Twitter Users’ Emotion, Emoticons and Scaling Metrics Based Categoric Interaction Analysis

Abstract

The popularity and use of social networks has also begun to increase in parallel with the worldwide increasing accessibility and means of access to the Internet. As one of the world's most popular social networks, Twitter is a platform where users are interacting through follow-up, sharing, messaging and appreciation tools, sharing their ideas and emotions in a variety of individual and corporate contexts. Therefore, Twitter is intense, dynamic and always an up-to-date data source. Identifying and correlating the physical and emotional interaction of users can be valuable in political, social, academic and commercial aspects. Users' physical networking with each other and emotional analysis can be done with many tools and applications. The character, tendency and impact analysis of the users can be used in the development of business intelligence applications and in the determination of social strategies. In this study, a large Twitter user group is divided into four categories: political, Entertainment, Sports, Trade Marks. Then, the physical and emotional interaction of each category was revealed.  The Physical interaction metrics determined as centrality, intensity, reciprocity and modularity while emotional interaction metrics were determined as resistance, passion, reach and emotionality. Positive, negative and neutral states of sharing were discussed in emotional measurement. Beside that, emoji-containing tweets have been transformed into texts and are especially included in emotion analysis. After all the metrics were calculated, physical and emotional interaction structures and overlap rates were revealed using "Interaction and Semantic Clustering Based Multinetwork Analysis" method.

Keywords

• Twitter
• Impact Quality Scaling Criteria
• Semantic and Emotion Analysis
• Interaction Analysis

References

1. [1] S. Kemp, “We are Social Global Digital Report 2018” (Online). Available: https://wearesocial.com/uk/blog/2018/01/global-digital-report-2018 last retrieved on March 30, 2018.
2. [2] Statista, “Most Famous Social Networks 2018” (Online). Available: https://www.statista.com/statistics/272014/global-social-networks-ranked-by-number-of-users/ last retrieved on March 30, 2018
3. [3] Global Web Index, “Global Social Media Research Summary 2017”(Online) Available: http://www.smartinsights.com/social-media-marketing/social-media-strategy/new-global-social-media-research/ ,last retrieved on March 30, 2018.
4. [4] H. H. Huang, H. C. Yang, “Semantic Clustering-Based Community Detection in an Evolving Social Network” in 6. International Conference on Genetic and Evolutionary Computing, IEEE, 2012.
5. [5] Y. Sun, J. Tang, J. Han, M.Gupta, and B. Zhao. “Community Evolution Detection in Dynamic Heterogeneous Information Networks” in Proc. KDD MLG, 2010.
6. [6] M. J. Preisendorfer, “Social Media Emoji Analysis, Correlations and Trust Modeling”, MSc. Thesis, SUNY Polytechnic Institute, 2018.
7. [7] P. K. Novak, J. Smailovic, B. Sluban, I. Mozetic, “Sentiment of Emojis” arXiv:1509.07761v2 [cs.CL], 2015.
8. [8] L. Zhao, C. Zeng, “Using Neutral Networks to Predict Emoji Usage from Twitter Data”, Stanford, 2017.

9. [9] R. Socher, A. Perelygin, J. Y. Wu, J. Chuang..,”Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank” Stanford, 2015.
10. [10] G. Guibon, M. Ochs, P. Bellot, “From Emojis to Sentiment Analysis”, WACAI, France, 2016.
11. [11] A. Bartolome, M. Bock, R. K. Vinayagam, R. Krishnamurthy, “Sentiment and Topic Analysis”, CS 6604 Digital Libraries, F. Report, 2017.
12. [12] A. Yeşilyurt, Ş.E. Şeker, “Metin Madenciliği Yöntemleri ve Twitter Duygu Analizi”, YBS Sözlük, Cilt 4, Sayı 2, 2017.
13. [13] J. A. Iglesias, A. G. Cuerva, A. Ledezma, A.Sanchis, “Social Network Analysis: Evolving Twitter Mining” in International Conference on Systems, Man, and Cybernetics, SMC, Budapeşt, Hungary, 2016.
14. [14] Netlytic, (Online), https://netlytic.org/home/?page_id=10834, last retrieved on March 30, 2018.
15. [15] Wikipedia, (Online), https://wikipedia.org/wiki/Metrics/, last retrieved on March 30, 2018.
16. [16] Tweet Sentiment Vizualization (Online), https://www.csc2.ncsu.edu/faculty/healey/tweet_viz/tweet_app/ last retrieved on March 30, 2018.
17. [17] B. Hutchins, “The Emoji Infographic: Stats to Back Up Your Obsession”, https://www.meltwater.com/blog/the-emoji-infographic-stats-to-back-up-your-obsession/ last retrieved on: March, 2018
18. [18] B. Shaul, ”Report: 92% of Online Consumers Use Emoji (Infographic)“, http://www.adweek.com/digital/report-92-of-online-consumers-use-emoji-infographic/, last retrieved on 2018
19. [19] N. Shah, “Introducing emoji targeting”, https://blog.twitter.com/marketing/en_us/a/2016/introducing-emoji-targeting.html, last retrieved on 2018.
20. [20] Oxford Dictionary (Online), “Oxford Dictionaries Word of the Year 2015”, http://blog.oxforddictionaries.com/2015/11/word-of-the-year-2015-emoji/, last retrieved on 2018.
21. [21] Emoji Translator (Online), http://superemojitranslator.com /emoji-translate, Last retrieved on March, 2018.
22. [22] Sentigem (Online), http://sentigem.com/#!, Last retrieved on March 2018.
23. [23] SocialMention (Online), http://www.socialmention.com, Last retrieved on March, 2018.