Algorithm applied to recommendation systems and implementation method thereof

Abstract

The invention discloses an algorithm applied to recommendation systems and an implementation method thereof. The algorithm and the implementation method aim to implement individualized expert recommendation based on user preference, so that the recommendation systems are more open, recommended and multi-index. In order to achieve the purpose, the algorithm realizes three dimension indexes including a quality index, a structural index and a matching index for evaluating experts by the aid of a principal component analysis method and an extremum method of a linear non-dimensionalization method. The implementation method includes: firstly, preliminarily finding out the experts according to the quality index (a correlation factor Y1, an authority factor Y2 and an activity factor Y3) and the structural index (a sociality factor Y4), then computing a reachable index (a reachable factor Y5) of each expert, and finally, obtaining one final recommendation expert according to the experts and reachability thereof.

Description

technical field [0001] The invention belongs to the technical field of Internet computing. Background technique [0002] In recent years, with the development of the Internet and e-commerce, recommender systems have been extensively developed. The informal concept of the recommendation system was proposed by Varian and Resnick in 1997. "It uses e-commerce websites to provide customers with product information and suggestions, to help users decide what products to buy, and to simulate the process of salespeople helping customers to complete the purchase." . In order to tap the needs of users, it has become the focus of major e-commerce systems to recommend products that users are satisfied with according to their interests and hobbies. At present, different forms of recommendation systems have been applied to various fields, such as e-commerce websites represented by Amazon and EBay, movie websites represented by Netflix and moviefinder, and music websites represented by so...

AI Technical Summary

Problems solved by technology

Although these algorithms have solved the problem of querying experts to a certain extent, their scope of application is limited to closed institutions, and all data needs to be updated by users or experts themselves, which has poor maintainability; secondly, only considered The value of experts as individuals, and as carriers of knowledge, experts are not isolated in the process of knowledge sharing and utilization, and their interpersonal relationships and social status can also reflect the differences between experts to a certain extent; Third, these systems are only query systems, listing experts that match the query terms, but there is no sorting function and lack of recommendation

Images