Topic Extraction and Classification Method Based on Comment Sets
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
In recent years, emotional text classification is one of the essential research contents in the field of naturallanguage processing. It has been widely used in the sentiment analysis of commodities like hotels, and othercommentary corpus. This paper proposes an improved WLDA (weighted latent Dirichlet allocation) topicmodel to improve the shortcomings of traditional LDA topic models. In the process of the topic of wordsampling and its word distribution expectation calculation of the Gibbs of the WLDA topic model. An averageweighted value is adopted to avoid topicrelated words from being submerged by highfrequency words, toimprove the distinction of the topic. It further integrates the highest classification of the algorithm of supportvector machine based on the extracted highquality documenttopic distribution and topicword vectors. Finally, an efficient integration method is constructed for the analysis and extraction of emotional words, topicdistribution calculations, and sentiment classification. Through tests on real teaching evaluation data and testset of public comment set, the results show that the method proposed in the paper has distinct advantagescompared with other two typical algorithms in terms of subject differentiation, classification precision, and F1measure.
Abstract
In recent years, emotional text classification is one of the essential research contents in the field of naturallanguage processing. It has been widely used in the sentiment analysis of commodities like hotels, and othercommentary corpus. This paper proposes an improved WLDA (weighted latent Dirichlet allocation) topicmodel to improve the shortcomings of traditional LDA topic models. In the process of the topic of wordsampling and its word distribution expectation calculation of the Gibbs of the WLDA topic model. An averageweighted value is adopted to avoid topicrelated words from being submerged by highfrequency words, toimprove the distinction of the topic. It further integrates the highest classification of the algorithm of supportvector machine based on the extracted highquality documenttopic distribution and topicword vectors. Finally, an efficient integration method is constructed for the analysis and extraction of emotional words, topicdistribution calculations, and sentiment classification. Through tests on real teaching evaluation data and testset of public comment set, the results show that the method proposed in the paper has distinct advantagescompared with other two typical algorithms in terms of subject differentiation, classification precision, and F1measure.
참고문헌 (19)
- L. Zhuang, "Text sentiment classification based on CSLSTM neural network", Computer Systems & Applications, 27(2), pp.230~235 (2018)
- H. Zhou, "Extracting and clustering features of evaluation object in Chinese user reviews", Microcomputer & its Applications, 2014(7), pp.72~75 (2014)
- Z. Hai, "Computational Linguistics and Intelligent Text Processing", pp.393~404 (2011), Springer
- W. Wang, "Implicit feature identification via hybrid association rule mining", Expert Systems with Applications, 40(9), pp.3518~3531 (2013)
- T. C. Chinsha, "A syntactic approach for aspect based opinion mining", Proceedings of the 2015 IEEE 9th International Conference on Semantic Computing(IEEE ICSC 2015), pp.24~31 (2015)
- Q. Zhang, "Research on text emotion classification based on deep belief networks", Journal of Northwestern Polytechnical University(Social Science Edition), 36(1), pp.62~66 (2016)
- D. Tang, "Document modeling with gated recurrent neural network for sentiment classification", Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pp.1422~1432 (2015)
- X. Tang, "Research on sentiment classification of online reviews based on emotional ontology and kNN algorithm", Information Studies : Theory & Application, 39(6), pp.110~114 (2016)
- H. Liu, "Comment target extraction and sentiment classification", Journal of Chinese Information Processing, 2010(1), pp.84~88 (2010)
- P. Yin, "Sentiment classification for Chinese online reviews at product feature level through domain ontology method", Journal of Systems & Management, 25(1), pp.103~114 (2016)
- Y. X. He, "A deep learning model enhanced with emotion semantics for microblog sentiment analysis", Chinese Journal of Computers, 40(4), pp.773~790 (2017)
- Y. LeCun, "Deep learning", Nature, 521(7553), pp.436~444 (2015)
- D. Blei, "Probabilistic topic models", IEEE Signal Processing Magazine, 27(6), pp.55~65 (2010)
- J. Jin, "Understanding big consumer opinion data for market-driven product design", International Journal of Production Research, 54(10), pp.3019~3041 (2016)
- H. Xia, "A comparative study on key technologies of the Chinese sentiment classification preprocessing", Journal of Intelligence, 30(9), pp.160~163 (2011)
- K. Lan, "A survey of data mining and deep learning in bioinformatics", Journal of Medical Systems, 42(8) (2018)
- W. B. A. Karaa, "Mining Multimedia Documents" (2017), CRC Press
- S. Xiao, "Nonlinear dynamic response of reciprocating compressor system with rub-impact fault caused by subsidence", Journal of Vibration and Control, 25(11), pp.1737~1751 (2019)
- S. Li, "Mining features of products from Chinese customer online reviews", Journal of Management Sciences in China, 12(2), pp.142~152 (2009)