Science of Emotion and Sensibility (감성과학)

Korean Society for Emotion and Sensibility (한국감성과학회)
권호 표지
DOI QR코드

DOI QR Code

Research on Sentiment Analysis in Social Media App Reviews: Focusing on Instagram

소셜 미디어 앱 리뷰에서의 감성 분석 연구: 인스타그램 중심으로

  • Wen-Qi Li ;
  • Yu-Hang Wu
  • 이문기 (부경대학교 미디어케뮤니게이션학과) ;
  • 우위항 (부경대학교 미디어케뮤니게이션학과)
  • Received : 2023.12.20
  • Accepted : 2024.02.27
  • Published : 2024.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

This study aimed to gain valuable insights into the performance and user satisfaction of applications (apps) through a thorough analysis of Instagram user reviews collected from Google Play. The study utilized text mining and sentiment analysis techniques and systematically identified emotions and opinions embedded in user reviews to deeply understand the areas of improvement and user experiences of the app. It analyzes how Instagram reviews reflect the diverse experiences of users and how they reveal the strengths and weaknesses of the app. For this purpose, sentiment analysis using the naive Bayes algorithm was conducted, and the results were expected to aid in the improvement of Instagram's services. In addition, the study aimed to assist developers in better understanding and utilizing user feedback, ultimately contributing to enhanced user satisfaction. This study explored the complex relationship between social media usage patterns and user opinions by seeking ways to provide a better user experience through these insights.

본 연구는 Google Play에서 수집된 Instagram 사용자 리뷰에 대한 심층 분석을 통해, 이 연구는 애플리케이션의 성능과 사용자 만족도에 대한 중요한 통찰력을 얻고자 한다. 텍스트 마이닝과 감성 분석 기술을 활용하여 사용자 리뷰에 담긴 감성과 의견을 체계적으로 파악하며, 이를 통해 앱의 개선점과 사용자 경험을 깊이 이해하려고 한다. 인스타그램 리뷰가 사용자들의 다양한 경험을 어떻게 반영하는지, 그리고 앱의 장단점을 어떻게 드러내는지를 분석한다. 이를 위해 나이브 베이즈 알고리즘을 사용한 감성 분석을 수행하며, 이 결과는 인스타그램 서비스 개선에 도움이 될 것으로 기대된다. 연구는 또한 개발자들이 사용자 피드백을 더 잘 이해하고 활용하는 데 도움을 주며, 결국 사용자 만족도를 향상시키는 데 기여할 것으로 예상된다. 이 연구는 소셜 미디어 사용 패턴과 사용자 의견의 복잡한 관계를 탐색하고, 이를 통해 더 나은 사용자 경험을 제공하는 방안을 모색한다.

