Tourism Review Text Mining in Guizhou Province Based on LDA Topic Model

Abstract

The rapid development of the Internet has brought more convenience to people's lives. Consumers begin to choose to obtain services through the Internet, and tourism has gradually become an important member of e-commerce. Most tourists choose to learn about scenic spots online in advance and buy tickets through tourism websites. In the face of a large number of consumers, scenic spot managers should pay attention to online feedback information so as to improve all aspects of the scenic spot. In this paper, the characteristic tourist attractions of Guizhou are taken as an example, the octopus collector is used to climb the comment information of major tourism websites, and the comment information is processed by python and R, the word cloud map of positive and negative emotion words is constructed, and the feature words with high attention are obtained. The topic extraction is carried out by the LDA model, and the comment information is further analyzed by combining the positive and negative emotion words. The paper focuses on exploring the concerns of tourists, and puts forward suggestions with commercial value for Guizhou characteristic tourist attractions.