related to research. Thereafter,

all the possible and relevant reviews/research work are captured/extracted for the development of

review protocol which includes following steps: Review Elicitation, Review Analysis, Review

Validation and Review Documentation. Detailed review planning phase is represented in figure 4

and discussed below.

Figure 3. Systematic Review Stages

Identify the Need

A huge amount of work has been done in the area of opinion mining of big data. Social web

statistics reveals that more than

o 3 billion people across world are online and use Internet for exchanging information and

ideas, sharing, searching facts, publishing amongst others; and

o 2 billion people worldwide maintaining social accounts (Twitter, Facebook, etc.) (Kumar &

Sharma, 2016).


Gamut of social networking sites has offered a platform for citizens to publically vocalize their

ideas in distinct areas which increases the significance of any issue/discussion very swiftly.

Hence, the necessity of incorporating public opinion and interest in the process of decision

making has grown exponentially throughout the years. And the fact that it is vitally important in

governance has been observed across literature studies.

Opinion mining is the key underpin of synergetic policymaking. It helps in listening to the

problem rather than asking it and hence ensures a detailed and precise speculation/impression of

verity (Osimo & Mureddu, 2012). It helps in concluding thousands of inferences by considering

various opinion, ideas, discussions and feedback from citizens and covers many spheres like

advisory, decision making, advocacy, early warning system detection etc.

The serviceability of collective and interactive aspect of social web by extracting and analyzing

sentiments can proffer a significant, incomparable platform for enormous association of citizens

in governance measures in order to make it smart and intelligent.

Analytics on the unstructured data can transform the governance model into a social governance

model, as an integration of social media and S-governance. In view of this, the need is to further

research and study this area to gain insights into public opinion by mining user generated data on

social web for empowering digital governance.


Figure 4. Review Planning Activities


Identify Research Questions

This review paper aims to explore opinion mining techniques in the field of advocacy in order to

develop, deepen and broaden the understanding of S-governance. Research questions (RQs) have

been formulated according to ensure complete coverage of existing research on the topic and the

scope of opinion mining to evolve in the area of advocacy. Table 1 enlists the research questions

identified for this SLR.

Table 1. Identified Research Questions

RQ# Research Questions

RQ1 What are the different application areas and techniques of

opinion mining?

RQ2 How much work has been done in the area of government

intelligence using opinion mining?

RQ3 Which (big) data sets are used for opinion mining in government

intelligence?

RQ4 Which techniques of opinion mining have been used in advocacy

so far?

Develop Review Protocol

This phase takes input as research questions, processes them with various sequential steps and

finally generates output as Systematic Literature Review Specification (SLRS) document. We

divide the activity of development of review protocol into four steps: Review Elicitation, Review

Analysis, Review Validation and Review Documentation as represented in Figure 5.


Research Question(s)

SLRS

Figure 5. Steps of Developing Review Protocol

Review Elicitation

Decomposition of Research Questions into individual and small units has been done for the

formation of search terms. Following were the search terms identified: opinion mining, big data,

and opinion mining applications. Identified search terms then get refined by incorporating

alternative terms and synonyms.

These equivalent terms were: sentiment analysis, sentiment extraction, and social web data.

Different permutation and combination of basic and refined terms was then applied with OR,

AND logistics for search of relevant studies.

Search of Available Primary Studies has been done considering the following digital four

libraries: Science Direct, Springer, ACM Digital Library and IEEE Transactions. The search

terms were explored in titles, keywords and abstracts. This step results all the available primary

studies put together in contemplation of answering research questions.

Review Analysis

Inclusion/Exclusion (I/E) Criteria is based to limit the scope of research to find fitting answers to

the RQs. The collected primary studies were input to this step and inclusion/exclusion criteria

was incorporated to select or reject the studies based on their relevancy.

Develop

Review

Protocol

Review Elicitation

Review Analysis

Review Documentation

Review Validation


Inclusion Criteria

o Journal Papers of Science Direct, Springer, ACM Digital Library and IEEE

o Papers from year 2011-2017

o Opinion mining and its applications

o Techniques of opinion mining

o Secondary studies of opinion mining

o Survey of opinion mining techniques in its various application areas

Exclusion Criteria

o Conference papers of: Science Direct, Springer, ACM Digital Library and IEEE

o Non textual representation of opinions/sentiments

o Big data process optimization and its survey

o Task classification of opinion mining.

Classification of Papers has been done based on the above mentioned inclusion/exclusion

criteria. Owing to the theoretical limitations and feasibility, we considered only journal papers

from selective digital databases, thus limiting the scope of our study. Table 2 depicts the process

of paper classification in different phases. Total 194 papers were selected as the outcome of this

step.

Table 2. Paper Classification Stats

Elicitation Phase Selected Papers Analysis Phase Selected Papers

Science Direct

Springer Journals

ACM Digital Library

IEEE Transactions

240 primary papers

Inclusion/exclusion

criteria (40)

Redundant papers (4) 194 final papers

Review Validation

Data Extraction and Synthesis accomplishes the appropriate analysis from the 257 final studies

for the extraction of relevant information required to answer the research queries. Selected

papers were referred for the extraction of information, such as year of publication, contributing

author(s), application areas, existing techniques, datasets used for assessment of performance,

etc. All such information is tabulated in order to get simplify the data synthesis process for

review. Data synthesis process summarizes the extracted information in the form of tables

(representation of categorized information) and diagrams like graphs and charts (representation

of measurable information). Table 3 and 6; Figure 6, 7, 8, 9, 10 and 11 were taken as output of

this step.


Review Documentation

Review Documentation results in a document call up as Systematic Literature Review

Specification (SLRS) document. It is a detailed description of relevant studies and their

systematic compilation and presentation in order to answer the research objectives in the

projected research area. Section 6 covers this step.

Review Conduct

Review conduct executes the proposal offered in planning phase. The actual process

implementation of studies elicitation and their analysis based on inclusion/exclusion criteria is

performed here by employing Search Strategy. The subsequent step is to explicate the

comprehensive Literature Survey by incorporating the diagrams, tabulations and answers of the

identified research questions.

Search Strategy

In this step, we see how much work is ongoing in this field. An extensive search of relevant

primary studies is done to capture the mounting interest. The chart in Figure 7 represents the year

wise number of papers published in different application areas, reflecting the attention. This area

is gaining from researchers. Some of the popular journals and conference in which the primary

studies have been published is listed in Table 3.

Figure 7. Year wise studies distribution


Table 3. Some popular journals of primary studies of this SLR

Publication Name Type

Decision Support Systems Journal

Expert Systems with Applications Journal

Knowledge-Based Systems Journal

Computers & Security Journal

IEEE Transactions on cybernetics Journal

Procedia Computer Science Journal

IEEE Intelligent Systems Journal

Information Processing &Management Journal

Applied Intelligence Journal

Quality & Quantity Journal

IEEE Transactions on Knowledge and Data Engineering Journal

ICT Express Journal

Applied Soft Computing Journal

Procedia Engineering Journal

Journal of Intelligent Information Systems Journal

Procedia Computer Science Journal

Computers in Human Behavior Journal

IEEE Transactions on Multimedia Journal

Literature Survey

In this phase, a state-of-art literature survey is carried to review the substantial research in the

area of opinion mining and its application area. The preliminary study from pertinent literature is

summarized in Table 4.

In the area of business intelligence, Tsai et al. (2011) proposed an algorithm for novel business

blogs detection by applying machine learning techniques. Chung et al. (2012) developed a new

framework based on ML techniques for extracting relation of customer rating and reviews. Jang

et al. (2013) proposed a method for reviewers' attitude extraction and their traits identification.

Zitnik et al. (2012) used ML techniques to enhance business process operations. In 2013, Pai et

al. proposed a ML based method to aid customers in better decision making and to support

organization for better understanding of product/service appraisals. Wu et al. (2014) and Li et al.

(2014) used ML techniques for stock markets prediction of automatic trading decisions and

market movements.

In the area of government intelligence, Cao et al., Katakis et al., Mohammad et al. in years 2011,

2014 and 2015 respectively worked in the field of voting advisory with use of machine learning

techniques. Tools are developed for providing voting advice to application users and political

parties are also facilitated by social media sites by making public opinion available to them.

Automation of manual processes has been done.


Panasyuk et al. in 2014, Cheng et al., Adedoyin-Olowe et al. and Gull et al. in 2016, worked in

the area of politics using machine learning techniques. Various requirements like public opinion

analysis system, identification of conflicting communities, trend prediction of elections etc. have

been fulfilled. Haselmeyer et al. in 2016 contributed in politics domain with the use of lexicon

based techniques. Objectives were to find sentiment polarity of user comments to provide

political portal recommendation and usage of sentiment dictionary in negative campaigning of

parties. De Fortuny et al. in 2012, Xianghua et al. in 2013, Makazhanov et al. in 2014, Kagan et

al. in 2015 use hybrid techniques in politics. They emphasizes over the automation of manual

process of opinion analysis over social communication channels and how political preference

prediction has been done on Twitter. In 2014, Rill et al. used concept level ontology based

technique to conclude that twitter detection of any topic is much faster than other communication

channels.

O'Leary, D. E in 2016 and Singh et al. in 2017 worked in the area of advocacy and policy making

using machine learning techniques. They aim to analyze public sentiments against government

policy. ElTayeby et al. in 2014 and Tayal et al. in 2016 applied machine learning techniques in

campaigning and developed tools to estimate their success. Also, clustering of similar opinions

has been performed to study the impact of media.

In the area of information and security analysis, Bhargava et al. in 2016 used lexicon based

techniques for opinion summarization. Framework has been proposed for entity based topic

oriented summarization. Proposed graph based technique summarizes redundant opinions.

He and Zhou, Liu et al. and Fan et al. in 2011, Zhai et al., Balahur et al., Cruz et al., Lane et al.

and Wang et al. in 2012, Bagheri et al., and Haddi et al. in 2013, Wu et al., Kansal and

Toshniwal, Dařena et al., Zheng et al., and Dehkharghani et al. in 2014, Wang and Liu, Kurian

and Asokan in 2015, Li et al. and Qiang et al. in 2016 and Rout et al. in 2017 applied various

machine learning algorithms for analyzing information like in summarization systems and their

automation and in security analysis like spam detection etc.

