Author(s)

Yejun Wu, Hongli Yang

Corresponding Author

Yejun Wu

Abstract

With the continuous development of information technology, the emergence of massive data makes recommendation algorithm become the core technology to solve the problem of user information acquisition. However, the traditional recommendation algorithm has some limitations. This article focuses on the fusion of probability statistics and recommendation algorithm. By expounding the basic theory of probability statistics and recommendation algorithm, this article deeply analyzes the application of probability statistics in data preprocessing, model construction and result assessment of recommendation algorithm, and discusses the advantages and challenges brought by the combination of them. Combining probability statistics with recommendation algorithm can improve the accuracy of recommendation algorithm, enhance its adaptability to complex data, and expand the diversity of recommendation. However, in this process, it also faces many challenges, such as the increase of computational complexity, the inconsistency between probability hypothesis and actual situation, and privacy protection. This study provides a new idea for the optimization of recommendation algorithm, promotes the application of probability and statistics theory in this field, and has important theoretical reference value and guiding significance for related research and practice.

Keywords

Probability Statistics; Recommendation Algorithm; Data Processing; Assessment Index