Sentiment-based Topic Model for Mining Usability Issues and Failures with User Products
https://doi.org/10.15514/ISPRAS-2015-27(4)-6
Abstract
This paper describes an approach to problem phrase extraction from texts that contain user experience with products. User reviews from online resources, that describe actual difficulties in use of products in addition to sentiment-oriented phrases, affect on other people's purpose decisions. In this paper we present two probabilistic graphical models which aims to extract problems with products from online reviews. We incorporate information about problem phrases with words’ sentiment polarities (negative, neutral or positive). The proposed models learn a distribution over words, associated with topics, both sentiment and problem labels. The algorithms achieve a better performance in comparison to several state-of-the-art models in terms of the likelihood of a held-out test and in terms of an accuracy of classification results, evaluated on reviews from several different domains in English and Russian. Our contribution is that summarizing sentiment and problem information about words with reviews’ topics by the model's asymmetric priors gives an improvement for problem phrase extraction from user reviews.
References
1. Sabirova I.M. Kachestvo - kluchevoy factor obespecheniya konkurentosposobnosti productov i uslug v usloviyah rynochnoy economiki [Quality-key factor of ensuring competition of products and services in the conditions of market economy]. Avtomatizaciya i upravlenie v tehnicheskih sistemah [Automation and management in technical systems], vol. 1, 2015, pp. 181-190. (In Russian)
2. Gupta N. K. Extracting descriptions of problems with product and services from twitter data. Proceedings of the 3rd Workshop on Social Web Search and Mining (SWSM2011). Beijing, China, 2011.
3. Solovyev V., Ivanov V. Dictionary-Based Problem Phrase Extraction from User Reviews. Text, Speech and Dialogue, Springer International Publishing, 2014, vol. 225-232.
4. Moghaddam S. Beyond Sentiment Analysis: Mining Defects and Improvements from Customer Feedback. Advances in Information Retrieval, Springer International Publishing, 2015, PP. 400-410.
5. Tutubalina E. Target-Based Topic Model for Problem Phrase Extraction. Advances in Information Retrieval, 2015, pp. 271-277.
6. Blei D. M., Ng A. Y., Jordan M. I. Latent dirichlet allocation. The Journal of machine Learning research., vol. 3, 2003, pp. 993-1022.
7. Liu B. Sentiment analysis and opinion mining. Synthesis Lectures on Human Language Technologies, Т. 5. , 2012, pp. 1-167.
8. Martínez-Cámara E, Martín-Valdivia M. T., Urena-López L. A., Montejo-Ráe, A. R. Sentiment analysis in twitter. Natural Language Engineering, Т. 20(1), 2014, PP. 1-28.
9. Moghaddam S., Ester M. On the design of LDA models for aspect-based opinion mining. Proceedings of the 21st ACM international conference on Information and knowledge management. – ACM, 2012., pp. 803-812.
10. Lin C., He Yu., Everson R., Ruger S. Weakly supervised joint sentiment-topic detection from text. Knowledge and Data Engineering, IEEE Transactions on, vol. 24(6), 2012, pp. 1134-1145.
11. Jo Y., Oh A. H. Aspect and sentiment unification model for online review analysis. Proceedings of the fourth ACM international conference on Web search and data mining, ACM, 2011, pp. 815-824.
12. Z. Yang, A. Kotov, A. Mohan S. Lu. Parametric and Non-parametric User-aware Sentiment Topic Models. Proceedings of the 38th ACM SIGIR, 2015.
13. Heinrich G. Parameter estimation for text analysis. Technical report, 2005.
14. Minka T., Lafferty J. Expectation-propagation for the generative aspect model. Proceedings of the Eighteenth conference on Uncertainty in artificial intelligence. – Morgan Kaufmann Publishers Inc., 2002., pp. 352-359.
15. Griffiths T. L., Steyvers M. Finding scientific topics. Proceedings of the National Academy of Sciences, vol. 101 (1), 2004, pp. 5228-5235.
16. Loukachevitch N., Blinov P., Kotelnikov E., Rubtsova Y., Ivanov V., Tutubalina E. SentiRuEval: testing object-oriented sentiment analysis systems in Russian. Proceedings of International Conference Dialog-2015, Moscow, Russia, 2015.
17. Tutubalina E.V. Izvlecheniye problemnykh vyskazyvaniy, svyazannykh s neispravnostyami i narusheniyem funktsional'nosti produktov , na osnovanii otzyvov pol'zovateley [Extracting problem phrases about product defects and malfunctions in user reviews about cars]. Vestnik KGTU im. A.N.Tupoleva [Proceeding of KGTU im. A.N.Tupoleva], vol. 3, 2015. (in Russian)
18. Ivanov V., Tutubalina E., Mingazov N., Alimova I. Extracting aspects, sentiment and categories of aspects in user reviews about restaurants and cars. Proceedings of International Conference "Dialog-2015", Moscow, Russia, 2015.
19. Vorontsov K.V., Potapenko A.A. Regulyarizatsiya, robastnost' i razrezhennost' veroyatnostnykh tematicheskikh modeley [Regularization, robustness and sparsity of probabilistic topic models]. Komp'yuternyye issledovaniya i modelirovaniye [Computer research and modeling], vol. 4 (4), 2012, pp. 693-706. (in Russian).
Review
For citations:
Tutubalina E. Sentiment-based Topic Model for Mining Usability Issues and Failures with User Products. Proceedings of the Institute for System Programming of the RAS (Proceedings of ISP RAS). 2015;27(4):111-128. (In Russ.) https://doi.org/10.15514/ISPRAS-2015-27(4)-6
Email this article 