Recurrent Entity Networks with Delayed Memory Update for Targeted Aspect-based Sentiment Analysis
While neural networks have been shown to achieve impressive results for sentence-level sentiment analysis, targeted aspect-based sentiment analysis (TABSA) --- extraction of fine-grained opinion polarity w.r.t. a pre-defined set of aspects --- remains a difficult task. Motivated by recent advances in memory-augmented models for machine reading, we propose a novel architecture, utilising external "memory chains" with a delayed memory update mechanism to track entities. On a TABSA task, the proposed model demonstrates substantial improvements over state-of-the-art approaches, including those using external knowledge bases.
PDF Abstract NAACL 2018 PDF NAACL 2018 Abstract
Datasets
CBT
Results from the Paper
Methods
No methods listed for this paper. Add
relevant methods here
