Informed Recommender Agent: Utilizing Consumer Product Reviews through Text Mining

SILVANA ACIAR; DEBBIE ZHANG; SIMEON SIMOFF; JOHN DEBENHAM

Hong kong, China

Workshop; Workshop on Interaction between Agents and Data Mining (IADM-06); 2006

Consumer reviews, opinions and shared experiences in the use of a product form a powerful source of information about consumer preferences that can be used for making recommendations. A novel framework, which utilizes this valuable information sources first time to create recommendations in recommender agents was recently developed by the authors. In this recommender agent, the most critical issue is how to convert the review comments into ontology instances that can be understood and utilized by computers. This problem was not addressed in our previous work. This paper presents an automatic mapping process using text mining techniques. The ontology contains a controlled vocabulary and their relationships. The attributes of the ontology are learnt from the semantic features in the review comments using supervised learning techniques. The proposed approach is demonstrated using a case study of digital camera reviews