In the sub component technology domain, Rosa et al. in 2015 used lexicon based techniques. The

studies done by them presents recommendation systems for restaurant reviews and music. In

2016, Lei et al. and Dong et al. and in 2017, X Ma et al. applied hybrid techniques to develop

recommender system for products and rating prediction. Qiao et al. and Bridge and Healy in

2011, Anooj in 2012, Kardan and Ebrahimi, Geva and Zahavi, García-Cumbreras et al., Chen

and Wang, Peleja et al., Scharkow in 2013, Sun et al. in 2014, Capdevila et al., Coussement et

al., Colace et al., Wei et al., Alahmadi and Zeng, Sohail et al. in 2015, Pröllochs et al., Li et al.,

Zhang et al., and Bilici & Saygın in 2016 and Gurini et al., Ma et al. in 2017 used machine

learning techniques in citation analysis, recommender systems etc.


In the area of smart society services, Zakzouk and Mathkour, Cambria et al. in 2012, Valsamidis

et al., Huh et al., Lu in 2013, Yan et al., de Oña et al., Bogdanova et al., Chen et al., Cheng et al.,

Cao et al., Wiley et al., Küçük et al. in 2014, İskender & Batı, Aliandu, Bucur, Ji et al., Bobicev

et al. in 2015, Lv & El-Gohary, Hu et al., Bui et al., Rossetti et al., Yang et al., Lee et al., in 2016

and James et al., Kuflik et al., Du et al., Dai & Hao in 2017 worked over machine learning

techniques. Muller et al. and Ali et al. in 2015 make use of ontology in concept level technique

for the automation of hotel reservation system and enhancing the data related to transportation by

mining social media. Hybrid techniques were used by Cameron et al., in 2013, Nguyen et al. and

Ortigosa et al. in 2014, Shen & Kuo in 2015, Asghar et al. and Ali et al., Xue et al.in

2017.Various areas such as drug epidemiology, online depression community analysis, tourism

monitoring, e-learning in education, health related issues have been considered for analysis. A

significant amount of work has been reported in this area using lexicon based techniques i.e.

Moreo et al. in 2012, Mostafa in 2013, Lei et al. in 2014, Teodorescu, Yu & Wang in 2015 and

Neidhardt et al. in 2017.

Marketing intelligence is one of the widely used application area of opinion mining. Various

researchers have applied machine learning techniques in review analysis of products and

services, reputation monitoring, placements of ads etc. Ghose & Ipeirotis, Sanchez-Monzon et

al., Fan et al., Park & Lee, Xu et al., Bai, Chen & Tseng in 2011, Yu et al., Liu et al., Chen et al.,

Schumaker et al., Min & Park, Reyes & Rosso, Esparza et al., Kang et al. in 2012, Wöllmer et

al., Bafna & Toshniwal, Liu et al., Yaakub et al., Basari et al., Yu et al. in 2013, Wu et al., Zhang

et al., Jiang et al., Vinodhini & Chandrasekaran, 2014a, 2014b, Yang & Chao, Wang et al., Zhang

et al., Smailović et al. in 2014, Zhang et al., Parkhe & Biswas, Bastı et al., He et al., Sandoval &

Hernández, Zimmermann et al., D’Avanzo & Pilato, Amarouche et al., Nguyen et al., Fernández-

Martínez et al. in 2015, Yan et al., Li et al., Vinodhini & Chandrasekaran, Cosma & Acampora,

Hur et al., Luo et al., Chua & Banerjee, Jin et al., Zhang et al., Malik & Shakshuki, Djouvas et

al., Lee et al., Khan et al., Manaman et al., Nayak et al., Chen et al., Feuerriegel & Prendinger,

Anand & Naorem in 2016 and Wang, Yuan et al., Tsirakis et al., Manek et al., Gui et al., Tsai &

Wang, Chan & Chong, Ding et al., Hu et al. in 2017 are the study of work done in MI using ML

techniques. Concept level techniques are implemented in year 2014 and 2015 by Ali et al. and

Liu et al. respectively. An approach has been proposed for identification of product features and

their relevant opinions. Cataldi et al. in 2013, Kang & Park and Li et al., in 2014, Van de Kauter

et al. in 2015, Li et al. in 2016 and Law et al., Daniel et al. in 2017 implemented hybrid

techniques in MI. Moreover, lexicon based techniques are used by Mostafa in 2013 to find

sentiment polarity towards famous brands and their services ; Cho et al. in 2014 proposed a

process for automated lexicon creation of stock market and a method is proposed for revising

sentiment dictionaries to classify sentiments of product reviews. Song et al. in 2015 proposed a

framework of customer review analysis for evaluating quality of service.


Table 4. Literature work in application area of opinion mining

S.

No

Application

Area

Technique

Used

Year Paper

Count

References

1 Business

Intelligence

Machine

Learning

2011 1 (Tsai & Kwee, 2011)

2012 2 (Chung & Seng, 2012), (Zitnik, 2012)

2013 2 (Jang et al., 2013), (Pai et al., 2013)

2014 2 (Wu et al., 2014), (Li et al., 2014)

2 Government

Intelligence

Concept

level

2014 1 (Rill et al., 2014)

Hybrid

2012 1 (De Fortuny et al., 2012)

2013 1 (Xianghua et al., 2013)

2014 1 ( Makazhanov et al., 2014)

2015 1 (Kagan et al., 2015)

Lexicon 2016 1 (Haselmayer & Jenny, 2016)

Machine

Learning

2011 1 (Cao et al., 2011)

2014 3 (Katakis et al., 2014), (ElTayeby et al., 2014),

(Panasyuk et al., 2014)

2015 1 (Mohammad et al., 2015)

2016 5 (Cheng et al., 2016), (Tayal & Yadav, 2016),

(Adedoyin-Olowe et al., 2016), (Gull et al.,

2016), (O'Leary, 2016)

2017 1 (Singh et al., 2017)

3

Information

and Security

Analysis

Lexicon 2016 1 (Bhargava et al., 2016)

Machine

Learning

2011 3 (He & Zhou, 2011), (Liu et al., 2011), (Fan &

Wu, 2011)

2012 5 (Zhai et al., 2012), (Balahur et al., 2012),

(Cruz et al., 2012), (Lane et al., 2012), (Wang

et al., 2012)

2013 2 ( Bagheri et al., 2013),( Haddi et al., 2013)

2014 5 (Wu et al., 2014), (Kansal & Toshniwal,

2014), (Dařena et al., 2014), (Zheng et al.,

2014), (Dehkharghani et al., 2014)

2015 2 (Wang & Liu, 2015), (Kurian & Asokan,

2015)

2016 2 (Li et al., 2016), (Qiang et al., 2016)

2017 1 (Rout et al., 2017)

4 Market

Intelligence

Concept

level

2014 1 (Ali et al., 2014)

2015 1 (Liu et al., 2015)

Hybrid

2013 1 (Cataldi et al., 2013)

2014 2 (Kang & Park, 2014), (Li et al., 2014)

2015 1 (Van de Kauter et al., 2015)

2016 1 (Li et al., 2016)

2017 2 (Law et al., 2017), (Daniel et al., 2017)

Lexicon 2013 1 (Mostafa, 2013)


2014 1 (Cho et al., 2014)

2015 1 (Song et al., 2015)

Machine

Learning

2011 7 (Ghose & Ipeirotis, 2011), (Sanchez-Monzon

et al., 2011), (Fan et al., 2011), (Park & Lee,

2011), (Xu et al., 2011), (Bai, 2011), (Chen &

Tseng, 2011)

2012 8 (Yu et al., 2012), (Liu et al., 2012), (Chen et

al., 2012), (Schumaker et al., 2012), (Min &

Park, 2012), (Reyes & Rosso, 2012), (Esparza

et al., 2012), (Kang et al., 2012)

2013 6 (Wöllmer et al., 2013), (Bafna & Toshniwal,

2013), (Liu et al., 2013), (Yaakub et al., 2013),

(Basari et al. 2013), (Yu et al., 2013)

2014 9 (Wu et al., 2014), (Zhang et al., 2014), (Jiang

et al., 2014), (Vinodhini & Chandrasekaran,

2014a, 2014b), (Yang & Chao, 2014), (Wang

et al., 2014), (Zhang et al., 2014), (Smailović

et al., 2014)

2015 10 (Zhang et al., 2015), ( Parkhe & Biswas,

2015), (Bastı et al., 2015), (He et al., 2015),

(Sandoval & Hernández, 2015),

(Zimmermann et al., 2015), (D’Avanzo &

Pilato, 2015), (Amarouche et al., 2015),

(Nguyen et al., 2015), (Fernández-Martínez et

al., 2015)

2016 19 (Yan et al., 2016), (Manek et al., 2017), (Li et

al., 2016), (Vinodhini & Chandrasekaran,

2016), (Cosma & Acampora, 2016), (Hur et

al., 2016), (Luo et al., 2016), (Chua &

Banerjee, 2016), (Jin et al., 2016), (Zhang et

al., 2016), (Malik & Shakshuki, 2016),

(Djouvas et al., 2016), (Lee et al., 2016),

(Khan et al., 2016), (Manaman et al., 2016),

(Nayak et al., 2016), (Anand & Naorem,

2016), (Chen et al., 2016), (Feuerriegel &

Prendinger, 2016)

2017 9 (Wang, 2017), (Yuan et al., 2017), (Guo et al.,

2017), (Tsirakis et al., 2017), (Gui et al.,

2017), (Tsai & Wang, 2017), (Chan & Chong,

2017), (Ding et al., 2017), (Hu et al., 2017b)

5

Sub

Component

Technology

Hybrid 2016 2 (Lei et al., 2016), (Dong et al., 2016)

2017 1 (Ma et al.,2017)

Lexicon 2015 1 (Rosa et al., 2015)

Machine

Learning

2011 2 (Qiao et al., 2011), (Bridge & Healy, 2011)

2012 1 (Anooj, 2012)

2013 6 (Kardan & Ebrahimi, 2013), (Geva & Zahavi,

2013), (García-Cumbreras et al., 2013), (Chen

& Wang, 2013), (Peleja et al., 2013),


(Scharkow, 2013)

2014 1 (Sun et al., 2014)

2015 6 (Capdevila et al., 2015), (Coussement et al.,

2015), (Colace et al., 2015), (Wei et al., 2015),

(Alahmadi & Zeng, 2015), (Sohail et al.,

2015),

2016 4 (Pröllochs et al., 2016), (Li et al., 2016),

(Zhang et al., 2016), (Bilici & Saygın, 2016)

2017 2 (Gurini et al., 2017), (Ma et al., 2017)

