Paper: Fine Granular Aspect Analysis using Latent Structural Models

ACL ID P12-2065
Title Fine Granular Aspect Analysis using Latent Structural Models
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2012

In this paper, we present a structural learning model for joint sentiment classification and as- pect analysis of text at various levels of gran- ularity. Our model aims to identify highly in- formative sentences that are aspect-specific in online custom reviews. The primary advan- tages of our model are two-fold: first, it per- forms document-level and sentence-level sen- timent polarity classification jointly; second, it is able to find informative sentences that are closely related to some respects in a review, which may be helpful for aspect-level senti- ment analysis such as aspect-oriented sum- marization. The proposed method was eval- uated with 9,000 Chinese restaurant reviews. Preliminary experiments demonstrate that our model obtains promising performance.