Developing Turkish Sentiment Lexicon: A Sentiment Analysis Study using Social Media Data

Year 2019, Issue: 16, 51 - 60, 31.08.2019

Serkan Ayvaz , Semra Yıldırım Yücel Batu Salman

https://doi.org/10.31590/ejosat.537085

Cited By: 4

Abstract

The objective of this study was to develop a comprehensive Sentiment Analysis Lexicon for Turkish language. The aim of this new lexicon is to facilitate effective sentiment analysis on Turkish social media posts. Previously developed lexicons have been examined and a comprehensive lexicon extending previous work has been developed. As an extension to the earlier work, the lexicon consist simple emoji characters and scoring infrastructure. To evaluate the performance of the Turkish Sentiment Analysis Lexicon, the tweets with certain hashtags were collected and analyzed. Analysis studies were carried out on two different topics. The first of these studies was done to measure the effect of weather changes on users. The data collected during the summer and winter were examined. The users who share Turkish post negative tweets in winter and appeared to share more positive posts in spring and summer compared to winter. The other analysis was conducted on Survivor TV Show, which has been aired on Turkish TVs, and placed at the top of the Trend Topic list for days. We observed that the users share negative posts about this program. Additionally, the habit of improper use of Turkish language on social media was noted in their posts.

Keywords

Emotion Analysis, Sentimental Analysis, Sentimental Lexicon for Turkish, Social Media, Twitter

Türkçe Duygu Kütüphanesi Geliştirme: Sosyal Medya Verileriyle Duygu Analizi Çalışması

Abstract

Bu çalışmanın amacı, Türkçe için kapsamlı yeni bir duygu kütüphanesi geliştirmektir. Bu yeni kütüphane ile Türkçe sosyal medya paylaşımlarında etkili duygu analizi çalışmalarının yapılmasına katkı sunmak hedeflenmektedir. Bu çalışmada, varolan diğer kütüphanelerden bazıları incelenmiş olup bunları genişleten kapsamlı bir kütüphane oluşturulmuştur. Daha önce yapılmış çalışmalara ek olarak kütüphaneye basit emoji karakterler ve puanlama altyapısı eklenmiştir. Geliştirilen Türkçe Duygu Kütüphanesi’nin verimliliğini ölçmek için Twitter’da belli etiketlerle oluşturulmuş paylaşımlar toplanmıştır ve bunların üzerinden duygu analizi çalışmaları yapılmıştır. Analiz çalışmaları, birbirinden bağımsız iki farklı konu başlığında gerçekleştirilmiştir. Bu analiz kapsamında yapılan çalışmalardan ilki hava değişikliklerinin kullanıcılar üzerindeki etkisini ölçmek adına yapılmıştır. Bu çalışma kapsamında yaz aylarında ve kış aylarında toplanan veri kümeleri incelenmiştir. Türkçe paylaşım yapan kullanıcıların kış aylarında daha olumsuz paylaşımlar yaptıkları, bahar ve yaz aylarında ise kış aylarına oranla daha olumlu paylaşımlar yaptıkları gözlemlenmiştir. Diğer analiz çalışması ise Türk televizyonlarında belirli bir süre boyunca yayınlanan, günlerce trend konu listesinin en üst sırasında yer alan Survivor adlı yarışma programı izleyicilerinin Türkçe paylaşımlarını konu almıştır. Analiz sonucunda kullanıcıların takip etmekte oldukları bu programla ilgili genelde olumsuz paylaşımlar yaptığı gözlemlenmiştir. Ayrıca, paylaşımlarda bozuk Türkçe kullanım alışkanlıkları tespit edilmiştir.