6

Smart

Society

Services

Concept

level

2015 2 (Grant-Muller et al., 2015), (Ali et al., 2015)

Hybrid

2013 1 (Cameron et al., 2013)

2014 2 (Nguyen et al., 2014), (Ortigosa et al., 2014)

2015 1 (Shen & Kuo, 2015)

2016 2 (Asghar et al., 2016), (Ali et al., 2016)

2017 1 (Xue et al., 2017)

Lexicon

2012 1 (Moreo et al., 2012)

2013 1 (Mostafa, 2013)

2014 1 (Lei et al., 2014)

2015 2 (Teodorescu, 2015), (Yu & Wang, 2015)

2017 1 (Neidhardt et al., 2017)

Machine

Learning

2012 2 (Zakzouk & Mathkour, 2012), (Cambria et al.,

2012)

2013 3 (Valsamidis et al., 2013), (Huh et al., 2013),

(Lu, 2013)

2014 8 (Yan et al., 2014), (de Oña et al., 2014),

(Bogdanova et al., 2014), (Chen et al., 2014),

(Cheng et al., 2014), (Cao et al., 2014), (Wiley

et al., 2014), (Küçük et al., 2014)

2015 5 (İskender & Batı, 2015), (Aliandu, 2015),

(Bucur, 2015), (Ji et al., 2015), (Bobicev et

al., 2015)

2016 6 (Lv & El-Gohary, 2016), (Hu et al., 2017a),

(Bui et al., 2016), (Rossetti et al., 2016),

(Yang et al., 2016), (Lee et al., 2016)

2017 4 (James et al., 2017), (Kuflik et al., 2017), (Du

et al., 2017), (Dai & Hao, 2017)

Review Reporting/Results

This section answers the research questions identified in section 4.2.

RQ1 What are the different application areas & techniques of opinion mining?

Answer Opinion mining is widely used in various real life application areas which could be

broadly classified as follows:


o Government Intelligence (GI) exploits opinion mining in order to provide public participation

for improving the process of planning and decision making. It help governments and

organizations in resource optimization, improves democratic measures/mechanism,

revolutionize the delivery mode of public services in order to develop a citizen centric model

of governance. Government Intelligence spectrum covers Politics, Advisory Voting,

Advocacy etc.

o Sub Component Technology take advantage of opinion mining which has a significant

promising role in enabling the mechanics of other systems. Expert Finding, Recommender

Systems, Citation Analysis are some of the action areas of opinion mining in sub component

technology.

o Information and Security Analysis (ISA) utilizes opinion mining for the analysis of

information as well as security. Information could be analyzed by summarization techniques,

content analysis etc. and security analysis deals with spam detection, fraud detection etc.

Both information and security analysis are generic domain could be applied in any

application sphere of opinion mining for the sake of making results/conclusions more

organized and genuine.

o Market Intelligence (MI) deals with the process of collecting and analyzing information

associated to market analytics and research for good decision making and strategies

conceptualization. With the digitization of market intelligence, numerous data resources are

used to get an overall illustration of company's reputation monitoring and its existing market,

consumer needs and priorities, market challenges, struggle/competition, and expansion

competency for fresh and latest products and services. Trend analyzers and researches

utilizes opinion mining as a tool/technique which aids effectual and proficient assessment of

consumer/customer viewpoint. Various areas fall under the umbrella of market intelligence

such as Product/Service Reviews, Market Analysis and Decision Making, Product Service

Quality Improvement etc.

o Business Intelligence (BI) is an industrial science of gathering, analyzing, presenting,

reporting and broadcasting of business information to aid business analysts, executives and

end users with well updating of business decision and conclusions. The goal is to exploit

upcoming opportunities and executes an effective plan of action for its accomplishment.

Opinion Mining helps in strengthening the various activities of business intelligence such as

business process optimization, trend analysis etc. by considering the opinion of different

stakeholders. A new social commerce paradigm has emerged.

o Smart Society Services (SSS) is an integration of various necessities and utilities, required for

the anatomy of an innovative and sustainable society by the use of internet and

communication technologies and digital networks. These services targets to resolve the

social, environmental and economical problems of the present era in view of improving

wellbeing, quality of life, to focus on resource optimization and ultimately to meliorate the

collective intelligence of society. Opinion mining being a powerful concept/technology to


grab the public sentiments, plays an important role in cultivating the foundation of smart

society by contributing the public/audience propensity towards smart services. Some key

areas of smart society services are education sector, banking, healthcare, transportation etc.

o Some examples of their respective field of action are listed in Table 5. The corresponding

field of action for the respective application areas are not limited to the one mentioned over

here. More action fields could be accommodated into the broader category of applications

which is an open scope for this SLR.

Table 5. Opinion Mining Application Areas

Techniques of Opinion Mining

o Opinion mining techniques can be broadly divided into following approaches:

Application Area(s) Field of Action(s)

Government Intelligence

Voting Advisor

Advocacy & Policy Making

Advocacy Evaluation & Monitoring

Politics

Detection of Early Warning system

Campaigning

Natural Resource Protection/Environment Utility

Sub Component

Technology

Expert Finding

Recommender System

Argument Mapping Software

Detection of "flames"

Citation Analysis

Collective Intelligence

Information and

Security Analysis

Automated Content Analysis

Buying a commodity or service

Spam Detection

Automated Summarization

Decision Making Tasks

Market Intelligence

Product/Service Reviews

Market Analysis and Decision Making

Product Service Quality Improvement

Ads Placement

Reputation Monitoring

Stock Exchange

Business Intelligence

Business Process Optimization

Automated Trading System

Trend Prediction in Sales

Smart Society Services

Banking

Education

Health Care

Sports

Tourism Industry

Resource Optimization


o Machine Learning Based Approaches which implement machine learning algorithms that

can learn from data and helps in predictive analysis. These can be classified into two

groups: supervised and unsupervised. Supervised approaches apply past learning from

training data to test new data whereas unsupervised approach itself identifies unrevealed

patterns in unlabeled data.

o Lexicon Based Approaches which are based on sentiment thesaurus, a collection of

known and precompiled sentiment terms. The two main approaches for compilation of

opinion lexicon are: dictionary based and corpus based. In dictionary based approach,

manual process is used for creation of a small set of opinion words reflecting their

orientation and gets it enhanced by searching their synonyms and antonyms. Corpus

based approach uses statistical or semantic methods for finding sentiment polarity and

determining words emotional affinity.

o Hybrid Approaches are a combination of both machine learning and lexicon based

approach, collaborated for better performance.

o Concept Based Approaches have been introduced recently & are based on huge

knowledge bases and analyzing the conceptual information associated to natural language

opinion behind the multi-word expressions. Semantic analysis is performed by the use of

web ontologies or semantic networks (Cambria et al., 2013).

Thus, approaches used in opinion mining so far are listed as under (K. Ravi & V. Ravi, 2015),

(Medhat et al., 2014):

Machine Learning Based

Supervised Learning

Decision Tree Classifier

Rule Based Classifier

Probabilistic Classifier

Naive Bayes

Bayesian Network

Maximum Entropy

Linear Classifier

Support Vector machine

Neural Network

Unsupervised Learning

Meta Classifiers

Lexicon Based

Dictionary Based

Corpus Based

Statistical

Semantic

Hybrid

Concept Based

Ontology Based

Context/ Non Ontology Based


Table 6 represents opinion mining technique and their respective application area in which it is

used as observed from the selected final studies.

Table 6. Usage of Opinion Mining Techniques in its application areas

Application Area

Techniques

BI GI ISA MI SCT SSS

Machine Learning ✓ ✓ ✓ ✓ ✓ ✓

Lexicon Based ✓ ✓ ✓ ✓ ✓

Hybrid ✓ ✓ ✓ ✓

Concept Level ✓ ✓ ✓

Charts in Figure 8 and 9 depict the percentage of work done in various application areas of

opinion mining and the percentage usage of its techniques/approaches in these application areas.

Figure 8. % Work done in OM Applications

Figure 9. % Work done in OM Techniques

RQ2 How much work has been done in the area of government intelligence using opinion

mining?


Answer Following are the various field of action in government intelligence application arena

as listed in Table 7 along with their motivation, detail and scope.

Table 7. Literature work in area of Government Intelligence

S. No. Author Year Publication Field of

Action

Technique(s) Used Dataset Used

Motivation Details and Scope Name

Algorithm

Used Scope Source

1 Cao et al. 2011

Elsevier,

Decision

Support Systems

Voting

Advisor ML

Factor

Analysis, Ordinal

Logistic

Regression

User

Reviews

CNET Download.c

om

Online user reviews

are effective on

product sales and consumer decision-

making. The

“helpfulness” property of online user reviews

aids consumers to

manage with the huge

amount of information

available on Web and

promotes decision-making.

This paper investigates how

the features of online user

reviews effects on the number of helpfulness votes received.

Basic, stylistic and semantic

are the three characteristics. Latest reviews inclined to be

more precise and complete as

they consider previews

reviews, therefore receive

more helpfulness vote.

2 De Fortuny

et al. 2012

Elsevier, Expert

Systems with

Applications

Politics Hybrid

KDD,

Lexicon based

News

articles

All major Flemish

newspapers

in 2011

Manual analysis of

contrasting and evaluating opinions for

large amount of online

available articles is troublesome job.

Requisite is to

overcome the method by automating the

process.

In this paper, newspapers are

compared on specific topics in order to help political parties

in decision making by

proposing a framework based on text-mining. Online

dashboard updates Regular

update of stats in real time by online dashboard promote

above process. Future vision

is to apply the framework over American presidential race.

3 Xianghua et

al. 2013

Elsevier,

Knowledge

Based Systems

Politics Hybrid

NB, KNN,

SVM,LDA,

Hownet Lexicon

Blogs

2000-SINA-Blog

dataset,

300-SINA-Blog

dataset

With the development

of communication

technologies, social

media provides a

platform to share public sentiment and

ideas over any topic.

Manual analysis of these online reviews is

becoming difficult due

to exponential rise in their number. Hence,

objective is to make

the process convenient.

A trained LDA model is used

for opinion mining of Chinese

social reviews for

identification of sliding

windows topic. How Net lexicon is used to analyze the

related sentiment. Proposed

method performs better in topic identification and

sentiment analysis. Future

work targets to use existing method and compare it with

TSM, JST in different

datasets.

