Incorporating Feature-based and Similarity-based Opinion Mining - CTL in NTCIR-8 MOAT.

Ruifeng Xu,Chunyu Kit
2010-01-01
Abstract:paper presents the design and implementation of an opinion mining system developed by NLPCity group for NTCIR-8 MOAT evaluation, named CTL-OM. CTL-OM incorporates two opinion mining approach, namely feature-based approach and similarity- based approach. The feature-based approach incorporates computational features at punctuation-, word-, collocation-, phrase-, sentence-, paragraph- and document-level in a coarse- fine multi-pass classification framework. The opinion holders and opinions targets in the opinionated sentences are then recognized. The similarity-based approach works in a different way. This approach estimates the similarity between the example sentences and testing sentence and identifies the similar example sentence- testing sentence pair. The opinion components annotated in the example sentence are utilized to recognize the corresponding components in the testing sentence. The analysis outputs by these two approaches are integrated to obtain the final opinion mining results. CTL-OM achieved promising results in Traditional Chinese and Simplified Chinese evaluation in MOAT-8, respectively. This result shows that the incorporation of feature- based and similarity-based opinion mining approach is effective.
What problem does this paper attempt to address?