A Topic Recommendation Control Method Based on Topic Relevancy and R-tree Index
-
Published:2024-09-02
Issue:5
Volume:19
Page:
-
ISSN:1841-9844
-
Container-title:INTERNATIONAL JOURNAL OF COMPUTERS COMMUNICATIONS & CONTROL
-
language:
-
Short-container-title:INT J COMPUT COMMUN, Int. J. Comput. Commun. Control
Author:
Yu Jing,Lu Zhixing,Li Xianghua,Wu Bin,Zhang Shunli,Cui Zongmin
Abstract
Topic recommendation control aims to suggest relevant topics to users based on their preferences and regional trends. However, existing methods often lack effective measures to evaluate topic-user relevancy and require comparing large amounts of regional information, leading to low accuracy and efficiency. Therefore, we propose a Topic Recommendation Control method based on topic Relevancy and R-tree index (named as TRCRR) to address these limitations. TRCRR introduces a novel personalized topic relevancy metric that quantifies the relevancy between topics and user preferences. To improve efficiency, an R-tree topic index is constructed to organize topics across different regions hierarchically. Experiments on a real-world dataset show that TRCRR achieves better recommendation accuracy and efficiency compared to several baseline methods. The proposed approach offers a promising solution for personalized and region-aware topic recommendation.
Publisher
Agora University of Oradea