Research on the Design of E-commerce Recommendation System
AUTHORS
Daniel Yang,Deakin University, Australia
Saman Grice*,Deakin University, Australia
ABSTRACT
In recent years, the amount of data in various fields, especially in the e-commerce industry, has continued to rise, which is inseparable from the rapid development of information technology and the application of massive data processing technology. To use massive amounts of data to serve users, and to promote user retention, the use of recommendation systems is very important. Based on a hybrid recommendation idea, this paper designs an e-commerce recommendation system based on big data technology. First, the system needs analysis, and the overall architecture of the system is designed. Then the design process of each recommended module is explained in detail. This article combines demographic-based recommendation algorithm, content-based recommendation algorithm, ALS-based collaborative filtering recommendation algorithm, and model-based real-time recommendation algorithm to form a hybrid recommendation module to provide users with recommendation services. Among them, the demographic-based recommendation can solve the user's cold start problem, the collaborative filtering recommendation does not require the content attributes of the item to be recommended, and the model-based recommendation algorithm can provide users with real-time recommendation services.
KEYWORDS
E-commerce, Recommendation system, Recommendation algorithm, Collaborative filtering