A hybrid recommendation technique using topic embedding for rating prediction and to handle cold-start problem
作者:
Highlights:
• A novel recommendation approach to handle the cold-start problem.
• Incorporating rating data and topic embedding into UBCF for rating prediction.
• Assessing the impact of topic embedding on effective recommendation.
• Comparative evaluation of the proposed approach with 9 baselines and 5 SOTA.
• Assessing efficacy of the proposed approach to handle the cold-start problem.
摘要
•A novel recommendation approach to handle the cold-start problem.•Incorporating rating data and topic embedding into UBCF for rating prediction.•Assessing the impact of topic embedding on effective recommendation.•Comparative evaluation of the proposed approach with 9 baselines and 5 SOTA.•Assessing efficacy of the proposed approach to handle the cold-start problem.
论文关键词:Collaborative filtering,Topic modeling,Word embedding,Topic embedding,Cold-start
论文评审过程:Received 29 August 2021, Revised 24 July 2022, Accepted 26 July 2022, Available online 30 July 2022, Version of Record 9 August 2022.
论文官网地址:https://doi.org/10.1016/j.eswa.2022.118307