With the rapid development of information technology, the informationization level of sport tourism has been improved in an all-around way, which makes a large number of data accumulated in the management system of road traffic. However, the traditional association rule algorithm cannot deal with huge data. To make up for the deficiency of predicting suitability from a single perspective, this paper, from the perspective of road daily travel planning, constructs a multi-dimensional characteristic model. Based on eigenvector processing as a research content, by carrying out a multi-dimensional prediction on the whole data set and considering the limitation problem of the influence of time, space, and items on daily travel routes, the invention provides a daily travel suitability road daily travel planning route prediction research method based on multiple dimensions. The invention also extracts the outstanding performance of a network model in a single variable problem based on deep interest and double characteristics. The combination of SVR and GBRT algorithm makes up for the one-sidedness of single perspective prediction. It uses the weighted fusion principle to fuse the results and establishes a multi-dimensional route suitability prediction model under this mode. Experiments verify that the prediction of multivariate dimensional data achieves the desired results. With the help of the dependence between the level and element dimension data, the future route planning trend can be judged. This algorithm compared to the ant colony algorithm and the traditional genetic algorithm increased by 15.6% and 15.1%. The system response time has been increased by more than 60%, which can effectively improve the accuracy of prediction. Therefore, in the VFC environment, the actual user needs can be improved, the planning and management of traffic routes can be guided, and the development of sport tourism systems of Taihang Mountain can be further promoted.