4 Katakis et al.

2014

IEEE

Transactions on

Cybernetics

Voting Advisor

ML

Dimensiona

lity Reduction

PCA

Preferenc

e Matcher research

Consortiu

m

www.preferencematche

r.org

To provide voting advice to the users of

social voting advice

applications is required.

To compare users political opinion and providing voting

recommendation, a new

online tool has been proposed for social voting advice

application. Future areas are

data privacy and transparency

in voting advice application

recommendation engine.

5 Makazhanov et al.

2014

Social

Network Analysis and

Mining

Politics

Hybrid

SVM Logic Regression

Model

Sentistrength

Twitter Election

Commissi

on of Pakistan

Twitter API

To understand users political preference

prediction problem on

Twitter network by forming contextual and

behavioral features

based prediction models.

Users political preference was predicted from the knowledge

derived from the content

generated and users behavior during the campaign.

Comparison concludes that

during preelection vocal users are unenthusiastic in their

preference change than silent

users.


6 ElTayeby et

al. 2014

Elsevier, Procedia

Computer

Science

Campaig

ning ML EM

User

Reviews Twitter

Social media plays a

vital role in spreading opinions over many

diversified subject

matters. Professionals take advantage of this

data for publicizing

their products and getting user feedback.

Hence, requirement is

to group similar opinions in to a cluster.

This paper inspects effect of

media over grouping of opinions. EM algorithm is

used for segregating opinions.

Grouping of opinion is done specific to certain political

topics and their percentage

media impact is calculated. Future works targets to collect

sentiment from multiple topics

and apply the proposed framework.

7 Panasyuk et al.

2014

Elsevier,

Procedia Computer

Science

Politics ML LDA News articles

Twitter

Twitter being a

significant social

networking site facilitates the

identification of

communities. Objective is to

recognize conflicting

communities.

This paper investigates the

usage of Twitter for

determining This paper addresses how

Twitter can be used for

identifying conflict among communities users.

Documents were aggregated

based on topic and community followed by applying SA to

find views of every

community on individual topic. Ranking of topic with

opposing views has been done. Future scope is to find

how the difference level is

used for detection of early warning system.

8 Rill et al. 2014

Elsevier,

Knowledge Based

Systems

Politics Concept level

Ontology based

User Reviews

Twitter,

Google

Trends

Social networking sites

are gaining popularity

over the past few years. Various

communication

channels are used for dissemination of

information. Aim is to

recognize that Twitter detects the topics faster

than other information

channels.

PoliTwi, a system is designed

to find upcoming political

topics in Twitter. Identified topics are broadcasted over

various channels. Results

were compared with Google Trends and results reflects

Twitter find topics earlier than

other communication channel.

9 Kagan et al. 2015

IEEE

Intelligent Systems

Politics Hybrid

SVM

classifier + Statistical

sentiment

diffusion model

India

Election

Tweet Database(

IET-DB)

-

Requirement is to

forecast the process of

spreading favor or denial toward a

candidate on Twitter

by implementing sentiment diffusion

forecasting.

Sentiment score is calculated

for each tweet to analyze

expressed sentiment Automatic learning is done for

diffusion estimation model in

order to find how a phenomenon diffuses over

network.

10 Mohammad et al.

2015

Elsevier,

Information & Processing

Management

Voting Advisor

ML SVM

Question

naires, User

Reviews

Amazon, Twitter

Social media sites

facilitate the political parties, candidates and

other stakeholders to make their views

accessible to public

during elections. Manual analysis of

opinions is difficult to

acquire an overall outlook of various

posts relevant to a

particular event. Hence, aim is to

automate the process

of analyzing opinions in electoral tweets.

In this paper, automatic

classifiers have been developed to identify emotion

and purpose labels in electoral tweets for 2012 US

presidential election.


11 Cheng et al. 2016 Applied Intelligence

Politics ML Recurrent NN

Micro

blogging

Websites

Twitter Sina Weibo

Requirement is to

implement sentiment parsing based methods

in opinion polling on

Chinese public figures using microblogs.

This paper uses RNN based

sequence labeling method for opinion target labeling and

their corresponding

sentiments. Result concludes that jointly representing

opinion targets and sentiments

make more sense in sentiment parsing of public figures in

microblogs.

12 Haselmeyer 2016 Quality &

Quantity Politics

Lexico

n

Dictionary

based

SentiWS

[22] -

Objective is to build a

negative sentiment dictionary by gathering

fine-grained sentiment

scores through crowdcoding.

In this paper a German

political language dictionary has been created specifically

for party statements and

media reports. New sentiment dictionary can be used in two

applications: negative

campaigning by parties and media tone.

13 Tayal et al. 2016 AI & Society Campaig

ning ML unigrams Twitter -

A sentiment analysis

tool is required for the estimation of success

rate of campaigns.

In this paper a detailed

analysis of Swachh Bharat Abhiyan has been computed

by the proposed tool which

extracts data from twitter. The tool computes weekly and

monthly analysis of campaign

related tweets.

14 Adedoyin-

Olowe et al. 2016

Elsevier, Expert

Systems with

Applications

Politics ML

Apriori,

Association Rule

Limimg,

TRCM algo proposed

Sports

and Politics

Twitter

Huge amount of public

opinion over social

media like twitter can be utilized for

computation of

detection and tracking of related topics. Aim

is to apply topic

detection and tracking methods for extraction

of valuable content..

In this paper Transaction-

based Rule Change Mining is

applied two domains- sports (The English FA Cup 2012)

and politics (US Presidential

Elections 2012 and Super Tuesday 2012). It

performed better on dataset in

the sports domain.

15 Gull et al. 2016

Elsevier,

Procedia Computer

Science

Politics ML NB, SVM User Reviews

Twitter

Objective is to use

linguistic analysis and opinion classifiers for

simplifying opinion

mining approach in order to determine

positive, negative and

neutral sentiments for Pakistan political

parties.

In this paper, information

from tweets has been extracted by preprocessing of

raw data. Comparison of SVM

and Naive Bayes has been done for tweets

classification where SVM

outperforms NB.

16 Singh et al. 2017 ICT Express

Advocacy &

Policy

Making

ML

Statistical

document classificatio

n with rule-

based

filtering

User

Reviews Twitter

Public opinion on social media can act as

a magnitude of

acceptance or rejection of certain ordinances.

Requirement is to

gauge sentiment of

public for the

government policy of

demonetization.

Indian government has implemented a

demonetization policy and

opinion mining concludes the favorable feedback of the

public for the same.

17 O'Leary, D. E.

2016

Elsevier,

Decision Support

Systems

Advocac

y & Policy

Making

ML

Decision

Trees, Regression

Analysis

Canada's

Digital

Compass

Sept–Oct

2014 from PWC's web

site

http://pwc-compass.ch

aordix.com/

Objective is to analyze

the data of a crowd

sourcing study to extract Canada's digital

future.

In this paper, relationship of

number of votes and number

of innovated comments has been validated with respect to

another case study. They

found that the number of votes is related positively with the

number of words and

magnitude of idea.


The following charts in figure 10, 11 and 12 represent the percentage of work done in various

field actions of government intelligence and the percentage use of opinion mining techniques in

these action areas.

Figure 10. % work done in GI

RQ3 Which (big) data sets are used for opinion mining in government intelligence?

Answer From the above literature review following are the big data sets used for opinion mining

in government intelligence:

User Reviews from CNET, Twitter, Google

Trends, Election India Tweet

News Articles from Newspapers, Twitter

Twitter API Preference Matcher Resort Consortium

Questionnaire from Amazon, Twitter Micro blogging sites Twitter Sina, Weibo

MyGov.in Facebook

Citymis application Sports and politics from Twitter

RQ4 Which techniques of opinion mining have been used in advocacy so far?

Answer Extensive literature review revealed that machine learning and hybrid techniques have

been used in the area of advocacy. Semantic Role Labelling, GATE Tool, Rule based method,

Naive Bayes, Corpus Based, Statistical document classification with rule-based filtering are

some of the approaches applied for opinion mining in advocacy so far.


Machine Learning

65%Lexicon based

6%

Hybrid23%

Concept Level6%

OM Techniques used in GI(%)

Figure 11. % work done in GI with OM techniques

2011 2012 2013 2014 2015 2016 2017

Year wise Number of Publication(s) in GI

1 1 1 5 2 6 1

0

1

2

3

4

5

6

7

Year wise Number of Publication(s) in GI

Figure 12. Year wise GI Publications

Conclusion

This paper presents a systematic, comprehensive review on the research work done in different

application areas of opinion mining during 2011-2017.The application areas are classified into

six broad domains i.e. Government Intelligence, Market Intelligence, Information and Security

Analysis, Sub Component Technology, Business Intelligence and Smart Society Services. Some

important conclusions have been drawn by answering identified research questions. From the

literature it is clear that Market Intelligence is the major application area of opinion mining while

GI and BI are the least explored ones. Machine learning techniques are mostly used in GI so far

with the maximum number of publication in GI is in the year 2016 till now. Certain research

gaps which are identified in this study are:


o To understand public opinion and sentiments by a human or machine is a difficult job due

to intricate and ambiguous nature of their verbalization. Therefore, automation of

sentiment extraction from text is a big challenge.

o User generated content are usually briefly written and vernacular in nature with

grammatical and typo mistakes. Therefore, opinion mining in such content is

comparatively difficult than regular texts.

o Due to domain specific characteristics of data, many sentimental words are not used for

expressing opinions. An expression could be positive or neutral in nature although

negative words are used.

o Affordability of opinion mining software are only with organizations and government and

not by citizens due to which content analysis has not been democratized.

o Few efforts have been reported in leveraging big social data for intelligent opinion

mining which is a dynamic field with problems stemming either from the velocity and

volume of data or the variety in application domain.

o There is no standard model or framework exists till date which incorporates opinion

mining in any action field(s) of government intelligence.

o Detection of Early Warning Systems & Advocacy Evaluation & Monitoring are the two

action field(s) which have been least explored in the area of government intelligence.

The study has certain limitations, specifically the restriction of covering four electronic databases

with a manual search of high quality journal or conference papers only. Future study will try to

cover more papers by automating searches.

References

Adedoyin-Olowe, M., Gaber, M. M., Dancausa, C. M., Stahl, F., & Gomes, J. B. (2016). A rule

dynamics approach to event detection in Twitter with its application to sports and politics.

Expert Systems with Applications, 55, 351-360.

Alahmadi, D. H., & Zeng, X. J. (2015). ISTS: Implicit social trust and sentiment based approach to

recommender systems. Expert Systems with Applications, 42(22), 8840-8849.

