Integrating Topic Model and Heterogeneous Information Network for Aspect Mining with Rating Bias

  • Yugang Ji
  • Chuan ShiEmail author
  • Fuzhen Zhuang
  • Philip S. Yu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11439)


Recently, there is a surge of research on aspect mining, where the goal is to predict aspect ratings of shops with reviews and overall ratings. Traditional methods assumed that aspect ratings in a specific review text are of the same level, which equal to the corresponding overall rating. However, recent research reveals a different phenomenon: there is an obvious rating bias between aspect ratings and overall ratings. Moreover, these methods usually analyze aspect ratings of reviews with topic models at textual level, while totally ignore potentially structural information among multiple entities (users, shops, reviews), which can be captured by a Heterogeneous Information Network (HIN). In this paper, we present a novel model integrating Topic model and HIN for Aspect Mining with rating bias (called THAM). Firstly, a phrase-level LDA model is designed to extract topic distributions of reviews by using textual information. Secondly, making full use of structural information, we constructs a topic propagation network, and propagate topic distributions in this heterogeneous network. Finally, by setting review as the sharing factor, the two parts are integrated into a uniform optimization framework. Experimental results on two real datasets demonstrate that THAM achieves significant performance improvement, compared to the state of the arts.


Aspect mining Rating bias Topic model Topic propagation network Heterogeneous information network 



This work is supported by the National Key Research and Development Program of China (2017YFB0803304) and the National Natural Science Foundation of China (No. 61772082, U1836206, 61702296, 61806020, 61375058).


  1. 1.
    Bauman, K., Liu, B., Tuzhilin, A.: Aspect based recommendations: recommending items with the most valuable aspects based on user reviews. In: The ACM SIGKDD International Conference, pp. 717–725 (2017)Google Scholar
  2. 2.
    Laddha, A., Mukherjee, A.: Aspect opinion expression and rating prediction via LDA-CRF hybrid. Nat. Lang. Eng. 24, 1–29 (2018)CrossRefGoogle Scholar
  3. 3.
    Li, H., Lin, R., Hong, R., Ge, Y.: Generative models for mining latent aspects and their ratings from short reviews. In: 2015 IEEE International Conference on Data Mining, ICDM 2015, Atlantic City, NJ, USA, 14–17 November 2015, pp. 241–250 (2015)Google Scholar
  4. 4.
    Li, Y., Shi, C., Zhao, H., Zhuang, F., Wu, B.: Aspect mining with rating bias. In: Frasconi, P., Landwehr, N., Manco, G., Vreeken, J. (eds.) ECML PKDD 2016. LNCS (LNAI), vol. 9852, pp. 458–474. Springer, Cham (2016). Scholar
  5. 5.
    Lu, Y., Zhai, C., Sundaresan, N.: Rated aspect summarization of short comments. In: Proceedings of the 18th International Conference on World Wide Web, WWW 2009, Madrid, Spain, 20–24 April 2009, pp. 131–140 (2009)Google Scholar
  6. 6.
    Luo, W., Zhuang, F., Cheng, X., He, Q., Shi, Z.: Ratable aspects over sentiments: predicting ratings for unrated reviews. In: 2014 IEEE International Conference on Data Mining, ICDM 2014, Shenzhen, China, 14–17 December 2014, pp. 380–389 (2014)Google Scholar
  7. 7.
    Luo, W., Zhuang, F., Zhao, W., He, Q., Shi, Z.: QPLSA: utilizing quad-tuples for aspect identification and rating. Inf. Process. Manag. 51(1), 25–41 (2015)CrossRefGoogle Scholar
  8. 8.
    Moghaddam, S., Ester, M.: The FLDA model for aspect-based opinion mining: addressing the cold start problem. In: International Conference on World Wide Web, pp. 909–918 (2013)Google Scholar
  9. 9.
    Pecar, S.: Towards opinion summarization of customer reviews. In: Proceedings of ACL 2018, Student Research Workshop, pp. 1–8 (2018)Google Scholar
  10. 10.
    Schouten, K., van der Weijde, O., Frasincar, F., Dekker, R.: Supervised and unsupervised aspect category detection for sentiment analysis with co-occurrence data. IEEE Trans. Cybern. 48(4), 1263–1275 (2018)CrossRefGoogle Scholar
  11. 11.
    Shi, C., Li, Y., Zhang, J., Sun, Y., Yu, P.S.: A survey of heterogeneous information network analysis. IEEE Trans. Knowl. Data Eng. 29(1), 17–37 (2017)CrossRefGoogle Scholar
  12. 12.
    Sun, Y., Han, J., Zhao, P., Yin, Z., Cheng, H., Wu, T.: RankClus: integrating clustering with ranking for heterogeneous information network analysis. In: ACM SIGKDD 2009, pp. 565–576 (2009)Google Scholar
  13. 13.
    Wang, H., Ester, M.: A sentiment-aligned topic model for product aspect rating prediction. In: Conference on Empirical Methods in Natural Language Processing, pp. 1192–1202 (2014)Google Scholar
  14. 14.
    Xiao, D., Ji, Y., Li, Y., Zhuang, F., Shi, C.: Coupled matrix factorization and topic modeling for aspect mining. Inf. Process. Manag. 54(6), 861–873 (2018)CrossRefGoogle Scholar
  15. 15.
    Yu, D., Mu, Y., Jin, Y.: Rating prediction using review texts with underlying sentiments. Inf. Process. Lett. 117, 10–18 (2017)MathSciNetCrossRefGoogle Scholar
  16. 16.
    Zhang, L., Liu, B.: Aspect and entity extraction for opinion mining. In: Chu, W.W. (ed.) Data Mining and Knowledge Discovery for Big Data. SBD, vol. 1, pp. 1–40. Springer, Heidelberg (2014). Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Yugang Ji
    • 1
  • Chuan Shi
    • 1
    Email author
  • Fuzhen Zhuang
    • 2
    • 3
  • Philip S. Yu
    • 4
  1. 1.Beijing Key Laboratory of Intelligent Telecommunications Software and MultimediaBeijing University of Posts and TelecommunicationsBeijingChina
  2. 2.Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), Institute of Computing TechnologyCASBeijingChina
  3. 3.University of Chinese Academy of SciencesBeijingChina
  4. 4.University of Illinois at ChicagoChicagoUSA

Personalised recommendations