Research Article
Network Resource Personalized Recommendation System Based on Collaborative Filtering Algorithm
@INPROCEEDINGS{10.1007/978-3-030-94551-0_50, author={Gang Qiu and Jie Cheng}, title={Network Resource Personalized Recommendation System Based on Collaborative Filtering Algorithm}, proceedings={Advanced Hybrid Information Processing. 5th EAI International Conference, ADHIP 2021, Virtual Event, October 22-24, 2021, Proceedings, Part I}, proceedings_a={ADHIP}, year={2022}, month={1}, keywords={Collaborative filtering algorithm Overall circuit module Bus timing Time weight function Recommendation model}, doi={10.1007/978-3-030-94551-0_50} }- Gang Qiu
Jie Cheng
Year: 2022
Network Resource Personalized Recommendation System Based on Collaborative Filtering Algorithm
ADHIP
Springer
DOI: 10.1007/978-3-030-94551-0_50
Abstract
Several existing personalized recommendation systems for network resources do not specifically classify user history information personalized classification criteria, resulting in a low degree of matching between system personalized recommendations and user interests. In order to improve the accuracy of personalized recommendations for network resources, we designed a collaborative filtering based Algorithm-based personalized recommendation system for network resources. In the hardware design, design the overall circuit module, configure the bus timing, and improve the operating speed of the system hardware. In software design, calculate the time weight function to construct the user's implicit scoring matrix; calculate the similarity between network resources and user browsing history items, design personalized classification criteria for user history information based on collaborative filtering algorithms; calculate predicted item scores and actual scores Establish a personalized recommendation model for network resources.In the experiment, the system is compared with several existing systems, and the average absolute error in different adjacent sets is tested. According to the data results, in the five data test sets, the average absolute error of the system is less than that of other systems, so the personalized recommendation system based on collaborative filtering algorithm has better recommendation accuracy.
