Research on Recommendation Strategies Integrating Emotional Tendency and User Influences

At present, the topic model based on LDA as an information recommendation method has the defect of neglecting the emotional information in social platform. Therefore, a microblog user recommendation strategy I-TES (influence topic emotionality similarity model) is proposed which combines emotional orientation with user influence model in this paper. Firstly, the user's theme distribution and emotional tendency is obtained based on JST model, the JS distance is used to calculate the similarity of user's theme probability distribution, and the two is combined to get the user's interest similarity score. Secondly, the user's influence measurement indicators are summarized into microblog influence, user's activity and fan's influence, and the weights of the three indicators are calculated according to AHP(analytic hierarchy process). Finally, the combination of the two is used to obtain the final score as the recommendation result. The strategy is tested on Sina Weibo user data set, and the experiment shows that this strategy has higher recommendation accuracy than the traditional recommendation method.

[1]  Liu Yu,et al.  A method of optimizing LDA result purity based on semantic similarity , 2017, 2017 32nd Youth Academic Annual Conference of Chinese Association of Automation (YAC).

[2]  Bofeng Zhang,et al.  Friend Recommendation in Online Social Networks Combining Interest Similarity and Social Interaction , 2018, 2018 International Conference on Audio, Language and Image Processing (ICALIP).

[3]  Benjin Zhu,et al.  Improving user recommendation by extracting social topics and interest topics of users in uni-directional social networks , 2018, Knowl. Based Syst..

[4]  Kathleen McKeown,et al.  A Hierarchical Model of Web Summaries , 2011, ACL.

[5]  Ling Huang,et al.  Topic Detection from Microblogs Using T-LDA and Perplexity , 2017, 2017 24th Asia-Pacific Software Engineering Conference Workshops (APSECW).

[6]  Wei Zheng,et al.  Building a TIN-LDA Model for Mining Microblog Users’ Interest , 2019, IEEE Access.

[7]  Ramesh Nallapati,et al.  Labeled LDA: A supervised topic model for credit attribution in multi-labeled corpora , 2009, EMNLP.

[8]  Xiaomin Sun,et al.  Topic Discovery for Streaming Short Texts with CTM , 2018, 2018 International Joint Conference on Neural Networks (IJCNN).

[9]  Changjun Jiang,et al.  Mining Coherent Topics With Pre-Learned Interest Knowledge in Twitter , 2017, IEEE Access.

[10]  Hua Xu,et al.  Implicit feature identification in Chinese reviews using explicit topic mining model , 2015, Knowl. Based Syst..

[11]  Hai Jin,et al.  Future Generation Computer Systems , 2022 .

[12]  Frank D. Wood,et al.  Hierarchically Supervised Latent Dirichlet Allocation , 2011, NIPS.

[13]  Michael I. Jordan,et al.  Hierarchical Dirichlet Processes , 2006 .

[14]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[15]  David M. Blei,et al.  Supervised Topic Models , 2007, NIPS.

[16]  Cong Wang,et al.  Microblog Topic Mining Based on FR-DATM , 2018 .

[17]  Thomas L. Griffiths,et al.  Hierarchical Topic Models and the Nested Chinese Restaurant Process , 2003, NIPS.

[18]  Feng Zhou,et al.  Dynamic topic mining for microblog fused with user's behavior and time window , 2017, 2017 International Conference on Asian Language Processing (IALP).

[19]  Hae-Chang Rim,et al.  Identifying interesting Twitter contents using topical analysis , 2014, Expert Syst. Appl..

[20]  Barbara Plank,et al.  Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies , 2011 .

[21]  Haitao Xiong,et al.  A hybrid model of VSM and LDA for text clusteing , 2017, 2017 2nd IEEE International Conference on Computational Intelligence and Applications (ICCIA).

[22]  Edward Y. Chang,et al.  PLDA: Parallel Latent Dirichlet Allocation for Large-Scale Applications , 2009, AAIM.