Research on Personalized Recommendation Algorithm of E-Commerce Platform Based on Big Data

This study aims to improve the accuracy and user satisfaction of personalized recommendation algorithms for e-commerce platforms. By analyzing the advantages and disadvantages of the existing recommendation algorithms, combined with big data and machine learning technology, this paper proposes an improvement method. During the research process, the parameter configurations of the two recommendation algorithms, SVD and NMF, are optimized using GridSearchCV. The experimental results show that the optimized SVD algorithm outperforms the NMF and the benchmark algorithm in terms of root mean square error (RMSE) and mean absolute error (MAE) indexes, and exhibits high recommendation accuracy. The study concludes that the optimized recommendation algorithm provides more accurate recommendations on e-commerce platforms, enhancing user satisfaction and platform conversion rates.