Design and Implementation of a Personalized Recommendation System Based on Deep Learning Distributed Collaborative Filtering Algorithm on Social Media Platforms

Social media platforms mainly provide personalized recommendation services for users, and usually adopt recommendation methods related to user or project similarity based on the characteristics of different platforms. However, current recommendation techniques mostly rely on simple similarity calculations, which can significantly reduce accuracy when dealing with large amounts of data. To address this issue, this paper proposes a collaborative filtering recommendation method based on social media gene maps. Specifically, this article draws on the concept of genes in biology, collects social media user behavior data, extracts biological genes, social genes, and network gene fragments, calculates the core gene matching degree of each gene fragment, and uses entropy weight method to assign different weights to gene fragments, thereby constructing a complete social media gene map and optimizing the recommendation mechanism. The experimental results show that this method outperforms traditional recommendation algorithms in terms of recommendation performance. Subsequently, this article proposes to combine social media gene maps with project similarity to further improve recommendation algorithms. When measuring project similarity, a fusion of scoring similarity and semantic similarity was used, combined with user gene matching degree, to optimize the weighted Slope One algorithm and achieve more accurate recommendation services. The results indicate that the algorithm proposed in this paper has significantly improved in evaluation metrics such as MAE and MRSE.