A consensus group decision making method for hotel selection with online reviews by sentiment analysis

This paper proposes a framework for hotel selection based on online reviews by sentiment analysis from the perspective of consensus group decision making. To identify multi-granularity sentiment strength in text reviews, a sentiment analysis method based on the Word2Vec algorithm and one-vs-one strategy based Support Vector Machine (OVO-SVM) algorithm is provided. Then, richer information content can be derived from online text reviews, which are used as the data source of this study. To help members make an aggregation on the preference of hotel attributes, a consensus model with an improved feedback mechanism is proposed, which can reasonably control the adjustment cost in the consensus reaching process. Combining the hotel performance obtained from online reviews and the group preference consensus, the optimal hotel for members can be selected. At the end of this paper, a case study is presented to illustrate the use of the proposed method.