DOI QR코드

DOI QR Code

Using Support Vector Machine to Predict Political Affiliations on Twitter: Machine Learning approach

  • Muhammad Javed (Institute of Computing and Information Technology, Gomal University) ;
  • Kiran Hanif (Institute of Computing and Information Technology, Gomal University) ;
  • Arslan Ali Raza (Department of Computer Science, COMSATS University Islamabad, Vehari Campus) ;
  • Syeda Maryum Batool (Institute of Computing and Information Technology, Gomal University) ;
  • Syed Muhammad Ali Haider (Institute of Computing and Information Technology, Gomal University)
  • 투고 : 2024.05.05
  • 발행 : 2024.05.30

초록

The current study aimed to evaluate the effectiveness of using Support Vector Machine (SVM) for political affiliation classification. The system was designed to analyze the political tweets collected from Twitter and classify them as positive, negative, and neutral. The performance analysis of the SVM classifier was based on the calculation of metrics such as accuracy, precision, recall, and f1-score. The results showed that the classifier had high accuracy and f1-score, indicating its effectiveness in classifying the political tweets. The implementation of SVM in this study is based on the principle of Structural Risk Minimization (SRM), which endeavors to identify the maximum margin hyperplane between two classes of data. The results indicate that SVM can be a reliable classification approach for the analysis of political affiliations, possessing the capability to accurately categorize both linear and non-linear information using linear, polynomial or radial basis kernels. This paper provides a comprehensive overview of using SVM for political affiliation analysis and highlights the importance of using accurate classification methods in the field of political analysis.

키워드

참고문헌

  1. Javed.M, Ziaudin, S.Kamal, Nasir.J.A, Raza.A.A, Habib.A (2020). Socio monitoring framework (SMF): Efficient sentiment analysis through informal and native terms. International Journal of advanced and applied sciences, 7(12), 113-126.
  2. Boutet, A., Kim, H., &Yoneki, E. (2012). What's in Your Tweets? I Know Who You Supported in the UK 2010 General Election. ICWSM, 12, (pp. 411-414).
  3. Conover, M. D., Goncalves, B., Ratkiewicz, J., Flammini, A., &Menczer, F. (2011, October). Predicting the political alignment of twitter users. In Privacy, Security, Risk and Trust (PASSAT) and 2011 IEEE Third Inernational Conference on Social Computing (SocialCom), 2011 IEEE Third International Conference on (pp. 192-199).
  4. Ellen, J., &Parameswaran, S. (2011). Machine Learning for Author Affiliation within Web Forums--Using Statistical Techniques on NLP Features for Online Group Identification. In Machine Learning and Applications and Workshops (ICMLA), 2011 10th International Conference on(Vol. 1, pp. 100-105).
  5. Kalmegh, S. (2015). Analysis of WEKA data mining algorithm REPTree, Simple CART and Random Tree for classification of Indian news. Int. J. Innov. Sci. Eng. Technol, 2(2), (pp. 438-446)
  6. Sharma, P., &Moh, T. S. (2016, December). Prediction of indian election using sentiment analysis on hindi twitter. In Big Data (Big Data), 2016 International Conference on (pp. 1966-1971).
  7. Khatua, A., Khatua, A., Ghosh, K., &Chaki, N. (2015). Can# twitter_trends predict election results? Evidence from 2014 indian general election. In System Sciences (HICSS), 2015 48th Hawaii International Conference on (pp. 1676-1685).
  8. Wicaksono, A. J. (2016). A proposed method for predicting US presidential election by analyzing sentiment in social media. In Science in Information Technology (ICSITech), 2016 2nd International Conference on (pp. 276-280).
  9. Sahu, R. K. (2019). Election result prediction for indian political parties using sentiment analysis of twitter data. Journal of the Gujarat Research Society, 21(14s), 369-374.
  10. Nayak A., Nataranjan S. (2016) Comparative study of Naive Bayes, Support Vector Machine and Random Forest Classifiers in Sentiment Analysis of Twitter feeds. International Journal of Advanced Studies in Computer Science and Engineering (IJASCSE) Issue 1, Volume 5.
  11. Kim, S. M., &Hovy, E. (2007). Crystal: Analyzing predictive opinions on the web. In Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL).
  12. Mejova, Y., Srinivasan, P., & Boynton, B. (2013). Gop primary season on twitter: popular political sentiment in social media. In Proceedings of the sixth ACM international conference on Web search and data mining (pp. 517-526). ACM.
  13. Raza, A. A., Habib, A., Ashraf, J., & Javed, M. (2019). Semantic orientation based decision making framework for big data analysis of sporadic news events. Journal of Grid Computing, 17, 367-383.
  14. Macrohon, Julio Jerison E., Charlyn Nayve Villavicencio, X. Alphonse Inbaraj, and Jyh-Horng Jeng. "A Semi-Supervised Approach to Sentiment Analysis of Tweets during the 2022 Philippine Presidential Election." Information 13, no. 10 (2022): 484.
  15. Chauhan, P., Sharma, N., & Sikka, G. (2021). The emergence of social media data and sentiment analysis in election prediction. Journal of Ambient Intelligence and Humanized Computing, 12, 2601-2627. https://doi.org/10.1007/s12652-020-02423-y
  16. Javed, M., & Kamal, S. (2018). Normalization of unstructured and informal text in sentiment analysis. International Journal of Advanced Computer Science and Applications, 9(10).