Ali, F., Kim, E. K., & Kim, Y. G. (2015). Type-2 fuzzy ontology-based opinion mining and information

extraction: A proposal to automate the hotel reservation system. Applied Intelligence, 42(3),

481-500.

Ali, F., Kwak, D., Khan, P., Islam, S. R., Kim, K. H., & Kwak, K. S. (2017). Fuzzy ontology-based

sentiment analysis of transportation and city feature reviews for safe traveling. Transportation

Research Part C: Emerging Technologies, 77, 33-48.

Ali, F., Kwak, K. S., & Kim, Y. G. (2016). Opinion mining based on fuzzy domain ontology and

Support Vector Machine: A proposal to automate online review classification. Applied Soft

Computing, 47, 235-250.

Aliandu, P. (2015). Sentiment Analysis to Determine Accommodation, Shopping and Culinary

Location on Foursquare in Kupang City. Procedia Computer Science, 72, 300-305.


Amarouche, K., Benbrahim, H., & Kassou, I. (2015). Product Opinion Mining for Competitive

Intelligence. Procedia Computer Science, 73, 358-365.

Anand, D., & Naorem, D. (2016). Semi-supervised Aspect Based Sentiment Analysis for Movies

Using Review Filtering. Procedia Computer Science, 84, 86-93.

Anooj, P. K. (2012). Clinical decision support system: Risk level prediction of heart disease using

weighted fuzzy rules. Journal of King Saud University-Computer and Information Sciences,

24(1), 27-40.

Arputhamary, B., & Jayapriya, V. (2016). Interpreting the public sentiment variations on twitter using

virtual shuffling for data movement in big data. International Journal of Control Theory and

Applications, 9(26), 389-395

Asghar, M. Z., Ahmad, S., Qasim, M., Zahra, S. R., & Kundi, F. M. (2016). SentiHealth: creating

health-related sentiment lexicon using hybrid approach. SpringerPlus, 5(1), 1139.

Bafna, K., & Toshniwal, D. (2013). Feature based summarization of customers’ reviews of online

products. Procedia Computer Science, 22, 142-151.

Bagheri, A., Saraee, M., & De Jong, F. (2013). Care more about customers: unsupervised domain-

independent aspect detection for sentiment analysis of customer reviews. Knowledge-Based

Systems, 52, 201-213.

Bai, X. (2011). Predicting consumer sentiments from online text. Decision Support Systems, 50(4),

732-742.

Balahur, A., Kabadjov, M., Steinberger, J., Steinberger, R., & Montoyo, A. (2012). Challenges and

solutions in the opinion summarization of user-generated content. Journal of Intelligent

Information Systems, 39(2), 375-398.

Basari, A. S. H., Hussin, B., Ananta, I. G. P., & Zeniarja, J. (2013). Opinion mining of movie review

using hybrid method of support vector machine and particle swarm optimization. Procedia

Engineering, 53, 453-462.

Bastı, E., Kuzey, C., & Delen, D. (2015). Analyzing initial public offerings' short-term performance

using decision trees and SVMs. Decision Support Systems, 73, 15-27.

Bhargava, R., Sharma, Y., & Sharma, G. (2016). ATSSI: Abstractive Text Summarization Using

Sentiment Infusion. Procedia Computer Science, 89, 404-411.

Bilici, E., & Saygın, Y. (2017). Why do people (not) like me?: Mining opinion influencing factors

from reviews. Expert Systems with Applications, 68, 185-195.

Bobicev, V., Sokolova, M., & Oakes, M. (2015). What goes around comes around: learning sentiments

in online medical forums. Cognitive Computation, 7(5), 609-621.

Bogdanova, D., Rosso, P., & Solorio, T. (2014). Exploring high-level features for detecting

cyberpedophilia. Computer Speech and Language, 28(1), 108-120.

Bridge, D., & Healy, P. (2012). The GhostWriter-2.0 case-based reasoning system for making content

suggestions to the authors of product reviews. Knowledge-Based Systems, 29, 93-103.

Bucur, C. (2015). Using Opinion Mining Techniques in Tourism. Procedia Economics and Finance,

23, 1666-1673.

Bui, N., Yen, J., & Honavar, V. (2016). Temporal Causality Analysis of Sentiment Change in a Cancer

Survivor Network. IEEE Transactions on Computational Social Systems, 3(2), 75-87.


Cambria, E., Benson, T., Eckl, C., & Hussain, A. (2012). Sentic PROMs: Application of sentic

computing to the development of a novel unified framework for measuring health-care quality.

Expert Systems with Applications, 39(12), 10533-10543.

Cambria, E., Schuller, B., Xia, Y., & Havasi, C. (2013). New avenues in opinion mining and sentiment

analysis. IEEE Intelligent Systems, 28(2), 15-21.

Cameron, D., Smith, G. A., Daniulaityte, R., Sheth, A. P., Dave, D., Chen, L., & Falck, R. (2013).

PREDOSE: a semantic web platform for drug abuse epidemiology using social media. Journal

of Biomedical Informatics, 46(6), 985-997.

Cao, J., Zeng, K., Wang, H., Cheng, J., Qiao, F., Wen, D., & Gao, Y. (2014). Web-based traffic

sentiment analysis: Methods and applications. IEEE transactions on Intelligent Transportation

systems, 15(2), 844-853

Cao, Q., Duan, W., & Gan, Q. (2011). Exploring determinants of voting for the “helpfulness” of online

user reviews: A text mining approach. Decision Support Systems, 50(2), 511-521.

Capdevila, J., Arias, M., & Arratia, A. (2016). GeoSRS: A hybrid social recommender system for

geolocated data. Information Systems, 57, 111-128.

Cataldi, M., Ballatore, A., Tiddi, I., & Aufaure, M. A. (2013). Good location, terrible food: detecting

feature sentiment in user-generated reviews. Social Network Analysis and Mining, 3(4), 1149-

1163.

Chan, S. W., & Chong, M. W. (2017). Sentiment analysis in financial texts. Decision Support Systems,

94, 53-64.

Chen, C. C., & Tseng, Y. D. (2011). Quality evaluation of product reviews using an information

quality framework. Decision Support Systems, 50(4), 755-768.

Chen, K., Luo, P., Xu, D., & Wang, H. (2016). The dynamic predictive power of company comparative

networks for stock sector performance. Information & Management, 53(8), 1006-1019.

Chen, L., & Wang, F. (2013). Preference-based clustering reviews for augmenting e-commerce

recommendation. Knowledge-Based Systems, 50, 44-59.

Chen, L., Qi, L., & Wang, F. (2012). Comparison of feature-level learning methods for mining online

consumer reviews. Expert Systems with Applications, 39(10), 9588-9601.

Chen, X., Vorvoreanu, M., & Madhavan, K. (2014). Mining social media data for understanding

students’ learning experiences. IEEE Transactions on Learning Technologies, 7(3), 246-259.

Cheng, J., Zhang, X., Li, P., Zhang, S., Ding, Z., & Wang, H. (2016). Exploring sentiment parsing of

microblogging texts for opinion polling on Chinese public figures. Applied Intelligence, 45(2),

429-442.

Cheng, V. C., Leung, C. H., Liu, J., & Milani, A. (2014). Probabilistic aspect mining model for drug

reviews. IEEE Transactions on Knowledge and Data Engineering, 26(8), 2002-2013.

Cho, H., Kim, S., Lee, J., & Lee, J. S. (2014). Data-driven integration of multiple sentiment

dictionaries for lexicon-based sentiment classification of product reviews. Knowledge-Based

Systems, 71, 61-71.

Chua, A. Y., & Banerjee, S. (2016). Helpfulness of user-generated reviews as a function of review

sentiment, product type and information quality. Computers in Human Behavior, 54, 547-554.

Chung, W., & Tseng, T. L. B. (2012). Discovering business intelligence from online product reviews:

A rule-induction framework. Expert Systems with Applications, 39(15), 11870-11879.


Colace, F., De Santo, M., Greco, L., Moscato, V., & Picariello, A. (2015). A collaborative user-

centered framework for recommending items in Online Social Networks. Computers in Human

Behavior, 51, 694-704.

Cosma, G., & Acampora, G. (2016). A computational intelligence approach to efficiently predicting

review ratings in e-commerce. Applied Soft Computing, 44, 153-162.

Coussement, K., Benoit, D. F., & Antioco, M. (2015). A Bayesian approach for incorporating expert

opinions into decision support systems: A case study of online consumer-satisfaction detection.

Decision Support Systems, 79, 24-32.

Cruz, F. L., Troyano, J. A., Enríquez, F., Ortega, F. J., & Vallejo, C. G. (2013). ‘Long autonomy or long

delay?’ The importance of domain in opinion mining. Expert Systems with Applications, 40(8),

3174-3184.

D’Avanzo, E., & Pilato, G. (2015). Mining social network users' opinions to aid buyers’ shopping

decisions. Computers in Human Behavior, 51, 1284-1294.

Dai, H., & Hao, J. (2017). Mining social media data on marijuana use for post traumatic stress

disorder. Computers in Human Behavior, 70, 282-290.

Daniel, M., Neves, R. F., & Horta, N. (2017). Company event popularity for financial markets using

Twitter and sentiment analysis. Expert Systems with Applications, 71, 111-124.

Dařena, F., Žižka, J., & Přichystal, J. (2014). Clients’ freely written assessment as the source of

automatically mined opinions. Procedia Economics and Finance, 12, 103-110.

Dave, K., Lawrence, S., & Pennock, D. M. (2003, May). Mining the peanut gallery: Opinion

extraction and semantic classification of product reviews. In Proceedings of the 12th

international conference on World Wide Web (pp. 519-528). ACM.

De Fortuny, E. J., De Smedt, T., Martens, D., & Daelemans, W. (2012). Media coverage in times of

political crisis: A text mining approach. Expert Systems with Applications, 39(14), 11616-11622.

de Oña, R., López, G., de los Rios, F. J. D., & de Oña, J. (2014). Cluster analysis for diminishing

heterogeneous opinions of service quality public transport passengers. Procedia-Social and

Behavioral Sciences, 162, 459-466.

Dehkharghani, R., Mercan, H., Javeed, A., & Saygin, Y. (2014). Sentimental causal rule discovery

from Twitter. Expert Systems with Applications, 41(10), 4950-4958.

