Simplifying Content-Based Neural News Recommendation: On User Modeling and Training Objectives

被引:3
|
作者
Iana, Andreea [1 ]
Glavas, Goran [2 ]
Paulheim, Heiko [1 ]
机构
[1] Univ Mannheim, Mannheim, Germany
[2] Univ Wurzburg, CAIDAS, Wurzburg, Germany
关键词
neural news recommendation; user modeling; late fusion; training objectives; contrastive learning; evaluation;
D O I
10.1145/3539618.3592062
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The advent of personalized news recommendation has given rise to increasingly complex recommender architectures. Most neural news recommenders rely on user click behavior and typically introduce dedicated user encoders that aggregate the content of clicked news into user embeddings (early fusion). These models are predominantly trained with standard point-wise classification objectives. The existing body of work exhibits two main shortcomings: (1) despite general design homogeneity, direct comparisons between models are hindered by varying evaluation datasets and protocols; (2) it leaves alternative model designs and training objectives vastly unexplored. In this work, we present a unified framework for news recommendation, allowing for a systematic and fair comparison of news recommenders across several crucial design dimensions: (i) candidate-awareness in user modeling, (ii) click behavior fusion, and (iii) training objectives. Our findings challenge the status quo in neural news recommendation. We show that replacing sizable user encoders with parameter-efficient dot products between candidate and clicked news embeddings (late fusion) often yields substantial performance gains. Moreover, our results render contrastive training a viable alternative to point-wise classification objectives.
引用
收藏
页码:2384 / 2388
页数:5
相关论文
共 50 条
  • [1] Content-Based News Recommendation
    Kompan, Michal
    Bielikova, Maria
    [J]. E-COMMERCE AND WEB TECHNOLOGIES, 2010, 61 : 61 - 72
  • [2] VideoTopic: Modeling User Interests for Content-Based Video Recommendation
    Zhu, Qiusha
    Shyu, Mei-Ling
    Wang, Haohong
    [J]. INTERNATIONAL JOURNAL OF MULTIMEDIA DATA ENGINEERING & MANAGEMENT, 2014, 5 (04): : 1 - 21
  • [3] Content-Based Collaborative Filtering for News Topic Recommendation
    Lu, Zhongqi
    Dou, Zhicheng
    Lian, Jianxun
    Xie, Xing
    Yang, Qiang
    [J]. PROCEEDINGS OF THE TWENTY-NINTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2015, : 217 - 223
  • [4] Deep User Modeling for Content-based Event Recommendation in Event-based Social Networks
    Wang, Zhibo
    Zhang, Yongquan
    Chen, Honglong
    Li, Zhetao
    Xia, Feng
    [J]. IEEE CONFERENCE ON COMPUTER COMMUNICATIONS (IEEE INFOCOM 2018), 2018, : 1304 - 1312
  • [5] News Recommendation Based on Content Fusion of User Behavior
    Li, Lin
    Wang, Li
    [J]. 2020 13TH INTERNATIONAL SYMPOSIUM ON COMPUTATIONAL INTELLIGENCE AND DESIGN (ISCID 2020), 2020, : 217 - 220
  • [6] Terms of a Feather: Content-Based News Recommendation and Discovery Using Twitter
    Phelan, Owen
    McCarthy, Kevin
    Bennett, Mike
    Smyth, Barry
    [J]. ADVANCES IN INFORMATION RETRIEVAL, 2011, 6611 : 448 - 459
  • [7] Graph Neural News Recommendation with User Existing and Potential Interest Modeling
    Qiu, Zhaopeng
    Hu, Yunfan
    Wu, Xian
    [J]. ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, 2022, 16 (05)
  • [8] A novel content-based recommendation approach based on LDA topic modeling for literature recommendation
    Bagul, Dhiraj Vaibhav
    Barve, Sunita
    [J]. PROCEEDINGS OF THE 6TH INTERNATIONAL CONFERENCE ON INVENTIVE COMPUTATION TECHNOLOGIES (ICICT 2021), 2021, : 954 - 961
  • [9] Content-Based Bipartite User-Image Correlation for Image Recommendation
    Meng Jian
    Ting Jia
    Lifang Wu
    Lei Zhang
    Dong Wang
    [J]. Neural Processing Letters, 2020, 52 : 1445 - 1459
  • [10] Content-Based Bipartite User-Image Correlation for Image Recommendation
    Jian, Meng
    Jia, Ting
    Wu, Lifang
    Zhang, Lei
    Wang, Dong
    [J]. NEURAL PROCESSING LETTERS, 2020, 52 (02) : 1445 - 1459