Learning-based aspect identification in customer review products

Aspect extraction is an important step in opinion mining to identify aspect in customer review products. Most existing works defines the pattern set manually or using heuristic approach. In this paper, we propose learning-based approach using decision tree and rule learning to generate pattern set based on sequence labelling. The patterns will be used to identify and extract aspect in customer product review combined with opinion lexicon. We use ID3, J48, RandomTree, Part and Prism to generate pattern that identifies aspect, based on sequence labelling. Our experiment results based on some generated pattern using Decision Tree and Rule Learning, show that the generated pattern can produced better performance than baseline model. However, there is significant increase in the number of patterns generated from learning-based aspect extraction compared with previous pattern.