This paper reports on the extension of our previous work on feature-based opinion mining for Thai language. In this paper, we present an approach for automatically constructing two main lexicons: features and polar words based on syntactic pattern analysis. The evaluation is performed with a case study on hotel reviews. The experimental results show that our approach is effective in performing its task. To illustrate the potential application, we implement a system called HotelOpinion for summarizing the hotel reviews written in the Thai language. Our system can also generate a comparison between hotels based upon the user's preferred features and then present the results in a user-friendly visualization. Results from our system can be used to determine public perceptions regarding selected hotels in order to allow the business to improve their customer intimacy and satisfaction. © 2011 IEEE.