An unsupervised topic-sentiment joint probabilistic model for detecting deceptive reviews.
Lu-yu DongShu-Juan JiChun-jin ZhangQi ZhangDickson K. W. ChiuLi-qing QiuDa LiPublished in: Expert Syst. Appl. (2018)
Keyphrases
- probabilistic model
- product reviews
- opinion mining
- topic modeling
- topic models
- sentiment analysis
- opinion summarization
- customer reviews
- sentiment classification
- online product reviews
- online reviews
- product features
- topic specific
- language model
- generative model
- topic discovery
- unsupervised learning
- bayesian networks
- latent dirichlet allocation
- semi supervised
- sentiment polarity
- text classification
- data driven
- user generated
- public opinion
- probabilistic topic models
- language modeling
- automatic detection
- supervised learning
- sentiment lexicon
- mental states
- definitional question answering
- information bottleneck
- topic detection
- positive or negative
- information retrieval
- test collection
- expectation maximization
- search engine