Abstract
Monitoring online customer reviews is important for business organizations to measure customer satisfaction and better manage their reputations. In this paper, we propose a novel dynamic Brand-Topic Model (dBTM) which is able to automatically detect and track brand-associated sentiment scores and polarity-bearing topics from product reviews organized in temporally ordered time intervals. dBTM models the evolution of the latent brand polarity scores and the topic-word distributions over time by Gaussian state space models. It also incorporates a meta learning strategy to control the update of the topic-word distribution in each time interval in order to ensure smooth topic transitions and better brand score predictions. It has been evaluated on a dataset constructed from MakeupAlley reviews and a hotel review dataset. Experimental results show that dBTM outperforms a number of competitive baselines in brand ranking, achieving a good balance of topic coherence and uniqueness, and extracting well-separated polarity-bearing topics across time intervals.1- Anthology ID:
- 2023.tacl-1.24
- Volume:
- Transactions of the Association for Computational Linguistics, Volume 11
- Month:
- Year:
- 2023
- Address:
- Cambridge, MA
- Venue:
- TACL
- SIG:
- Publisher:
- MIT Press
- Note:
- Pages:
- 404–418
- Language:
- URL:
- https://aclanthology.org/2023.tacl-1.24
- DOI:
- 10.1162/tacl_a_00555
- Cite (ACL):
- Runcong Zhao, Lin Gui, Hanqi Yan, and Yulan He. 2023. Tracking Brand-Associated Polarity-Bearing Topics in User Reviews. Transactions of the Association for Computational Linguistics, 11:404–418.
- Cite (Informal):
- Tracking Brand-Associated Polarity-Bearing Topics in User Reviews (Zhao et al., TACL 2023)
- PDF:
- https://preview.aclanthology.org/jeptaln-2024-ingestion/2023.tacl-1.24.pdf