臺灣博碩士論文加值系統

A wide variety of events occur on daily basis that triggers the responses, opinions and feelings of the public. Such occurrences usually form part of public news issues which tend to be broadcasted by news media companies through different news forums such as social media, websites, television and others. To investigate the perceptions of the public towards the different news affairs, sentiment analysis is performed on the responses of the public on each news content. Machine learning algorithms are employed to perform the investigations and predictions on the public opinions and perceptions from the different sources of the news issues like Facebook.

Social media platforms like Facebook and Twitter are expansively utilized by people and establishments such as news media companies and governments to share information. Facebook is the most used social networking site in the world with more than 2.3 billion active users a month and an average of more than 1.5 billion active users daily as of March 31, 2019. Facebook is one of the social media which allows its members to post status updates on their timelines and comments on other user’s posts. Facebook allows application developers to access public Facebook posts and comments. NETVIZZ is a data collection and extraction tool that enables researchers to export data in standard file formats from different sections of the Facebook social networking service. Sentiment Analysis (SA) is one small application of Natural Language Processing, computational linguistics, and text analytics to find out and extract subjective information from social media.

Therefore, in this thesis, Facebook posts and comments are extracted and sentiment analysis is implemented on them. Based on the results of the sentiment analysis, predictions can be drawn to determine the perceptions of the public on the Facebook News posts. Machine learning techniques, Naïve Bayes, Support Vector Machine and Logistic Regression are compared to determine the best classifier which is further employed to do the sentiment analysis of the news comments and determine the general views of the public from the text they post on Facebook.

Abstract…………………………………………………………………………………… i
Acknowledgements………………………………………………………………………. iii
Table of Contents………………………………………………………………………… iv
List of tables……………………………………………………………………………… vi
List of figures…………………………………………………………………………….. vii
1. Chapter 1: Introduction………………………………………………………………. 1
1.1. Background……………………………………………………………………… 1
1.2. Related Works…………………………………………………………………... 3
1.3. Problem Overview………………………………………………………………. 4
1.4. Contributions……………………………………………………………………... 4
1.5. Structure of the thesis…………………………………………………………… 5
2. Chapter 2: Literature Review………………………………………………………… 6
2.1. Sentiment Analysis……………………………………………………………… 6
2.1.1. Naïve Bayes……………………………………………………………… 7
2.1.2. Support Vector Machine…………………………………………………. 8
2.1.3. Logistic Regression………………………………………………………. 8
2.2. Feature Extraction………………………………………………………………. 9
2.3. Term Frequency – Inverse Document Frequency (TF-IDF)……………………. 10
2.4. Term Correlation………………………………………………………………... 11
2.5. About Facebook………………………………………………………………… 11
2.5.1. NETVIZZ………………………………………………………………… 12
3. Chapter 3: Methodology……………………………………………………………… 14
3.1. Downloading Facebook Data……………………………………………………. 14
3.2. Proposed Model Architecture…………………………………………...……….. 16
3.3. Data Pre-processing……………………………………………………………… 17
3.3.1. Tokenizing and Removing Stopwords…………………………………… 17
3.3.2. Stemming………………………………………………………………… 18
3.4. Data Splitting……………………………………………………………………. 18
3.5. Feature Extraction……………………………………………………..………… 19
3.6. Performance Measure…………………………………………………..……….. 21
4. Chapter 4: Analysis and Discussion of Results………………………………..…….. 23
4.1. Presentation of Results……………………………………………………...…… 23
4.1.1. Sentiment analysis categories…………………………………………… 23
4.1.2. Term Correlation………………………………………………………… 26
4.1.3. Training and prediction with Naïve Bayes………………………………. 27
4.1.4. Training and prediction with Support Vector Machine………………….. 28
4.1.5. Training and prediction with Logistic Regression……………………….. 28
4.1.6. Comparison of Machine Learning Classifiers……………………………. 29
4.2. Evaluation of Results…………………………………………………………….. 31
4.3. Discussion of Results…………………………………………………………….. 32
4.4. Limitations……………………………………………………………………….. 32
5. Chapter 5: Conclusion and Future Works…….……………………………………… 33
5.1. Conclusion………………………………………………………………………. 33
5.2. Future Works……………………………………………………………………. 33
References………………………………………………………………………………... 34

