CollRec: Pre-Trained Language Models and Knowledge Graphs Collaborate to Enhance Conversational Recommendation System

被引:0
|
作者
Liu, Shuang [1 ]
Ao, Zhizhuo [1 ]
Chen, Peng [2 ]
Kolmanic, Simon [3 ]
机构
[1] Dalian Minzu Univ, Sch Comp Sci & Engn, Dalian 116600, Peoples R China
[2] Dalian Neusoft Univ Informat, Sch Comp & Software, Dalian 116023, Peoples R China
[3] Univ Maribor, Fac Elect Engn & Comp Sci, Maribor 2000, Slovenia
来源
IEEE ACCESS | 2024年 / 12卷
关键词
Knowledge graphs; Oral communication; Task analysis; Recommender systems; Motion pictures; Costs; Accuracy; Large language models; Conversational recommendation system; knowledge graph; large language model; end-to-end generation; fine-tuning; ReDial; WebNLG; 2020; challenge;
D O I
10.1109/ACCESS.2024.3434720
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Existing conversational recommender systems (CRS) use insufficient generality in incorporating external information using knowledge graphs. The recommendation module and generation module are loosely connected during model training and shallowly integrated during inference. A simple switching or copying mechanism is used to merge recommended items into generated responses. These problems significantly degrade the recommendation performance. To alleviate this problem, we propose a novel unified framework for collaboratively enhancing conversational recommendations using pre-trained language models and knowledge graphs (CollRec). We use a fine-tuned pre-trained language model to efficiently extract knowledge graphs from conversational text descriptions, perform entity-based recommendations based on the generated graph nodes and edges, and fine-tune a large-scale pre-trained language model to generate fluent and diverse responses. Experimental results on the WebNLG 2020 Challenge dataset, ReDial dataset, and Reddit-Movie dataset show that our CollRec model significantly outperforms the state-of-the-art methods.
引用
收藏
页码:104663 / 104675
页数:13
相关论文
共 50 条
  • [1] ReLMKG: reasoning with pre-trained language models and knowledge graphs for complex question answering
    Xing Cao
    Yun Liu
    Applied Intelligence, 2023, 53 : 12032 - 12046
  • [2] ReLMKG: reasoning with pre-trained language models and knowledge graphs for complex question answering
    Cao, Xing
    Liu, Yun
    APPLIED INTELLIGENCE, 2023, 53 (10) : 12032 - 12046
  • [3] ProSide: Knowledge Projector and Sideway for Pre-trained Language Models
    He, Chaofan
    Lu, Gewei
    Shen, Liping
    NATURAL LANGUAGE PROCESSING AND CHINESE COMPUTING, PT II, NLPCC 2024, 2025, 15360 : 56 - 68
  • [4] Assisted Process Knowledge Graph Building Using Pre-trained Language Models
    Bellan, Patrizio
    Dragoni, Mauro
    Ghidini, Chiara
    AIXIA 2022 - ADVANCES IN ARTIFICIAL INTELLIGENCE, 2023, 13796 : 60 - 74
  • [5] Knowledge Graphs and Pretrained Language Models Enhanced Representation Learning for Conversational Recommender Systems
    Qiu, Zhangchi
    Tao, Ye
    Pan, Shirui
    Liew, Alan Wee-Chung
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2024, : 1 - 15
  • [6] Pre-trained language models: What do they know?
    Guimaraes, Nuno
    Campos, Ricardo
    Jorge, Alipio
    WILEY INTERDISCIPLINARY REVIEWS-DATA MINING AND KNOWLEDGE DISCOVERY, 2024, 14 (01)
  • [7] KG-prompt: Interpretable knowledge graph prompt for pre-trained language models
    Chen, Liyi
    Liu, Jie
    Duan, Yutai
    Wang, Runze
    KNOWLEDGE-BASED SYSTEMS, 2025, 311
  • [8] CokeBERT: Contextual knowledge selection and embedding towards enhanced pre-trained language models
    Su, Yusheng
    Han, Xu
    Zhang, Zhengyan
    Lin, Yankai
    Li, Peng
    Liu, Zhiyuan
    Zhou, Jie
    Sun, Maosong
    AI OPEN, 2021, 2 : 127 - 134
  • [9] Pre-trained language models for keyphrase prediction: A review
    Umair, Muhammad
    Sultana, Tangina
    Lee, Young-Koo
    ICT EXPRESS, 2024, 10 (04): : 871 - 890
  • [10] Billion-scale pre-trained knowledge graph model for conversational chatbot
    Wong, Chi-Man
    Feng, Fan
    Zhang, Wen
    Chen, Huajun
    Vong, Chi-Man
    Chen, Chuangquan
    NEUROCOMPUTING, 2024, 606