octs 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · airline customer...

15
ISBN: 978-625-83525-1-0 OCTS 2019 International Conference on Operating Systems, Cyber Security, Engineering Technology & Applied Sciences December 14-15, 2019 / Istanbul, Turkey

Upload: others

Post on 25-Jun-2020

7 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

ISBN: 978-625-83525-1-0

OCTS 2019

International Conference on Operating Systems,Cyber Security, Engineering Technology & Applied Sciences

December 14-15, 2019 / Istanbul, Turkey

Page 2: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

SEASOCTS - 19Program Book and Schedule

International Conference on Operating Systems, Cyber Security,Engineering Technology & Applied Sciences

Date: December 14-15, 2019

Istanbul Gonen Hotel, Istanbul Turkey

Email [email protected]

Hosted by:

Page 3: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

TABLE OF CONTENTEditorial Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Conference Secretariat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Contact Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5AGENDA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6AGENDA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Improvement of e-Learning Platforms with a Profile Identification Web Service . . . . . . . . . . . . . . . . . 10A new convolutional neural network for Facial Expression Recognition . . . . . . . . . . . . . . . . . . . . . . 11Airline customer purchase segmentation using machine learning techniques . . . . . . . . . . . . . . . . . . . 12Banking Trojan Trickbot Analysis and Injectable Situations to Different . . . . . . . . . . . . . . . . . . . . . 13Upcoming Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3

Page 4: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

Editorial Board

Prof. Mohammed Abdel Razek, King Abdulazize University, Saudi Arabia

Prof. Badlishah Ahmad, Universiti Sultan Zainal Abidin (UniSZA), Malaysia

Prof. Zeeshan Ahmed, University of Connecticut Health Center, USA

Prof. Majida Alasady, University of Tikrit, Iraq

Prof. Modafar Ati, Abu Dhabi University, United Arab Emirates

Prof. Eduard Babulak, Fort Hays State University, USA

Prof. Konstantinos Blekas, University of Ioannina, Greece

Prof. Rodrigo Campos Bortoletto, Sao Paulo Federal Institute of Education, Brazil

Prof. Bong Jun Choi, The State University of New York (SUNY) Korea, Korea

Prof. Deepak Choudhary, LPU, India

Prof. George Dekoulis, Aerospace Engineering Institute, Cyprus

Prof. Darya Filatova, Kielce University of Technology, Poland

Prof. Atul Gonsai, Saurashtra Universtiy, India

Prof. Ching-Ting Hsu, University of Taipei, Taiwan

Prof. Alex Pappachen James, Nazarbayev University, Kazakhstan

Prof. Ming-shen Jian, National Formosa University, Taiwan

Prof. Li-Wei Kang, National Yunlin University of Science and Technology, Taiwan

Prof. Dimitrios Karras, Sterea Hellas Institute of Technology, Athens, Greece

Prof. Abdelmajid Khelil, Landshut University of Applied Sciences, Germany

Prof. Muralidhar Kulkarni, National Institute of Technology Karnataka, India

Prof. Datong Liu, Harbin Institute of Technology, P.R. China

Prof. Wenyu Liu, Huazhong University. of Sci. & Tech. Wuhan, P.R. China

Prof. Valeri Mladenov, Technical University of Sofia, Bulgaria

Prof. Marco Mugnaini, University of Siena, Italy

Prof. Yi Lu Murphey, University of Michigan, USA

Prof. Shashikant Patil, SVKM NMIMS Mumbai India, India

Prof. Grienggrai Rajchakit, Maejo University, Thailand

Prof. Rabie Ramadan, Cairo University, Egypt

Prof. Priya Ranjan, Amity University, India

Prof. Addisson Salazar, Universidad Politecnica de Valencia, Spain

Prof. Manuel Silva, Polytechnic Institute of Porto, Portugal

Prof. China Sonagiri, MRIET JNTUH Hyderabad, India

Prof. Hung-Min Sun, National Tsing Hua University, Taiwan

Prof. Theo Swart, University of Johannesburg, South Africa

Prof. G Viju, Karary University, Sudan

Prof. Renyong Wu, Hunan University, P.R. China

K. Martin Sagayam, Department Of ECE, Karunya Institute Of Technology & Sciences (Deemed to be University), India

4

Page 5: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

Conference Secretariat

Aminath Gayoom

[email protected]

Contact Details

[email protected]

[email protected]

5

Page 6: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

AGENDA- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

OCTSInternational Conference on Operating Systems, Cyber Security,Engineering Technology & Applied Sciences

Istanbul Gonen Hotel, Istanbul TurkeyDecember 14-15, 2019Event Objectives:

• Creating the conditions for concluding new agreements and partnerships with prestigious international institutionsin the academic field.