[1]N. A. Abdulla, N. A. Ahmed, M. A. Shehab, and M. Al-Ayyoub, “Arabic sentiment analysis: Lexicon-based and corpus-based,” in Applied Electrical Engineering and Computing Technologies (AEECT), 2013. IEEE Jordan Conference on. IEEE, 2013, pp. 1–6.
[2] Facebook, (15th May, 2019) https://newsroom.fb.com/company-info/
[3]W. Medhat, A. Hassan and H. Korashy, “Sentiment analysis algorithms and applications: A survey,” Ain Shams Engineering Journal (2014) 5, 1093-1113.
[4]P. Santidhanyaroj, T. A. Khan, C. M. Gelowitz and L. Benedicenti, “A Sentiment Analysis Prototype System for Social Network Data,” IEEE, 2014.
[5]A. Trilla and F. Alías, “Sentence-based sentiment analysis for expressive text-to-speech,” Audio, Speech, Lang. Process. IEEE Trans., vol. 21, no. 2, pp. 223–233, 2013.
[6]T. T. Nguyen and A. G. Kravets, “Analysis of the Social Network Facebook Comments,” 7th International Conference on Information, Intelligence, Systems & Applications (IISA). IEEE, 2016.
[7]A. Jose, “Twitter Sentiment Analysis,” Seminar Report, National Institute of Technology Calicut, 2010.
[8]D. I. G. Amalarethinam and V. J. Nirmal, “Real-Time Sentiment Prediction on Streaming Social Network Data Using In-Memory Processing,” World Congress on Computing and Communication Technologies (WCCCT). IEEE, 2017.
[9]A. Noureen, U. Qamar and M. Ali, “Semantic Analysis of Social Media and Associated Psychotic Behavior,” 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD 2017). IEEE, 2017.
[10]A. Ul Hassan, J. Hussain, M. Hussain, M. Sadiq and S. Lee, “Sentiment Analysis of Social Networking Sites (SNS) Data using Machine Learning Approach for Measurement of Depression,” IEEE, 2017.
[11]A. Elouardighi, M. Maghfour, H. Hammia and F. Aazi, “A Machine Learning Approach for Sentiment Analysis in the Standard or Dialectal Arabic Facebook Comments,” IEEE, 2017.
[12]A. Prabhat and V. Khullar, “Sentiment classification on Big Data using Naïve Bayes and Logistic Regression,” 2017 International Conference on Computer Communication and Informatics (ICCCI – 2017), Jan. 05 – 07, 2017, Coimbatore, INDIA.
[13]H. Liu and H. Motoda, “Feature selection for knowledge discovery and data mining.” Springer Science & Business Media, 2012, vol. 454.
[14]F. M. bin Naina Hanif and G. A. Putri Saptawati, “Correlation Analysis of User Influence and Sentiment on Twitter Data,” IEEE, 2014
[15]B. Rieder, “Studying Facebook via data extraction: The Netvizz application.” in WebSci, H. C. Davis, H. Halpin, A. Pentland, M. Bernstein, and L. A. Adamic, Eds. ACM, 2013, pp. 346–355. [Online]. Available: http://dblp.unitrier.de/db/conf/websci/websci2013.htmlRieder13
[16]D. Terrana, A. Augello and G. Pilato, “Facebook Users Relationships Analysis based on Sentiment Classification,” 2014 IEEE International Conference on Semantic Computing.
[17]A. Baj-Rogowska, “Sentiment Analysis of Facebook Posts: the Uber case,” The 8th IEEE International Conference on Intelligent Computing and Information Systems (ICICIS 2017), IEEE, 2017
[18]S. Akter and M. T. Aziz, “Sentiment Analysis On Facebook Group Using Lexicon Based Approach,” IEEE, 2016.
[19]K.C. Lin, S.H. Wu, L.P. Chen, T. Ku and G.D. Chen, “Mining the User Clusters on Facebook Fan Pages Based on Topic and Sentiment Analysis,” IEEE IRI 2014, August 13 – 15, 2014, San Francisco, Carlifornia, USA, IEEE, 2014.
[20]J. Ahktar, and S. Soria. 2009. “Sentiment Analysis: Facebook Status Messages.” Stanford University, Technical Report.
[21]V. B. Vaghela and B. M. Jadav, “Analysis of Various Sentiment Classification Techniques,” International Journal of Computer Applications (0975 – 8887) Volume 140 – No. 3, April, 2016.