Digital India – A programme to transform India into digital empowered society and knowledge

economy. (2014, August 20). Retrieved August 20, 2017, from

http://pib.nic.in/newsite/PrintRelease.aspx?relid=108926

Digital India: Transforming India into a Knowledge Economy. Retrieved August 20, 2017, from

http://www.makeinindia.com/article/-/v/digital-india-transforming-india-into-a-knowledge-

economy

Ding, Z., Liu, Z., Zhang, Y., & Long, R. (2017). The contagion effect of international crude oil price

fluctuations on Chinese stock market investor sentiment. Applied Energy, 187, 27-36.

Djouvas, C., Mendez, F., & Tsapatsoulis, N. (2016). Mining online political opinion surveys for

suspect entries: An interdisciplinary comparison. Journal of Innovation in Digital Ecosystems,

3(2), 172-182.

http://pib.nic.in/newsite/PrintRelease.aspx?relid=108926

http://www.makeinindia.com/article/-/v/digital-india-transforming-india-into-a-knowledge-economy

http://www.makeinindia.com/article/-/v/digital-india-transforming-india-into-a-knowledge-economy


Dong, R., O’Mahony, M. P., Schaal, M., McCarthy, K., & Smyth, B. (2016). Combining similarity and

sentiment in opinion mining for product recommendation. Journal of Intelligent Information

Systems, 46(2), 285-312.

Du, J., Xu, J., Song, H., Liu, X., & Tao, C. (2017). Optimization on machine learning based

approaches for sentiment analysis on HPV vaccines related tweets. Journal of Biomedical

Semantics, 8(1), 9.

ElTayeby, O., Molnar, P., & George, R. (2014). Measuring the Influence of Mass Media on Opinion

Segregation through Twitter. Procedia Computer Science, 36, 152-159.

Esparza, S. G., O’Mahony, M. P., & Smyth, B. (2012). Mining the real-time web: a novel approach to

product recommendation. Knowledge-Based Systems, 29, 3-11.

Fan, M., & Wu, G.. (2012). Opinion Summarization of Customer Comments. Physics Procedia, 24,

2220-2226.

Fan, T. K., & Chang, C. H. (2011). Blogger-centric contextual advertising. Expert Systems with

Applications, 38(3), 1777-1788.

Fernández-Martínez, F., García, A. H., & de María, F. D. (2015). Succeeding metadata based

annotation scheme and visual tips for the automatic assessment of video aesthetic quality in car

commercials. Expert Systems with Applications, 42(1), 293-305.

Feuerriegel, S., & Prendinger, H. (2016). News-based trading strategies. Decision Support Systems,

90, 65-74.

García-Cumbreras, M. Á., Montejo-Ráez, A., & Díaz-Galiano, M. C. (2013). Pessimists and optimists:

Improving collaborative filtering through sentiment analysis. Expert Systems with Applications,

40(17), 6758-6765.

Geva, T., & Zahavi, J. (2014). Empirical evaluation of an automated intraday stock recommendation

system incorporating both market data and textual news. Decision Support Systems, 57, 212-

223.

Ghose, A., & Ipeirotis, P. G. (2011). Estimating the helpfulness and economic impact of product

reviews: Mining text and reviewer characteristics. IEEE Transactions on Knowledge and Data

Engineering, 23(10), 1498-1512.

Grant-Muller, S. M., Gal-Tzur, A., Minkov, E., Nocera, S., Kuflik, T., & Shoor, I. (2014). Enhancing

transport data collection through social media sources: methods, challenges and opportunities

for textual data. IET Intelligent Transport Systems, 9(4), 407-417.

Gui, L., Zhou, Y., Xu, R., He, Y., & Lu, Q. (2017). Learning representations from heterogeneous

network for sentiment classification of product reviews. Knowledge-Based Systems, 124, 34-45.

Gull, R., Shoaib, U., Rasheed, S., Abid, W., & Zahoor, B. (2016). Pre processing of twitter's data for

opinion mining in political context. Procedia Computer Science, 96, 1560-1570.

Guo, K., Sun, Y., & Qian, X. (2017). Can investor sentiment be used to predict the stock price?

Dynamic analysis based on China stock market. Physica A: Statistical Mechanics and its

Applications, 469, 390-396.

Gurini, D. F., Gasparetti, F., Micarelli, A., & Sansonetti, G. (2018). Temporal people-to-people

recommendation on social networks with sentiment-based matrix factorization. Future

Generation Computer Systems, 78(1), 430-439.


Haddi, E., Liu, X., & Shi, Y. (2013). The role of text pre-processing in sentiment analysis. Procedia

Computer Science, 17, 26-32.

Haselmayer, M., & Jenny, M. Sentiment analysis of political communication: combining a dictionary

approach with crowdcoding. Quality & Quantity, 1-24.

He, W., Wu, H., Yan, G., Akula, V., & Shen, J. (2015). A novel social media competitive analytics

framework with sentiment benchmarks. Information & Management, 52(7), 801-812.

He, Y., & Zhou, D. (2011). Self-training from labeled features for sentiment analysis. Information

Processing & Management, 47(4), 606-616.

Hu, Y. H., Chen, K., & Lee, P. J. (2017a). The effect of user-controllable filters on the prediction of

online hotel reviews. Information & Management, 54(6), 728-744.

Hu, Y. H., Chen, Y. L., & Chou, H. L. (2017b). Opinion mining from online hotel reviews–A text

summarization approach. Information Processing & Management, 53(2), 436-449.

Huh, J., Yetisgen-Yildiz, M., & Pratt, W. (2013). Text classification for assisting moderators in online

health communities. Journal of Biomedical Informatics, 46(6), 998-1005.

Hur, M., Kang, P., & Cho, S. (2016). Box-office forecasting based on sentiments of movie reviews and

Independent subspace method. Information Sciences, 372, 608-624.

İskender, E., & Batı, G. B. (2015). Comparing Turkish Universities Entrepreneurship and

Innovativeness Index's Rankings with Sentiment Analysis Results on Social Media. Procedia -

Social and Behavioral Sciences, 195, 1543-1552.

James, T. L., Calderon, E. D. V., & Cook, D. F. (2017). Exploring patient perceptions of healthcare

service quality through analysis of unstructured feedback. Expert Systems with Applications, 71,

479-492.

Jang, H. J., Sim, J., Lee, Y., & Kwon, O. (2013). Deep sentiment analysis: Mining the causality

between personality-value-attitude for analyzing business ads in social media. Expert Systems

with applications, 40(18), 7492-7503.

Janowski, T. (2015). Digital government evolution: From transformation to contextualization.

Government Information Quarterly, 32(3), 221-236.

Ji, X., Chun, S. A., Wei, Z., & Geller, J. (2015). Twitter sentiment classification for measuring public

health concerns. Social Network Analysis and Mining, 5(1), 1-25.

Jiang, C., Liang, K., Chen, H., & Ding, Y. (2014). Analyzing market performance via social media: a

case study of a banking industry crisis. Science China Information Sciences, 57(5), 1-18.

Jin, J., Ji, P., & Gu, R. (2016). Identifying comparative customer requirements from product online

reviews for competitor analysis. Engineering Applications of Artificial Intelligence, 49, 61-73.

Kagan, V., Stevens, A., & Subrahmanian, V. S. (2015). Using twitter sentiment to forecast the 2013

pakistani election and the 2014 indian election. IEEE Intelligent Systems, 30(1), 2-5.

Kang, D., & Park, Y. (2014). Review-based measurement of customer satisfaction in mobile service:

Sentiment analysis and VIKOR approach. Expert Systems with Applications, 41(4), 1041-1050.

Kang, H., Yoo, S. J., & Han, D. (2012). Senti-lexicon and improved Naïve Bayes algorithms for

sentiment analysis of restaurant reviews. Expert Systems with Applications, 39(5), 6000-6010.

Kansal, H., & Toshniwal, D. (2014). Aspect based summarization of context dependent opinion words.

Procedia Computer Science, 35, 166-175.


Kardan, A. A., & Ebrahimi, M. (2013). A novel approach to hybrid recommendation systems based on

association rules mining for content recommendation in asynchronous discussion groups.

Information Sciences, 219, 93-110.

Katakis, I., Tsapatsoulis, N., Mendez, F., Triga, V., & Djouvas, C. (2014). Social voting advice

applications—definitions, challenges, datasets and evaluation. IEEE Transactions on

Cybernetics, 44(7), 1039-1052.

Khan, A. U. R., Khan, M., & Khan, M. B. (2016). Naïve Multi-label Classification of YouTube

Comments Using Comparative Opinion Mining. Procedia Computer Science, 82, 57-64.

Kitchenham, B., Brereton, O. P., Budgen, D., Turner, M., Bailey, J., & Linkman, S. (2009). Systematic

literature reviews in software engineering–a systematic literature review. Information and

Software Technology, 51(1), 7-15.

Küçük, A. E., Orhan, Z., & Kabil, M. (2014). Psychodroid: A Mobile Psychological Disorder

Detection Application by Dynamic Question Generation and Content Analysis. Procedia -

Social and Behavioral Sciences, 159, 691-696.

Kuflik, T., Minkov, E., Nocera, S., Grant-Muller, S., Gal-Tzur, A., & Shoor, I. (2017). Automating a

framework to extract and analyse transport related social media content: The potential and the

challenges. Transportation Research Part C: Emerging Technologies, 77, 275-291.

Kumar, A., & Joshi, A. (2017, March). Ontology driven sentiment analysis on social web for

government intelligence. In Proceedings of the Special Collection on eGovernment Innovations

in India (pp. 134-139). ACM.

Kumar, A., & Sharma, A. (2016). Paradigm shifts from e-governance to s-governance. The Human

Element of Big Data: Issues, Analytics, and Performance, 213.

Kumar, A., & Sharma, A. (2015). SWOT analysis of requirements engineering for web applications.

Journal of Advances Research in Science and Engineering, 4, 34-43.

Kumar, A., & Teeja, M. S. (2012). Sentiment analysis: A perspective on its past, present and future.

International Journal of Intelligent Systems and Applications, 4(10), 1.

Kurian, N., & Asokan, S. (2015). Summarizing user opinions: A method for labeled-data scarce

product domains. Procedia Computer Science, 46, 93-100.

Lane, P. C., Clarke, D., & Hender, P. (2012). On developing robust models for favourability analysis:

Model choice, feature sets and imbalanced data. Decision Support Systems, 53(4), 712-718.

Law, D., Gruss, R., & Abrahams, A. S. (2017). Automated defect discovery for dishwasher appliances

from online consumer reviews. Expert Systems with Applications, 67, 84-94.

