Integrating Item Relevance in Training Loss for Sequential Recommender Systems

Andrea Bacciu, Federico Siciliano, Nicola Tonellotto, Fabrizio Silvestri
5/18/2023
cs.IRcs.AI

No official code found. Best implementation: andreabac3/Integrating-Item-Relevance-in-Training-Loss-for-Sequential-Recommender-Systems (7 stars, last commit 2023-06-10).

Abstract

Sequential Recommender Systems (SRSs) are a popular type of recommender system that learns from a user's history to predict the next item they are likely to interact with. However, user interactions can be affected by noise stemming from account sharing, inconsistent preferences, or accidental clicks. To address this issue, we (i) propose a new evaluation protocol that takes multiple future items into account and (ii) introduce a novel relevance-aware loss function to train a SRS with multiple future items to make it more robust to noise. Our relevance-aware models obtain an improvement of ~1.2% of NDCG@10 and 0.88% in the traditional evaluation protocol, while in the new evaluation protocol, the improvement is ~1.63% of NDCG@10 and ~1.5% of HR w.r.t the best performing models.

View on arXivView PDF

Code Implementations(1)

andreabac3/Integrating-Item-Relevance-in-Training-Loss-for-Sequential-Recommender-SystemsCommunity71%

Integrating Item Relevance in Training Loss for Sequential Recommender Systems

70Shell, Python3 years ago
loss-functionsrecommendation-systemrecommender-systemrecommender-systemssequential-recommendation

Similar Papers

SPR:Supervised Personalized Ranking Based on Prior Knowledge for Recommendation

86% match2022

Large Language Model Enhanced Hard Sample Identification for Denoising Recommendation

83% match2024

Unbiased Recommender Learning from Missing-Not-At-Random Implicit Feedback

83% match2 repos302019

LLM4DSR: Leveraging Large Language Model for Denoising Sequential Recommendation

82% match1 repo102024

On Softmax Direct Preference Optimization for Recommendation

82% match1 repo1002024

Learning from Negative User Feedback and Measuring Responsiveness for Sequential Recommenders

82% match2023

Cite this paper

@article{bacciu2023integrating,
  title  = {Integrating Item Relevance in Training Loss for Sequential Recommender Systems},
  author = {Andrea Bacciu and Federico Siciliano and Nicola Tonellotto and Fabrizio Silvestri},
  year   = {2023},
  eprint = {2305.10824},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.10824v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.10824.svg)](https://codeofpaper.com/paper/2305.10824)

Discussion