• Organization of events with international participation on the topic of internationalization of higher education,challenges of global approaches in the educational process and optimization of their development.

• SEAS aims to extend knowledge and innovation with the help of high quality journals.

Times Activity/Session

09:00 am 09:20 am Registration and Reception

09:20 am 09:30 am am Opening ceremony

09:30 am 09:40 am Welcome & Introductions

09:40 am 10:00 am Coffee & Networking

6

Page 7: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

AGENDA- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

OCTSInternational Conference on Operating Systems, Cyber Security,Engineering Technology & Applied Sciences

SESSION ACTIVITIES (10:00 am 11:00 pm)DAY 01: SaturdayDecember 14, 2019

Track A: Engineering, Technology & Applied SciencesRef ID Presenter Details Topic

OCTS-129-P1 Oral PresentationALI SERIDI, Labstic Laboratory,Computer science department, 8 May1945 University - Guelma; Algeria

Improvement of e-Learning Platforms with aProfile Identification Web Service

OCTS-129-P2 Oral PresentationYamina Bordjiba, Department of Com-puter Science, Labstic Laboratory, 8May 1945 University, Algeria

A new convolutional neural network for FacialExpression Recognition

OCTS-129-P9 Oral PresentationIlkay YELMEN, Vor Software andR&D Center, Turkey

Airline customer purchase segmentation usingmachine learning techniques

OCTS-129-P10 Oral PresentationRuveyda Celik, Kayseri Universty,Turkey

Banking Trojan Trickbot Analysis and In-jectable Situations to Different Banking WebSites

Closing Ceremony & Lunch (11:00 pm - 12:00 pm)

7

Page 8: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

OCTSInternational Conference on Operating Systems, Cyber Security,Engineering Technology & Applied Sciences

DAY 02: SundayDecember 15, 2019

The second day of the conference is for leisure activities. Participants and guests are free toexplore the city at their own.

8

Page 9: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

Track A: Engineering, Technology and Applied Sciences

9

Page 10: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

Improvement of e-Learning Platforms with a Profile Identifica-tion Web ServiceAli Seridi1*, Lynda Dib2, Riad Bourbia3, Yamina Bordjiba4

1,3,4 Department of Computer Science, Labstic Laboratory,8 May 1945 University, Guelma, Algeria, 2 Department ofComputer Science, Badji Mokhtar University Annaba, AlgeriaCorresponding email: a [email protected]

E-Learning environments are constantly evolving, while taking into account two main objectives, namely, on the onehand, to promote a process of knowledge transfer and acquisition in the most effective way possible and, on the otherhand, to motivate the learner to continue his or her studies and not to give up, by providing the most attractive and ap-propriate learning environment and styles (personality, cognitive level, and presentation preference). This could partlyreplace the lack of direct contact of face-to-face learning. In order to adapt the learning environment to the learner’sprofile, we must first detect the learner’s profile through the detection of all possible information concerning him in astatic way (by passing tests and filling out questionnaires on personal info) and dynamic (by capturing behaviour andfollowing the traces of learning, collaboration and communication with other learners, tutors and teachers) Our con-tribution consists in designing and implementing an independent module as a web service that can be solicited by anyLearning Management System (LMS) to provide it with an overview of the learners’ profile. This web service, calledthe Profile Identification Web Service (PIWS), will provide questionnaires for learners to complete and tests designedby specialists in psycho-pedagogy and educational science. The PIWS will return to the LMS, for each learner, thecognitive level, personality traits, learning and presentation preferences, as well as many other parameters that can beused by the LMS to adapt the learning style, course difficulty and appropriate presentation to learners. The choice ofprofile identification modelling, as a web service, is justified by the advantage offered by service-oriented architecture,which allows web services to be easily reused with any other system, even heterogeneous ones. In order to validate ourapproach, we developed and tested the PIWS on a remote server. It allows the LMS to identify the profile of each learner.

Index Terms: Profile identification, Adaptative e-learning systems, Service Oriented Architecture, Service basedeLearning, Web service.

10

Page 11: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

A new convolutional neural network for Facial Expression Recog-nitionYamina Bordjiba1*, Hayet Farida Merouani2, Nabiha Azizi3,Ali Seridi14

1,4Department of Computer Science, Labstic Laboratory,8 May 1945 University, Guelma, Algeria, 2Department ofComputer Science, LRI Laboratory, Badji Mokhtar University Annaba, Algeriav, 3Department of Computer Science,Labged Laboratory, Badji Mokhtar University Annaba, AlgeriaCorresponding email: [email protected]

