project synopsis on opinion mining
TRANSCRIPT
-
8/10/2019 Project Synopsis on Opinion Mining
1/4
1. Introduction
In recent past, due to existence of numerous forums, discussion groups, and
blogs, individual users are participating more actively and are generating vast
amount of new data termed as user-generated contents. These new Web contentsinclude customer reviews and blogs that express opinions on products and services
which are collectively referred to as customer feedback data on the Web. As
customer feedback on the Web influences other customers decisions, these
feedbacks have become an important source of information for businesses to take
into account when developing marketing and product development plans. Recent
works have shown that the distribution of an overwhelming majority of reviews
posted in online markets is bimodal. Reviews are either allotted an extremely high
rating or an extremely low rating. In such situations, the average numerical star
rating assigned to a product may not convey a lot of information to a prospective
buyer. Instead, the reader has to read the actual reviews to examine which of the
positive and which of the negative aspect of the product are of interest. Several
sentiment analysis approaches have proposed to tackle this challenge up to some
extent. However, most of the classical sentiment analysis mapping the customer
reviews into binary classes positive or negative,and thus fails to identify the
product features liked or disliked by the customers.
2. Motivation
This project results from the need of extracting useful information from thelarge amount of unstructured and unorganized data available on the web. Because
of the explosion of data on the internet , there is a growing need to analyze this
unprocessed data and obtain meaningful information that can be used in other
applications.
There is a need to implement a system which can help consumers to directly get the
positive or negative opinion about the products without wasting time in reading the
reviews as stated by other users of those products. In this project, a framework has
been presented which first extracts the feature, modifier and opinion from thedataset and then using clustering mechanism divides them into discrete clusters on
the basis of users opinion, in which the intra-cluster similarity between the
features are high whereas the inter-cluster similarity is very low.
-
8/10/2019 Project Synopsis on Opinion Mining
2/4
3. Objective
1)To design and in feature based clustering techniques in sentiment analysis to
improve customer review summarization.
2)To process and analyze twitter or Facebook feeds to determine the responses and
feedbacks of the customers. Using sentiment analysis , we can determine the
content of the posts and how many customers have given positive or negative
reviews.
3)To use sentiment analysis and opinion mining to analyze customer reviews about
a specific product or service. We can determine how many users liked/disliked the
product/service, what are the strong and weak points of the product reviewed.
As an example , we can analyze the customer feedbacks about a smartphone. Usingsentiment analysis we can determine how many customers described the product
as good and how many disliked it. The positive features like battery , LCD display ,
RAM ,etc. that the users have rated high can be displayed in accordance with their
rankings. Similarly, the drawbacks of the product as d escribed by the customers
can be listed with their rankings.
4)To use opinion mining in improving the efficiency of web mining. Company
officials can directly analyze the general response and feedback of the customers
about their product or service without spending hours over reading the reviews
manually.
5) To implement a system which helps consumers to directly get the positive or
negative opinion about the products without wasting time in reading the reviews as
stated by other users of those products.
-
8/10/2019 Project Synopsis on Opinion Mining
3/4
4. Scope of the project
Fig. 1 presents the architectural details of the proposed opinion mining system,
which consists of five major modules Document Processor, Subjectivity/
Objectivity Analyzer, Document Parser, Feature and Opinion Learner, andReview
Summarizer and Visualizer. The working principles of these components are
explained in the following steps:
1) First step involves the collecting of review documents from various sources like e-
commerce websites such as Flipkart, amazon, etc. and social networking sites like
twitter,Facebook, etc.
2) In next step,Document Processor and Subjectivity/Objectivity Analyzer module is
employed, which consists of a Markup Language (ML) tag filter that divides an
unstructured web document into individual record-size chunks, cleans them byremoving ML tags, and presents them as individual unstructured record documents
for further processing.
3) ThenDocument Parser, and Feature and Opinion Learner module is
implemented. TheDocument Parser module uses Stanford parser, which assigns
-
8/10/2019 Project Synopsis on Opinion Mining
4/4