Lee, A. J., Yang, F. C., Chen, C. H., Wang, C. S., & Sun, C. Y. (2016). Mining perceptual maps from

consumer reviews. Decision Support Systems, 82, 12-25.

Lee, S. W., Kim, I., Yoo, J., Park, S., Jeong, B., & Cha, M. (2016). Insights from an expressive writing

intervention on Facebook to help alleviate depressive symptoms. Computers in Human

Behavior, 62, 613-619.

Lei, J., Rao, Y., Li, Q., Quan, X., & Wenyin, L. (2014). Towards building a social emotion detection

system for online news. Future Generation Computer Systems, 37, 438-448.

Lei, X., Qian, X., & Zhao, G. (2016). Rating prediction based on social sentiment from textual

reviews. IEEE Transactions on Multimedia, 18(9), 1910-1921.


Li, H., Cui, J., Shen, B., & Ma, J. (2016). An intelligent movie recommendation system through

group-level sentiment analysis in microblogs. Neurocomputing, 210, 164-173.

Li, J., Xu, Z., Yu, L., & Tang, L. (2016). Forecasting oil price trends with sentiment of online news

articles. Procedia Computer Science, 91, 1081-1087.

Li, Q., Jin, Z., Wang, C., & Zeng, D. D. (2016). Mining opinion summarizations using convolutional

neural networks in Chinese microblogging systems. Knowledge-Based Systems, 107, 289-300.

Li, Q., Wang, T., Gong, Q., Chen, Y., Lin, Z., & Song, S. K. (2014). Media-aware quantitative trading

based on public Web information. Decision Support Systems, 61, 93-105.

Li, S., Ming, Z., Leng, Y., & Guo, J. (2016). Product ranking using hierarchical aspect structures.

Journal of Intelligent Information Systems, 1-22

Li, X., Xie, H., Chen, L., Wang, J., & Deng, X. (2014). News impact on stock price return via

sentiment analysis. Knowledge-Based Systems, 69, 14-23.

Liu, C. L., Hsaio, W. H., Lee, C. H., Lu, G. C., & Jou, E. (2012). Movie rating and review

summarization in mobile environment. IEEE Transactions on Systems, Man, and Cybernetics,

Part C (Applications and Reviews), 42(3), 397-407.

Liu, L. Z., Liu, H., Wang, H. S., Song, W., & Zhao, X. L. (2015). Generating domain-specific affective

ontology from Chinese reviews for sentiment analysis. Journal of Shanghai Jiaotong University

(Science), 20, 32-37.

Liu, R., An, Y., & Song, L. (2011). Chinese multi-document summarization based on opinion

similarity. Procedia Engineering, 15, 2026-2031.

Liu, Y., Jin, J., Ji, P., Harding, J. A., & Fung, R. Y. (2013). Identifying helpful online reviews: a

product designer’s perspective. Computer-Aided Design, 45(2), 180-194.

Lu, Y. (2013). Automatic topic identification of health-related messages in online health community

using text classification. SpringerPlus, 2(1), 309.

Luo, B., Zeng, J., & Duan, J. (2016). Emotion space model for classifying opinions in stock message

board. Expert Systems with Applications, 44, 138-146.

Lv, X., & El-Gohary, N. (2016). Text Analytics for Supporting Stakeholder Opinion Mining for Large-

scale Highway Projects. Procedia Engineering, 145, 518-524.

Ma, X., Lei, X., Zhao, G., & Qian, X. Rating prediction by exploring user’s preference and sentiment.

Multimedia Tools and Applications, 1-20.

Ma, Y., Chen, G., & Wei, Q. (2017). Finding users preferences from large-scale online reviews for

personalized recommendation. Electronic Commerce Research, 17 (1), 3-29.

Makazhanov, A., Rafiei, D., & Waqar, M. (2014). Predicting political preference of Twitter users.

Social Network Analysis and Mining, 4(1), 1-15.

Malik, H., & Shakshuki, E. M. (2016). Mining Collective Opinions for Comparison of Mobile Apps.

Procedia Computer Science, 94, 168-175.

Manaman, H. S., Jamali, S., & AleAhmad, A. (2016). Online reputation measurement of companies

based on user-generated content in online social networks. Computers in Human Behavior, 54,

94-100.

Manek, A. S., Shenoy, P. D., Mohan, M. C., & Venugopal, K. R. (2017). Aspect term extraction for

sentiment analysis in large movie reviews using Gini Index feature selection method and SVM

classifier. World Wide Web, 20(2), 135-154.


Mcnulty E. (2014, May 22). Understanding big data: The seven v's. Retrieved August 20, 2017, from

http://dataconomy.com/seven-vs-big-data/

Medhat, W., Hassan, A., & Korashy, H. (2014). Sentiment analysis algorithms and applications: A

survey. Ain Shams Engineering Journal, 5(4), 1093-1113.

Min, H. J., & Park, J. C. (2012). Identifying helpful reviews based on customer’s mentions about

experiences. Expert Systems with Applications, 39(15), 11830-11838.

Mohammad,S. M., Zhu, X., Kiritchenko, S., & Martin, J. (2015). Sentiment, emotion, purpose, and

style in electoral tweets. Information Processing &Management, 51(4), 480-499.

Moreo, A., Romero, M., Castro, J. L., & Zurita, J. M. (2012). Lexicon-based comments-oriented news

sentiment analyzer system. Expert Systems with Applications, 39(10), 9166-9180.

Mostafa, M. M. (2013). An emotional polarity analysis of consumers’ airline service tweets. Social

Network Analysis and Mining, 3(3), 635-649.

Mostafa, M. M. (2013). More than words: Social networks’ text mining for consumer brand

sentiments. Expert Systems with Applications, 40(10), 4241-4251.

Nath, V. (2003). Digital governance models: Moving towards good governance in developing

countries. E-commerce, 2(4). Retrieved August 20, 2017, from

http://www.amarc.org/documents/articles/nath-digital.pdf

Nayak, A., Pai, M. M., & Pai, R. M. (2016). Prediction models for indian stock market. Procedia


Neidhardt, J., Rümmele, N., & Werthner, H. (2017). Predicting happiness: user interactions and

sentiment analysis in an online travel forum. Information Technology & Tourism, Information

Technology & Tourism, 17(1), 101–119.

Nguyen, T. H., Shirai, K., & Velcin, J. (2015). Sentiment analysis on social media for stock movement

prediction. Expert Systems with Applications, 42(24), 9603-9611.

Nguyen, T., Phung, D., Dao, B., Venkatesh, S., & Berk, M. (2014). Affective and content analysis of

online depression communities. IEEE Transactions on Affective Computing, 5(3), 217-226.

O'Leary, D. E. (2016). On the relationship between number of votes and sentiment in crowdsourcing

ideas and comments for innovation: A case study of Canada's digital compass. Decision Support

Systems, 88, 28-37.

Ortigosa, A., Martín, J. M., & Carro, R. M. (2014). Sentiment analysis in Facebook and its application

to e-learning. Computers in Human Behavior, 31, 527-541.

Osimo, D., & Mureddu, F. (2012). Research challenge on opinion mining and sentiment analysis.

Universite de Paris-Sud, Laboratoire LIMSI-CNRS.

Pai, M. Y., Chu, H. C., Wang, S. C., & Chen, Y. M. (2013). Electronic word of mouth analysis for

service experience. Expert Systems with Applications, 40(6), 1993-2006.

Panasyuk, A., Yu, E. S. L., & Mehrotra, K. G. (2014). Controversial topic discovery on members of

congress with twitter. Procedia Computer Science, 36, 160-167.

Park, Y., & Lee, S. (2011). How to design and utilize online customer center to support new product

concept generation. Expert Systems with Applications, 38(8), 10638-10647.

Parkhe, V., & Biswas, B. (2016). Sentiment analysis of movie reviews: finding most important movie

aspects using driving factors. Soft Computing, 20(9), 3373-3379.


Peleja, F., Dias, P., Martins, F., & Magalhães, J. (2013). A recommender system for the TV on the web:

integrating unrated reviews and movie ratings. Multimedia Systems, 19(6), 543-558.

Pröllochs, N., Feuerriegel, S., & Neumann, D. (2016). Negation scope detection in sentiment analysis:

Decision support for news-driven trading. Decision Support Systems, 88, 67-75.

Qiang, J. P., Chen, P., Ding, W., Xie, F., & Wu, X. (2016). Multi-document summarization using

closed patterns. Knowledge-Based Systems, 99, 28-38.

Qiao, Y. N., Yong, Q., & Di, H. (2011). Tensor Field Model for higher-order information retrieval.

Journal of Systems and Software, 84(12), 2303-2313.

Ravi, K., & Ravi, V. (2015). A survey on opinion mining and sentiment analysis: Tasks, approaches

and applications. Knowledge-Based Systems, 89, 14-46.

Remus, R., Quasthoff, U., & Heyer, G. (2010, May). SentiWS-A Publicly Available German-language

Resource for Sentiment Analysis. In: Proceedings of the 7th International Language Resources

and Evaluation (LREC'10), pp. 1168-1171.

Reyes, A., & Rosso, P. (2012). Making objective decisions from subjective data: Detecting irony in

customer reviews. Decision Support Systems, 53(4), 754-760.

Rill, S., Reinel, D., Scheidt, J., & Zicari, R. V. (2014). Politwi: Early detection of emerging political

topics on twitter and the impact on concept-level sentiment analysis. Knowledge-Based Systems,

69, 24-33.

Rosa, R. L., Rodriguez, D. Z., & Bressan, G. (2015). Music recommendation system based on user's

sentiments extracted from social networks. IEEE Transactions on Consumer Electronics, 61(3),

359-367.

Rossetti, M., Stella, F., & Zanker, M. (2016). Analyzing user reviews in tourism with topic models.

Information Technology & Tourism, 16(1), 5-21.

Rout, J. K., Dalmia, A., Choo, K. K. R., Bakshi, S., & Jena, S. K. (2017). Revisiting semi-supervised

learning for online deceptive review detection. IEEE Access, 5(1), 1319-1327.

Sanchez-Monzon, J., Putzke, J., & Fischbach, K. (2011). Automatic generation of product association

networks using latent dirichlet allocation. Procedia-Social and Behavioral Sciences, 26, 63-75.

Sandoval, J., & Hernández, G. (2015). Computational visual analysis of the order book dynamics for

creating high-frequency foreign exchange trading strategies. Procedia Computer Science, 51,

1593-1602.

