Recommendation Algorithm Research Based on MPA

In the era of big data, recommendation system has been widely used in many applications, and the research on accurate and fast recommendation algorithm has become the focus of society. At present, the main applications are based on content-based recommendation algorithm and collaborative filtering recommendation algorithm. This paper integrates the two recommendation methods above to construct a “host-router” model, adopting MPA (multi path routing algorithm) to select multiple routes with strong connectivity for the host, so as to realize the recommendation process for users. Experimental results on the MovieLens dataset show that compared with the traditional open-source collaborative filtering algorithm, the proposed algorithm increases the recall rate by about 26% and the F1 value by about 30% at the same time. More accurate, wide scenario and comprehensively stronger recommendation can be achieved while innovating the recommendation architecture.