박성민 석박통합과정, 윤민철 석사과정, 최민진 석박통합과정 WSDM 2025 국제 학술대회 논문 채택

24 Oct 2024

DIAL 연구실 소속 인공지능학과 박성민(석·박통합과정, 공동 1저자) 학생,  윤민철(석사과정, 공동 1저자) 학생, 최민진(석·박통합과정, 제3저자) 학생, 이종욱(교신저자) 교수가 참여한 논문 “Temporal Linear Item-Item Model for Sequential Recommendation” 이 데이터마이닝 분야 최우수 국제 학술대회인 The 18th ACM International Conference on Web Search and Data Mining (WSDM 2025)에 최종 게재가 승인되었으며 2025년 3월에 발표될 예정입니다.

Abstract

In sequential recommendation (SR), neural models have been actively explored due to their remarkable performance, but they suffer from inefficiency inherent to their complexity. On the other hand, linear SR models exhibit high efficiency and achieve competitive or superior accuracy compared to neural models. However, they solely deal with the sequential order of items (i.e., sequential information) while overlooking the actual timestamp (i.e., temporal information). It is limited to effectively capturing various user preference drifts over time. To address this issue, this paper introduces a simple-yet-effective linear SR model, named TemporAl LinEar item-item model (TALE), incorporating temporal information while preserving training and inference efficiency, with three key components. (i) Single-target augmentation concentrates on a single target item, enabling linear models to learn the temporal correlation for the target item. (ii) Time interval-aware weighting utilizes the actual timestamp to discern item correlation patterns over various time intervals. (iii) Trend-aware normalization reflects the dynamic shift of item popularity over time. Our empirical studies show that TALE outperforms ten competing SR models by up to 18.71% gains on five benchmark datasets. TALE also exhibits remarkable effectiveness in evaluating long-tail items by up to 30.45% gains.