Recommendation system evaluation method and device, electronic equipment and storage medium

Abstract

The invention discloses a recommendation system evaluation method and device, electronic equipment and a storage medium, and the recommendation system evaluation method comprises the steps: responding to a diversity evaluation trigger instruction of a recommendation system; the recommendation system is evaluated according to the detection result of the diversity index of the set recommendation system, if it is determined that the recommendation strategy updating condition is met, the current recommendation strategy is updated, and the detection result is obtained by detecting the diversity index in the current detection period. The diversity index comprises at least one of the following indexes: coverage degree, individuation degree, novelty and surprise degree, the coverage degree represents the coverage capability of recommendation information of the recommendation system, and the individuation degree represents the balance degree of individuation recommendation and non-individuation recommendation of the recommendation system; the novelty represents the recommendation degree of the recommendation system to the information which is not recommended by the recommendation object, and the surprise represents the recommendation degree of the recommendation system to the information which is not recommended by the recommendation object and is interested in.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular, to a recommendation system evaluation method, apparatus, electronic device, and storage medium. Background technique [0002] This section is intended to provide a background or context for the embodiments of the application that are recited in the claims. The descriptions herein are not admitted to be prior art by inclusion in this section. [0003] The diversity of the recommender system refers to the fact that the information recommendation to the recommender contains different types of content sets, which are different from the past preferences of the recommender. Since the interests of the recommended objects are extensive, on an information flow platform, in order to satisfy the wide interests of the recommended objects, the recommended content needs to cover different interest fields of the recommended objects, that is, the recommendation results need to be ...

Images