Keywords

References

  1. Bengio, Y., Ducharme, R., Vincent, P., & Jauvin, C. (2003). A neural probabilistic language model. Journal of Machine Learning Research, 3, 1137-1155. DOI: 10.1162/153244303322533223
  2. Byeon, H. M., & Shim, S. W. (2020). The relationship between engagement with social media and social media advertising: The differentiating role of platform type (Instagram, Facebook and Youtube). Journal of Advertising & PR Practical Research, 13(2), 58-90. DOI: 10.21331/jprapr.2020.13.2.003
  3. Chung, D. (2022). A case study of visualizing emotions with social media emotion analysis - focused on media art cases. Archives of Design Research, 35(1), 237-257. DOI: 10.15187/adr.2022.02.35.1.237
  4. Chung, J., Yoo, K., & Koo, C. (2017). A study on social media sentiment analysis for exploring public opinions related to education policies. National Information Society Agency, 24(4), 3-16. DOI: 10.22693/ NIAIP.2017.24.4.003
  5. Choi, S.-J., Son, M.-Y., & Kim, Y.-H. (2016). Associated keyword recommendation system for keyword-based blog marketing. KIISE Transactions on Computing Practices, 22(5), 246-251.
  6. Hong, T., Kim, J., & Hyun, S. J. (2016). A user sentiment classification using instagram image and text analysis. Journal of Smart Media, 5(1), 61-68.
  7. Jacobi, C., van Atteveldt, W., & Welbers, K. (2015). Quantitative analysis of large amounts of journalistic texts using topic modelling. Digital Journalism, 4(1), 1-18. DOI: 10.1080/21670811.2015.1093271
  8. Jang, K., Park, S., & Kim, W. J. (2015). Automatic construction of a negative/positive corpus and emotional classification using the internet emotional sign. Journal of KIISE, 42(4), 512-521. DOI: 10.5626/JOK.2015.42.4.512
  9. Jeong, J, W., Lee, J. M., & Choi, S. Y. (2018). Analysis of news regarding the disabled labor using text mining techniques. Reinterpretation of Disability, 1, 48-100.
  10. Kang, C. M., Eo, K. S., & Lee, K. C. (2022). Investigating the performance of bayesian-based feature selection and classification approach to social media sentiment analysis. Journal of Management Information Systems Research, 24(1), 1-19.
  11. Kang, S., Kwon, B., Kwon, C., Park, S., & Yun, I. (2018). Development of incident detection algorithm using naive bayes classification. The Journal of The Korea Institute of Intelligent Transport Systems, 17(6), 25-39. DOI: 10.12815/kits.2018.17.6.25
  12. Kim, I.-G., Kim, H.-M., Lim, B., & Lee, K.-K. (2016). Relationship between result of sentiment analysis and user satisfaction - the case of Korean meteorological administration. The Journal of the Korea Contents Association, 16(10), 393-402. DOI: 10.5392/JKCA.2016.16.10.393
  13. Kim, J. I., Park, S. J., Kim, H. J., Choi, J. H., Kim, H. I., & Kim, P. K. (2020). Sensitivity identification method for new words of social media based on naive bayes classification. Smart Media Journal, 9(1), 51-59. DOI: 10.30693/SMJ.2020.9.1.51
  14. Kim, N., & Park, J. (2012). Personal information detection by using naive bayes methodology. Journal of Intelligence and Information Systems, 18(1), 91-107. DOI: 10.13088/jiis.2012.18.1.091
  15. Kim, R. A., & Han, E. K. (2016). The impacts of use motivation of instagram brand account on satisfaction and continuous use intention focusing on the mediating effect of flow. Journal of Advertising Research, 111, 5-39. DOI: 10.16914/ar.2016.111.5
  16. Kim, S., Lee, Y. J., Shin, J., & Park, K. Y. (2019). Text Mining for Economic Analysis. Bank of Korea WP, 2019(18), 1-66.
  17. Lee, J. H., & Gil, W. Y. (2019). News agenda classification and media diversity analysis using topic modeling: Through the analysis of news on the presidential new year press conference. Korean Journal of Broadcasting Studies, 33(1), 161-196. DOI: 10.22876/kab.2019.33.1.005
  18. Lee, S., & Kim, H. J. (2009). Keyword extraction from news corpus using modified TF-IDF. Journal of the Korean Electronic Commerce Association, 14(4), 59-73.
  19. Lee, S.-M., & Hong, S.-G. (2020). Policy agenda proposals from text mining analysis of patents and news articles. Journal of Digital Convergence, 18(3), 1-12. DOI: 10.14400/JDC.2020.18.3.001
  20. Lee, S. M., Ryu, S. E., & Ahn, S. (2020). Mass media and social media agenda analysis using text mining: focused on '5-day rotation mask distribution system'. Journal of Korean Contents Association, 20(6), 460-469. DOI: 10.5392/JKCA. 2020.20.06.460
  21. Salton, G., & McGill, M. J. (1983). Introduction to modern information retrieval. McGraw-Hill Book Co., New York.
  22. Yun, Y., Jo, J., Hur, Y., & Lim, H. (2017). A comparative analysis of cognitive change about big data using social media data analysis. Journal of Software & Data Engineering (Information Processing Society Journal), 6(7), 371-378. DOI: 10.3745/KTSDE.2017.6.7.371