With the rapid development of the Internet, the research and application of recommendation systems are developing in multiple directions and fields. Traditional recommendation algorithms no longer meet the requirements for accuracy in some specific fields. Considering that the user's online review information can obtain the user's preference information and preference for the product, this article proposes a tourist attraction recommendation algorithm based on the user's online review. First, crawler software and Jieba word segmentation are used to obtain and preprocess the online review information of users visiting tourist attraction. Second, the sentiment intensity analysis method is used to determine the evaluation scale of each comment's relative attractions and attributes. Third, the user's weight on the attributes of the tourist attraction is calculated based on the processed online review information. Finally, the TOPSIS ranking method is used to recommend users' tourist attractions. Experiments show that the algorithm can effectively determine the user's preference for attractions and effectively improve the accuracy of tourist attraction recommendations.