Traditional keyword search method lacks the semantic processing of search content, and the search results may appear incomplete and inaccurate, unable to meet user's requirements. In order to improve the quality of retrieval, on the basis of the construction of tourism domain ontology, the retrieval module was used to analyze the search conditions selected in the process of retrieval preprocessing at semantic level, and the semantic similarity calculation method was used to expand the semantic analysis of search keywords, combining Jena reasoning mechanism and tourism domain. The reasoning rules suited to this field were formulated according to the various attribute relations of the actual information contained in it. The qualified reasoning rules were screened out through the query conditions, and the semantic reasoning query was carried out with the established database of data resources. A semantic retrieval model based on tourism domain ontology was proposed. From the experiments, it can be seen that the model can make up for the shortcomings of traditional search methods and make the results more accurate.