Recently, interest in automatic facial expressions recognition has evolved with the rapid development of artificial in-telligent techniques. Facial expression recognition (FER) is an important problem, traditional methods such as SVM,Adaboost, and Random Forest have been proposed since the past decades. Recently, deep learning is widely appliedin the field of computer vision as well as in facial expression recognition. VGG16 is a convolutional neural networkmodel proposed by K. Simonyan and A. Zisserman, The model reaches an accuracy of 92.7In this work, we propose anarchitecture of a CNN for the recognition of the facial expression, inspired by the VGG16 model. Our CNN consistsof eleven (11) layers (after the elimination of the first two layers because of the size of the image already reduced atthe input, and the last three layers fully connected), including 10 convolutional layers using filters of different sizes andwe apply a batch normalization (BatchNormalization), then corrections ReLU are applied to eliminate negative values.At the end, we added an average Pooling Global layer, and we applied the Softmax function to calculate the rate of theseven expression classes (the six universal expressions and the neutral state. The richness of the layers, which constitutesthe VGG network, makes it possible to treat a maximum of parameters. Experimental results on the FER-2013 datasetshow that our model offers precision comparable to advanced methods and the higher level of traditional methods. Wesee through the confusion matrix of our CNN, that some classes are more oriented towards the class ”Anger”. Theseconfusions are probably due to the size and composition of the database.

Index Terms: Facial expressions Recognition, CNN, VGG16.

11

Page 12: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

Airline customer purchase segmentation using machine learningtechniquesSerpil Ustebay1*, lkay Yelmen2, Metin Zontul31stanbul Medeniyet University, 2Vor Software and R&D Center, 3Istanbul Arel University, TurkeyCorresponding email: [email protected]

Customer segmentation allows to differentiate marketing strategies and activities according to groups by dividing theaddressed market into homogeneous subgroups of customers with similar behaviors, needs and expectations. It providesaccess to the right customer with the right methods by knowing the customer better. In this study, customer segmenta-tion was performed using sales data from an airline between 2017 and 2018. Customers with similar sales tendencieswere included in the same cluster. Flight sales records for each customer have been used as features then this featuresare clustered with KMeans++ algorithm. Calinski Harabasz and distortion score metrics have been used to find theoptimum number of clusters. Also, unknown customers have been classifed by Random forest and K-NN classificationalgorithms, one of the most common machine learning techniques.

Index Terms: Airline Customer.

12

Page 13: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

Banking Trojan Trickbot Analysis and Injectable Situations toDifferentRuveyda Celik1*, Ali Gezer2

Kayseri Universty, TurkeyCorresponding email: [email protected]

Trickbot is a banking trojan designed to steal users’ personal information, private data. This virus first appeared in lateOctober 2016 and had targeted banks in Australia. Later, it has targeted other countries’ financial institutions, banks andcredit card providers. As of April 2017, there have been attacks on leading banks in the UK, US, Switzerland, Germany,Canada, New Zealand, France, Ireland. In late 2017, however, the encryption technique of Trickbot became very pop-ular, and therefore Trickbot malware code started some updates. Due to these updates, Trickbot is constantly evolvingand gaining new features. In addition, with new modules, Trickbot is able to spread itself and infect as many computersas possible. Trickbot spreads itself with spam e-mail campaigns which include e-mail attachments that are often hiddenas a Microsoft Office document with active macros. Executes and secretly downloads malware when the file is opened.In November 2018, Trickbot developed a new module for stealing credentials from the most known applications, such asFilezilla, Microsoft, Outlook, and WinSCP, and was added to such popular applications. From the beginning of January2019, the latest versions were available through seasonal themed spam, emails, as if coming from a major financial ad-visory firm. In addition, multiple search engines (Google Chrome, Mozilla, Firefox, Internet Explorer, Microsoft Edge)credentials and personal data has been revealed. In this study, we conducted analyses to reveal Trickbot behavior whileits code is injected into Banking web sites. We performed static and dynamic analyses using different tools to identifyTrickBot-associated streams and detect a TrickBot infection. As a result of the analyses, we learned that malware usersused one of two methods-obtaining credentials or a fake phishing page resembling a real account-to learn login detailsand access the account. In addition, our malicious virus analysis also revealed that Trickbot targets many internationalbanks. In our study, we also discovered that Trickbot updates itself and uses different interfaces and files to replicate.

Index Terms: Trickbot, Banking Web Sites, Malware Analysis

13

Page 14: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation

Upcoming Events

https://society-eas.com/aedac2020/

https://society-eas.com/idcei2020/

https://society-eas.com/retc2020/

https://society-eas.com/tsera2020/

https://society-eas.com/cmeia2020/

https://society-eas.com/eatnp2020/

14

Page 15: OCTS 2019 - society-eas.com › wp-content › uploads › 2020 › 01 › ... · Airline customer purchase segmentation using machine learning techniques OCTS-129-P10 Oral Presentation