skip to main content
10.1145/3640457.3687094acmconferencesArticle/Chapter ViewAbstractPublication PagesrecsysConference Proceedingsconference-collections
tutorial

A Tutorial on Feature Interpretation in Recommender Systems

Published: 08 October 2024 Publication History

Abstract

Data-driven techniques have greatly empowered recommender systems in different scenarios. However, many mainstream algorithms rely on black-box models, making them difficult to interpret, debug, and evolve. Therefore, effectively and efficiently interpreting the behaviors and impacts of features in different stages of recommendation pipelines is essential in industrial recommender systems to master a clear picture of the features they use and bring new insights to system improvement and product design. In this tutorial, we present a systematic overview of feature interpretation technologies in the recommendation field from various aspects including algorithms, applications, and challenges. We first provide a systematic taxonomy of previous feature interpretation methods based on their interpretation perspectives, then introduce the experience and lessons of feature interpretation in large-scale and real-time industrial recommender systems. Finally, we summarize several remaining theoretical and practical challenges in feature interpretation and present corresponding future directions to help feature interpretation better empower recommender systems. From this tutorial, the RecSys community can obtain insights into the methodology and real-world applications of feature interpretation to make more transparent, targeted, and intelligent system optimization.

References

[1]
André Altmann, Laura Toloşi, Oliver Sander, and Thomas Lengauer. 2010. Permutation importance: a corrected feature importance measure. Bioinformatics 26, 10 (2010), 1340–1347.
[2]
Avishek Anand, Lijun Lyu, Maximilian Idahl, 2022. Explainable information retrieval: A survey. arXiv preprint arXiv:2211.02405 (2022).
[3]
Xu Chen, Jingsen Zhang, Lei Wang, 2023. REASONER: An Explainable Recommendation Dataset with Multi-aspect Real User Labeled Ground Truths Towards more Measurable Explainable Recommendation. NeurIPS (2023).
[4]
Anupam Datta, Matt Fredrikson, Klas Leino, 2021. Machine learning explainability and robustness: connected at the hip. In KDD. 4035–4036.
[5]
Zhaocheng Du, Junhao Chen, Qinglin Jia, 2023. LightCS: Selecting Quadratic Feature Crosses in Linear Complexity. In WWW.
[6]
Zhaocheng Du, Xianquan Wang, Jieming Zhu, 2024. FairFS: Unbiased Feature Importance Regularization for Recommender System. In arXiv preprint.
[7]
Mark A Hall. 1999. Correlation-based feature selection for machine learning. Ph. D. Dissertation. The University of Waikato.
[8]
Noah Hollmann, Samuel Müller, and Frank Hutter. 2024. Large language models for automated data science: Introducing caafe for context-aware automated feature engineering. NeurIPS 36 (2024).
[9]
Pengyue Jia, Yejing Wang, Zhaocheng Du, 2024. ERASE: Benchmarking Feature Selection Methods for Deep Recommender Systems. In KDD.
[10]
Yuanfei Luo, Mengshuo Wang, Hao Zhou, Quanming Yao, Wei-Wei Tu, Yuqiang Chen, Wenyuan Dai, and Qiang Yang. 2019. Autocross: Automatic feature crossing for tabular data in real-world applications. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 1936–1945.
[11]
Mia C Mayer, Muhammad Bilal Zafar, 2023. Hands-on Tutorial:" Explanations in AI: Methods, Stakeholders and Pitfalls". In KDD. 5783–5785.
[12]
Inzamam Mashood Nasir, Muhammad Attique Khan, Mussarat Yasmin, Jamal Hussain Shah, Marcin Gabryel, Rafał Scherer, and Robertas Damaševičius. 2020. Pearson correlation-based feature selection for document classification using balanced training. Sensors 20, 23 (2020), 6793.
[13]
Milos Radovic, Mohamed Ghalwash, Nenad Filipovic, and Zoran Obradovic. 2017. Minimum redundancy maximum relevance feature selection approach for temporal gene expression data. BMC bioinformatics 18 (2017), 1–14.
[14]
Benedikt Schifferer, Chris Deotte, and Even Oldridge. 2020. Tutorial: feature engineering for recommender systems. In Recsys. 754–755.
[15]
Yejing Wang, Zhaocheng Du, Xiangyu Zhao, 2023. Single-shot Feature Selection for Multi-task Recommendations. In SIGIR. 341–351.
[16]
Beichuan Zhang, Chenggen Sun, Jianchao Tan, Xinjun Cai, Jun Zhao, Mengqi Miao, Kang Yin, Chengru Song, Na Mou, and Yang Song. 2023. SHARK: A Lightweight Model Compression Approach for Large-scale Recommender Systems. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. 4930–4937.
[17]
Jingsen Zhang, Xu Chen, Jiakai Tang, 2023. Recommendation with causality enhanced natural language explanations. In WWW. 876–886.
[18]
Yongfeng Zhang, Xu Chen, 2020. Explainable recommendation: A survey and new perspectives. Foundations and Trends® in Information Retrieval 14, 1 (2020), 1–101.
[19]
Yuang Zhao, Zhaocheng Du, Qinglin Jia, Linxuan Zhang, Zhenhua Dong, and Ruiming Tang. 2024. Retrievable Domain-Sensitive Feature Memory for Multi-Domain Recommendation. arXiv preprint arXiv:2405.12892 (2024).

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences
RecSys '24: Proceedings of the 18th ACM Conference on Recommender Systems
October 2024
1438 pages
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 08 October 2024

Permissions

Request permissions for this article.

Check for updates

Qualifiers

  • Tutorial
  • Research
  • Refereed limited

Conference

Acceptance Rates

Overall Acceptance Rate 254 of 1,295 submissions, 20%

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • 0
    Total Citations
  • 158
    Total Downloads
  • Downloads (Last 12 months)158
  • Downloads (Last 6 weeks)12
Reflects downloads up to 18 Feb 2025

Other Metrics

Citations

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

HTML Format

View this article in HTML Format.

HTML Format

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media