{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/unifying-topic-sentiment-preference-in-an-hdp","title":"Unifying Topic, Sentiment & Preference in an HDP-Based Rating Regression Model for Online Reviews","arxiv_id":"1812.07805","date":"2018-12-19","proceeding":null,"authors":["Zheng Chen","Yong Zhang","Yue Shang","Xiaohua Hu"],"abstract":"This paper proposes a new HDP based online review rating regression model\nnamed Topic-Sentiment-Preference Regression Analysis (TSPRA). TSPRA combines\ntopics (i.e. product aspects), word sentiment and user preference as regression\nfactors, and is able to perform topic clustering, review rating prediction,\nsentiment analysis and what we invent as \"critical aspect\" analysis altogether\nin one framework. TSPRA extends sentiment approaches by integrating the key\nconcept \"user preference\" in collaborative filtering (CF) models into\nconsideration, while it is distinct from current CF models by decoupling \"user\npreference\" and \"sentiment\" as independent factors. Our experiments conducted\non 22 Amazon datasets show overwhelming better performance in rating\npredication against a state-of-art model FLAME (2015) in terms of error,\nPearson's Correlation and number of inverted pairs. For sentiment analysis, we\ncompare the derived word sentiments against a public sentiment resource\nSenticNet3 and our sentiment estimations clearly make more sense in the context\nof online reviews. Last, as a result of the de-correlation of \"user preference\"\nfrom \"sentiment\", TSPRA is able to evaluate a new concept \"critical aspects\",\ndefined as the product aspects seriously concerned by users but negatively\ncommented in reviews. Improvement to such \"critical aspects\" could be most\neffective to enhance user experience.","url_abs":"http://arxiv.org/abs/1812.07805v1","url_pdf":"http://arxiv.org/pdf/1812.07805v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"unifying-topic-sentiment-preference-in-an-hdp","repo_url":"https://github.com/tonyrivermsfly/TSPRA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"collaborative-filtering","task_name":"Collaborative Filtering"},{"task_slug":"online-review-rating","task_name":"Online Review Rating"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}