Towards Designing an Intelligent Recommender System using Adaptive Collaborative Filtering Technique
Qi Zhang, Mideth Abisado · 2023
In recent years, intelligent recommendation systems using adaptive collaborative filtering have received great attention due to their ability to provide personalized and relevant recommendations. The innovation of this study lies in proposing an adaptive collaborative intelligent recommendation model (ACIR), which combines adaptive collaborative filtering with context, user behavior, and movie feature information. It utilizes automatic encoders and deep learning algorithms to capture the characteristic relationships of the context, explore the internal connections between the context and users, and improve the accuracy and interpretability of the recommendation system to meet the personalized needs of users. The aim of this study is to predict user preferences by considering the characteristics of users, projects, and contexts, and to evaluate the potential changes in user preferences over time. In the experimental stage, LDOS CoMoDa was used to test and train the dataset, and the model was validated. Choose traditional collaborative recommendation, k-means algorithm, CAMF, and CAMF_ CM algorithm for comparison. After experimental verification, this article proposes that compared with other traditional recommendation algorithms, ACIR has higher accuracy and stronger application value