Most of the existing recommendation algorithms are based on the recommendation of similar products, which can easily lead to “information cocoon rooms”. In order to solve the limitations of similar product recommendation, the recommendation algorithm is extended to different categories of product recommendation. This paper proposes a heterogeneous product recommendation algorithm based on item similarity. Based on the application of item similarity, a cross-correlation recommendation theory is proposed to solve the problem of heterogeneous recommendation of target products and recommended product sets. Finally, this article extracts the product data from the Tianchi Taobao clothing matching data set, and applies the proposed algorithm to the programming language to analyze the data set. According to the obtained experimental results, the algorithm has a high recommendation success rate and a good recommendation effect.
展开▼
