Paper: Bi-directional Inter-dependencies of Subjective Expressions and Targets and their Value for a Joint Model

ACL ID P13-2147
Title Bi-directional Inter-dependencies of Subjective Expressions and Targets and their Value for a Joint Model
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

Opinion mining is often regarded as a clas- sification or segmentation task, involving the prediction of i) subjective expressions, ii) their target and iii) their polarity. In- tuitively, these three variables are bidirec- tionally interdependent, but most work has either attempted to predict them in isolation or proposing pipeline-based approaches that cannot model the bidirectional interac- tion between these variables. Towards bet- ter understanding the interaction between these variables, we propose a model that allows for analyzing the relation of target and subjective phrases in both directions, thus providing an upper bound for the im- pact of a joint model in comparison to a pipeline model. We report results on two public datasets (cameras and cars), show- ing that our model outpe...