The Journal of Korean Institute of Information Technology (한국정보기술학회논문지)

Korean Institute of Information Technology (한국정보기술학회)
권호 표지
DOI QR코드

DOI QR Code

A Technique of the Approval Rating Analysis for Political Party Using Opinion Mining

오피니언 마이닝을 통한 정당지지도 분석 기법

  • 김원상 (숭실대학교 SW특성화대학원) ;
  • 이종혁 (숭실대학교 SW특성화대학원) ;
  • 박제원 (숭실대학교 SW특성화대학원) ;
  • 최재현 (숭실대학교 SW특성화대학원)
  • Received : 2014.08.19
  • Accepted : 2014.09.20
  • Published : 2014.10.31
⟨ Previous Next ⟩

Abstract

With development of SNS, users communicate with each other and share information online. Especially, field of politics that is sensitive to public sentiment grasps flow of the public using data with various method. SO-PMI, one of the popular techniques, sometimes shows a low precision because this method only computes positive words and negative words regardless of subject. So, this paper suggests opinion mining applied distance between subject and opinion word about election that would be begin on June, 2014. First, this process collects tweets about candidates for mayor of Seoul, Incheon and governor of Gyeonggi Province using Twitter api. After that, classify collected tweets as mach1.0, part of speech using morphological analyzer and it deducts result through suggested algorithm in this paper. The result data of experiment makes correspond approximately real data. Moreover, F-Score of this technique's performance has improved by 16% and 15% of positive/negative ratio and previous study.

SNS 발달과 함께 사용자들은 온라인상에서 자유로운 의사소통과 정보공유를 한다. 특히, 대중의 민심에 민감한 분야인 정치 분야에서는 이러한 데이터를 토대로 다양한 방법으로 대중의 흐름을 파악한다. 본 논문에는 트위터 상에서 2014년 6월에 있었던 지방선거에 대한 정치적인 발언들을 수집하여 주어와 감정단어 간에 거리가중치를 적용한 오피니언 마이닝 분석 기법을 소개한다. 서울, 인천시장 및 경기도지사의 후보와 관련된 트윗을 수집한 후 Mach1.0 형태소 분석기로 품사를 분류 후, 주어와 감정단어간의 거리에 가중치를 적용하여 결과를 도출하였다. 실험 결과, 실제 선거 결과와 거의 일치 하였으며, 일반적으로 사용해왔던 긍정/부정의 비율의 값에 비해 F-score가 약 16%, 기존에 선거 극성 분류 기법에 비해 15% 개선되어 우수한 성능을 나타내었다.

Keywords

References

  1. Howard, P. N. and Hussain, M. M., "The role of digital media", Journal of Democracy, Vol. 22, Issue 3, pp. 35-48, July 2011. https://doi.org/10.1353/jod.2011.0041
  2. Wegrzyn-Wolska K. and Bougueroua L., "Tweets mining for french presidential election", 2012 Fourth International Conference on Computational Aspects of Social Networks (CASoN), pp. 138-143, Nov. 2012.
  3. Hong-gu Choi and Eenjun Hwang, "Emotion-based music recommendation system based on twitter document analysis", Journal of KISS(C): Computing Practices and Letters, Vol. 18, Issue 11, pp. 762-767, Nov. 2012.
  4. Jung-hwan Bae, Ji-eun Son, and Min Song, "Analysis of twitter for 2012 south korea presidential election by text mining techniques", Journal of Intelligent Information Systems, Vol. 19, Issue 3, pp. 141-156, Sep. 2012.
  5. Jho Whasun and Kim JeongYeon, "Political communication and civic participation through blogs and twitter", Journal of Cybercommunication Academic Society, Vol. 29, Issue 2, pp. 95-130, June 2012.
  6. Hae-Sung Lee and Joon-Hee Kwon, "Social network-based mobile context awareness recommendation technique", Journal of Korean Institute of Information Technology, Vol. 10, Issue 1, pp. 157-167, Jan. 2012
  7. Michael D. Conover, Bruno Goncalves, Jacob Ratkiewicz, Alessandro Flammini, and Filippo Menczer, "Predicting the Political Alignment of Twitter Users", 2011 IEEE Third International Conference on, pp. 192-199, Oct. 2011.
  8. Mahmood T, Iqbal T, Amin F, Lohanna W., and Mustafa A., "Mining Twitter big data to predict 2013 Pakistan election winner", Multi Topic Conference (INMIC), 2013 16th International, pp. 49-54, Dec. 2013.
  9. Kwangseob Shim and Jaehyung Yang, "High Speed Korean Morphological Analysis based on Adjacency Condition Check", Journal of KISS : Software and Applications, Vol. 31, Issue 1, pp. 89-99, Jan. 2004.
  10. Sukjoong Kim and Byung-Yeon Hwang, "Propensity analysis of political attitude of twitter users by extracting sentiment from timeline", Journal of Korea Multimedia Society, Vol. 17, Issue 1, pp. 43-51, Jan. 2014. https://doi.org/10.9717/kmms.2014.17.1.043
  11. Yeounoh Chung, Sola Hong, and Jee-hyong Lee, "Distributed twitter opinion mining system using MongoDB aggregation framework", HCI2014 Conference, pp. 461-464, Feb. 2014
  12. Xiaowen Ding, Bing Liu, and Phips S. Yu, "A Holistic Lexicon-Based Approach to Opinion Mining", Proceedings of the 2008 International Conference on Web Search and Data Mining (WSDM '08). ACM, pp. 231-240, Feb. 2008.
  13. Jong-Seok Song and Soo-Won Lee, "Automatic Construction of Positive/Negative Feature-Predicate Dictionary for Polarity Classification of Product Reviews", The Korean Institute of Information Scientists and Engineers: Software and Application, Vol. 38, Issue 3, pp. 157-168, Mar. 2011.
  14. http://korean.abcthesaurus.com/
  15. In-Jo Park and Kyung-Hwan Min, "Making a List of Korean Emotion Terms and Exploring Dimensions Underlying Them", Korean Journal of Social and Personality Psychology, Vol. 19, Issue 1, pp. 109-129, Feb. 2005.
  16. http://www.nec.go.kr