Tourist behaviour data access and intelligent recommendation method based on distributed streaming mode

With the rapid growth of tourism and the advancing maturity of intelligent technology, the recommendation system in tourism has become more and more important. However, the traditional approach is unable to capture the real-time behavioural changes of tourists during the travel process in real time. Therefore, this study combines the advantages of the fuzzy C-means clustering algorithm and the alternate least squares method. The fuzzy C-means is employed to divide the clustering, and the alternate least squares method is utilized to supplement the missing items of the scoring matrix. A combinatorial recommendation algorithm is then designed. The results showed that the area under the curve of the algorithm was 0.783 in the receiver operation characteristic curve, which proved its superiority and reliability. The recall rate of the combinatorial recommendation algorithm was 2.1% and 4.5% higher than that of other algorithms, respectively, which proved that its prediction accuracy was high. The coverage rate of the combinatorial recommendation algorithm was 70.9%, which proved that the system based on this algorithm had high user satisfaction. The algorithm can quickly capture, analyze and process tourist behaviour data, provide intelligent recommendation services for the tourism industry, which is of great significance for improving the quality of tourism services and promoting the development of the tourism industry.