Scharkow, M. (2013). Thematic content analysis using supervised machine learning: An empirical

evaluation using German online news. Quality & Quantity, 47(2), 761-773.

Schumaker, R. P., Zhang, Y., Huang, C. N., & Chen, H. (2012). Evaluating sentiment in financial news

articles. Decision Support Systems, 53(3), 458-464.

Shen, C. W., & Kuo, C. J. (2015). Learning in massive open online courses: Evidence from social

media mining. Computers in Human Behavior, 51, 568-577.

Singh, P., Sawhney, R. S., & Kahlon, K. S. (2017). Sentiment analysis of demonetization of 500 &

1000 rupee banknotes by Indian government. ICT Express.

https://doi.org/10.1016/j.icte.2017.03.001

Smailović, J., Grčar, M., Lavrač, N., & Žnidaršič, M. (2014). Stream-based active learning for

sentiment analysis in the financial domain. Information Sciences, 285, 181-203.


Sohail, S. S., Siddiqui, J., & Ali, R. (2015). OWA based book recommendation technique. Procedia


Song, B., Lee, C., Yoon, B., & Park, Y. (2016). Diagnosing service quality using customer reviews: An

index approach based on sentiment and gap analyses. Service Business, 10(4), 775-798.

Sun, J., Wang, G., Cheng, X., & Fu, Y. (2015). Mining affective text to improve social media item

recommendation. Information Processing & Management, 51(4), 444-457.

Tayal, D. K., & Yadav, S. K. (2016). Sentiment analysis on social campaign “Swachh Bharat Abhiyan”

using unigram method. AI & Society, 1-13.

Teodorescu, H. N. (2015). Using analytics and social media for monitoring and mitigation of social

disasters. Procedia Engineering, 107, 325-334.

The Digital Governance Initiative. (n.d.). Digital Governance Concept. Retrieved August 20, 2017,

from http://www.digitalgovernance.org/index.php/concept

Tsai, F. S., & Kwee, A. T. (2011). Database optimization for novelty mining of business blogs. Expert

Systems with Applications, 38(9), 11040-11047.

Tsai, M. F., & Wang, C. J. (2017). On the risk prediction and analysis of soft information in finance

reports. European Journal of Operational Research, 257(1), 243-250.

Tsirakis, N., Poulopoulos, V., Tsantilas, P., & Varlamis, I. (2017). Large scale opinion mining for

social, news and blog data. Journal of Systems and Software, 127, 237-248.

United Nations Human Rights: Office of the High Commissioner. (n.d.). Good Governance and

Human Rights. Retrieved August 20, 2017, from

http://www.ohchr.org/EN/Issues/Development/GoodGovernance/Pages/GoodGovernanceIndex.aspxm

Valsamidis, S., Theodosiou, T., Kazanidis, I., & Nikolaidis, M. (2013). A framework for opinion

mining in blogs for agriculture. Procedia Technology, 8, 264-274.

Van de Kauter, M., Breesch, D., & Hoste, V. (2015). Fine-grained analysis of explicit and implicit

sentiment in financial news articles. Expert Systems with Applications, 42(11), 4999-5010.

Vinodhini, G., & Chandrasekaran, R. M. (2014). Measuring the quality of hybrid opinion mining

model for e-commerce application. Measurement, 55, 101-109.

Vinodhini, G., & Chandrasekaran, R. M. (2014). Opinion mining using principal component analysis

based ensemble model for e-commerce application. CSI transactions on ICT, 2(3), 169-179

Vinodhini, G., & Chandrasekaran, R. M. (2016). A comparative performance evaluation of neural

network based approach for sentiment classification of online reviews. Journal of King Saud

University-Computer and Information Sciences, 28(1), 2-12.

Wang, D., & Liu, Y. (2015). Opinion summarization on spontaneous conversations. Computer Speech

& Language, 34(1), 61-82

Wang, D., Zhu, S., & Li, T. (2013). SumView: A Web-based engine for summarizing product reviews

and customer opinions. Expert Systems with Applications, 40(1), 27-33.

Wang, T., Cai, Y., Leung, H. F., Lau, R. Y., Li, Q., & Min, H. (2014). Product aspect extraction

supervised with online domain knowledge. Knowledge-Based Systems, 71, 86-100.

Wang, Y. (2017). Stock market forecasting with financial micro-blog based on sentiment and time

series analysis. Journal of Shanghai Jiaotong University (Science), 22(2), 173-179.


Wei, C. P., Lin, W. B., Chen, H. C., An, W. Y., & Yeh, W. C. (2015). Finding experts in online forums

for enhancing knowledge sharing and accessibility. Computers in Human Behavior, 51, 325-

335.

Wikipedia contributors. (2017). Advocacy. In Wikipedia, The Free Encyclopedia. Retrieved August 20,

2017, from https://en.wikipedia.org/wiki/Advocacy

Wiley, M. T., Jin, C., Hristidis, V., & Esterling, K. M. (2014). Pharmaceutical drugs chatter on online

social networks. Journal of Biomedical Informatics, 49, 245-254.

Wöllmer, M., Weninger, F., Knaup, T., Schuller, B., Sun, C., Sagae, K., & Morency, L. P. (2013).

Youtube movie reviews: Sentiment analysis in an audio-visual context. IEEE Intelligent

Systems, 28(3), 46-53.

Wu, D. D., Zheng, L., & Olson, D. L. (2014). A decision support approach for online stock forum

sentiment analysis. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 44(8),

1077-1087.

Wu, J. L., Yu, L. C., & Chang, P. C. (2014). An intelligent stock trading system using comprehensive

features. Applied Soft Computing, 23, 39-50.

Wu, M., Wang, L., Li, M., & Long, H. (2014). An approach of product usability evaluation based on

Web mining in feature fatigue analysis. Computers & Industrial Engineering, 75, 230-238.

Xianghua, F., Guo, L., Yanyan, G., & Zhiqiang, W. (2013). Multi-aspect sentiment analysis for Chinese

online social reviews based on topic modeling and HowNet lexicon. Knowledge-Based Systems,

37, 186-195.

Xu, K., Liao, S. S., Li, J., & Song, Y. (2011). Mining comparative opinions from customer reviews for

Competitive Intelligence. Decision Support Systems, 50(4), 743-754.

Xue, W., Li, T., & Rishe, N. Aspect identification and ratings inference for hotel reviews. World Wide

Web, 20(1), 23-37.

Yaakub, M. R., Li, Y., & Zhang, J. (2013). Integration of sentiment analysis into customer relational

model: the importance of feature ontology and synonym. Procedia Technology, 11, 495-501.

Yan, D., Zhou, G., Zhao, X., Tian, Y., & Yang, F. (2016). Predicting stock using microblog moods.

China Communications, 13(8), 244-257.

Yan, G., He, W., Shen, J., & Tang, C. (2014). A bilingual approach for conducting Chinese and English

social media sentiment analysis. Computer Networks, 75, 491-503.

Yang, F. C., Lee, A. J., & Kuo, S. C. (2016). Mining health social media with sentiment analysis.

Journal of Medical Systems, 40(11), 236.

Yang, H. L., & Chao, A. F. (2015). Sentiment analysis for Chinese reviews of movies in multi-genre

based on morpheme-based features and collocations. Information Systems Frontiers, 17(6),

1335-1352.

Yu, X., Liu, Y., Huang, X., & An, A. (2012). Mining online reviews for predicting sales performance:

A case study in the movie domain. IEEE Transactions on Knowledge and Data engineering,

24(4), 720-734.

Yu, Y., & Wang, X. (2015). World cup 2014 in the twitter world: A big data analysis of sentiments in

US sports fans’ tweets. Computers in Human Behavior, 48, 392-400.

Yu, Y., Duan, W., & Cao, Q. (2013). The impact of social and conventional media on firm equity

value: A sentiment analysis approach. Decision Support Systems, 55(4), 919-926.


Yuan, H., Xu, W., Li, Q., & Lau, R. (2017). Topic sentiment mining for sales performance prediction

in e-commerce. Annals of Operations Research, 1-24.

Zakzouk, T. S., & Mathkour, H. I. (2012). Comparing text classifiers for sports news. Procedia

Technology, 1, 474-480.

Zhai, Z., Liu, B., Wang, J., Xu, H., & Jia, P. (2012). Product feature grouping for opinion mining.

IEEE Intelligent Systems, 27(4), 37-44.

Zhang, H., Sekhari, A., Ouzrout, Y., & Bouras, A. (2016). Jointly identifying opinion mining elements

and fuzzy measurement of opinion intensity to analyze product features. Engineering

Applications of Artificial Intelligence, 47, 122-139.

Zhang, L., Hua, K., Wang, H., Qian, G., & Zhang, L. (2014). Sentiment analysis on reviews of mobile

users. Procedia Computer Science, 34, 458-465.

Zhang, W., Li, C., Ye, Y., Li, W., & Ngai, E. W. (2015). Dynamic business network analysis for

correlated stock price movement prediction. IEEE Intelligent Systems, 30(2), 26-33.

Zhang, X., Cui, L., & Wang, Y. (2014). Commtrust: Computing multi-dimensional trust by mining e-

commerce feedback comments. IEEE Transactions on Knowledge and Data Engineering, 26(7),

1631-1643.

Zhang, Y., Chen, M., Huang, D., Wu, D., & Li, Y. (2017). iDoctor: Personalized and professionalized

medical recommendations based on hybrid matrix factorization. Future Generation Computer

Systems, 66, 30-35.

Zheng, X., Lin, Z., Wang, X., Lin, K. J., & Song, M. (2014). Incorporating appraisal expression

patterns into topic modeling for aspect and sentiment word identification. Knowledge-Based

Systems, 61, 29-47.

Zimmermann, M., Ntoutsi, E., & Spiliopoulou, M. (2015). Discovering and monitoring product

features and the opinions on them with OPINSTREAM. Neurocomputing, 150, 318-330.

Zitnik, M. (2012). Using sentiment analysis to improve business operations. XRDS: Crossroads, The

ACM Magazine for Students, 18(4), 42-43.

Bibliographic information of this paper for citing:

Kumar, Akshi, & Sharma, Abhilasha (2017). "Systematic Literature Review on Opinion Mining

of Big Data for Government Intelligence." Webology, 14(2), Article 156. Available at:

http://www.webology.org/2017/v14n2/a156.pdf

Copyright © 2017, Akshi Kumar and Abhilasha Sharma.