Keywords

Duygu Analizi, Sentimental Analiz, Türkçe Duygu Kütüphanesi, Sosyal Medya

References

• Akgul, M. (2011, July 19). Internet Yasaklari, Bilgi Toplumu ve Demokrasi.
• Aylin, Y. A., & Elif, S. (2017). Social Media in Social Organization. European Scientific Journal.
• Ayvaz, S., & Shiha, M. O. (2018). A Scalable Streaming Big Data Architecture for Real-Time Sentiment Analysis. In Proceedings of the 2018 2nd International Conference on Cloud and Big Data Computing (pp. 47–51). ACM.
• Christine B., W., & Girish J., G. (2007). Social networks in political campaigns: Facebook and the 2006 midterm elections. Annual Meeting of the American Political Science Association.
• Christine B., W., & Girish Jeff, G. (2008). What is a Social Network Worth? Facebook and Vote Share in the 2008 Presidential Primaries. Boston, MA: Department of International Studies and Government Bentley College.
• Cihan, C., Mustafa, E., Evren, K., H. Kaan, T., & Duygu, A. (n.d.). Sosyal Medyanın Politik Katılım ve Hareketlerdeki Rolü. Bilkent Universitesi, Bilgisayar Teknolojisi ve Bilişim Sistemleri Bölümü, Ankara.
• Claster, W. B., Dinh, H., & Cooper, M. (2010). Na #x00EF;ve Bayes and unsupervised artificial neural nets for Cancun tourism social media data analysis. In 2010 Second World Congress on Nature and Biologically Inspired Computing (NaBIC) (pp. 158–163). https://doi.org/10.1109/NABIC.2010.5716370
• Demirci, S. (2014). Emotion Analysis on Turkish Tweets. Middle East Technical University.
• Dokoohaki, N., Zikou, F., Gillblad, D., & Matskin, M. (2015). Predicting Swedish Elections with Twitter: A Case for Stochastic Link Structure Analysis. In Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining 2015 (pp. 1269–1276). New York, NY, USA: ACM. https://doi.org/10.1145/2808797.2808915
• Duncombe, C. (2011). The Twitter revolution? Social media, representation and crisis in Iran and Libya (pp. 1–12). Presented at the Australian Political Science Association Conference (APSA) 2011, Australian National University, School of Politics and International Relations. Retrieved from https://espace.library.uq.edu.au/view/UQ:269526
• Duwairi, R. M. (2015). Sentiment analysis for dialectical Arabic. In 2015 6th International Conference on Information and Communication Systems (ICICS) (pp. 166–170). https://doi.org/10.1109/IACS.2015.7103221
• Evirgen, E. (2016). Türkçe Tweetlerin Duygu Analizi. Bahcesehir Universitesi.
• Eyup Sercan, A., Caner, E., & Banu, D. (2017, September 26). Twitter verileri ile duygu analiz. Retrieved September 26, 2017, from http://docplayer.biz.tr/28952591-Pamukkale-universitesi-muhendislik-bilimleri-dergisi-pamukkale-university-journal-of-engineering-sciences.html
• Ganapathibhotla, M., & Lıu, B. (n.d.). Mining Opinions in Comparative Sentences (pp. 241–248). Presented at the Proceedings of the 22nd International Conference on Computational Linguistics - Volume 1 (COLING ’08), Vol. 1. Association for Computational Linguistics, Stroudsburg, PA, USA: Department of Computer Science University of Illinois at Chicago.
• Java, A., Song, X., Finin, T., & Tseng, B. (2007). Why We Twitter: Understanding Microblogging Usage and Communities. In Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 Workshop on Web Mining and Social Network Analysis (pp. 56–65). New York, NY, USA: ACM. https://doi.org/10.1145/1348549.1348556
• Kayahan, D., Sergin, A., & Banu, D. (n.d.). Twitter ile TV Program Reytinglerinin Belirlenmesi. Bilgisayar Mühendisligi Bölümü Yıldız Teknik Üniversitesi.
• Liu, B. (2010). Sentıment Analysıs and Subjectıvıty. Handbook of Natural Language Processing, 2, 627–666.
• Liu, B. (2012). Sentiment analysis and opinion mining. Synthesis Lectures on Human Language Technologies, 5(1), 1–167.
• Nassar, I. N., & Khamayseh, F. T. (2015). Constructing activity diagrams from Arabic user requirements using Natural Language Processing tool. In 2015 6th International Conference on Information and Communication Systems (ICICS) (pp. 50–54). https://doi.org/10.1109/IACS.2015.7103200
• Nasukawa, T., & Yi, J. (2003). Sentiment Analysis: Capturing Favorability Using Natural Language Processing. In Proceedings of the 2Nd International Conference on Knowledge Capture (pp. 70–77). New York, NY, USA: ACM. https://doi.org/10.1145/945645.945658
• Ozturk, N., & Ayvaz, S. (2017). Sentiment Analysis on Twitter: A Text Mining Approach to the Syrian Refugee Crisis.Telematics and Informatics. Bahcesehir University. https://doi.org/10.1016/j.tele.2017.10.006
• Pak, A., & Paroubek, P. (2010). Twitter as a corpus for sentiment analysis and opinion mining. In LREc (Vol. 10, pp. 1320–1326).
• Pang, B., & Lee, L. (2005). Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales (pp. 115–124). Presented at the Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics, Association for Computational Linguistics. Retrieved from https://scholar.google.com/citations?user=qCdLtIoAAAAJ
• Pang, B., Lee, L., & others. (2008). Opinion mining and sentiment analysis. Foundations and Trends® in Information Retrieval, 2(1–2), 1–135.
• Parycek, P. (2012). CeDEM 12 Conference for E-Democracy and Open Government 3-4 May 2012 Danube-University Krems, Austria. Edition-Donau-Univ. Krems.
• Setty, S., Jadi, R., Shaikh, S., Mattikalli, C., & Mudenagudi, U. (2014). Classification of facebook news feeds and sentiment analysis. In 2014 International Conference on Advances in Computing, Communications and Informatics (ICACCI) (pp. 18–23). https://doi.org/10.1109/ICACCI.2014.6968447
• Sevindi, B. I. (2013). Türkçe metinlerde denetimli ve sözlük tabanlı duygu analizi yaklaşımlarının karşılaştırılması. Gazi Universitesi Fen Bilimleri Enstitusu. Retrieved from http://acikarsiv.gazi.edu.tr/index.php?menu=2&secim=10&YayinBIK=9431
• Shiha, M., & Ayvaz, S. (2017). The effects of emoji in sentiment analysis. Int. J. Comput. Electr. Eng.(IJCEE.), 9(1), 360–369.
• Tang, L., & Liu, H. (2010). Toward Predicting Collective Behavior via Social Dimension Extraction. IEEE Intelligent Systems, 25(4), 19–25. https://doi.org/10.1109/MIS.2010.36
• Turney, P. D. (2002). Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews. arXiv:cs/0212032. Retrieved from http://arxiv.org/abs/cs/0212032
• Turney, P. D., & Mohammad, S. (2011). Crowdsourcing a Word–Emotion Association Lexicon. Institute for Information Technology, National Research Council Canada. Ottawa, Ontario, Canada, K1A 0R6.
• Vargas, J. (2008, November 20). Obama raised half a billion online. Washington Post. Retrieved from fromvoices.washingtonpost.com/44/2008/11/obama-raised-half-a-billion-on.html
• Wollmer, 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.