Emotional Analysis Of Hotel Reviews

The rapid development of online travel also indicates the generation of massive commentary information,how can tourism consumers in the face of these massive comments can quickly and effectively access their own needs of information,in the face of more travel options,more online travel products Make the most favorable decision for yourself.The purpose of this paper is to go to the online five-star hotel in Kunming,comment on the use of data mining algorithm analysis to obtain comments on the hotel service attributes of the word,and extract the comments used to include the feature words of the sentence.The naive Bayesian classification method is used to analyze the emotional tendencies of these features.This article first introduces the knowledge of online travel.Based on the introduction of the basic concept of online tourism and the development of online tourism,the development trend of online tourism is put forward,which lays the foundation for further research.Then the paper introduces the theory and development process of text categorization and text emotion analysis,and the key question of text classification is the selection of feature or characteristic item.In this paper,we propose an optimization method based on part of speech and Apriori association rule and TF-IDF standard feature selection.Through the experiment,it is concluded that the selection of the feature selection method is not only the dimensionality of the eigenvector,but also greatly improved the complexity of the computational complexity and the algorithm.In this paper,we use the optimized feature selection method to collect and sort out the five-star hotel reviews in Kunming,and use the naive Bayesian classification method to get the emotional tendencies of comments.This study can quickly obtain the positive and negative evaluation of the commentary about the characteristics of a hotel service from the massive comments,which can effectively assist the readers in making the decision.And extended the emotional analysis,I believe that emotional analysis in the future can be more widely used practice.