Interest-based Modeling Based on Group Wisdom and Its Application in Recommendation Algorithms

The existing related work fails to capture the shortcomings of the user's real interest when using resource features for user interest modeling. This paper proposes a user intelligence modeling method based on group intelligence. Taking into account several user behaviors including browsing and evaluation, we use group intelligence and representation learning technology to model users' interes, in order to improving the performance of personalized recommendation. In the experiment, this paper takes the film recommendation task as an example to verify the effectiveness of the proposed method on real data. The experimental results show that compared with the traditional project-based user interest modeling method, the group intelligence-based user interest modeling method can more accurately describe the user's interest and obtain better recommendation effect.