Application of Computer Big Data and Cloud Computing Technology in the Promotion of E-commerce Advertising

Abstract

This article aims to solve the information overload problem faced by large-scale e-commerce systems in the context of big data applications. The scheme of building a distributed e-commerce advertising fixed-point promotion system based on Hadoop is studied. The algorithm based on MapReduce model has high scalability and performance, and can perform offline data analysis efficiently. This paper proposes a decomposition personalized Markov chain (FPMC) model, which combines the MC model with the MF model. The final experimental results verify that the average absolute error (MAE) of the personalized Markov chain (FPMC) model based on content and collaborative filtering is reduced by 15% and 6% compared with traditional content-based or collaborative filtering algorithms. The model can accurately match user information and product information, and match them to complete e-commerce advertising fixed-point push.