Time Matters: Enhancing Sequential Recommendations with Time-Guided Graph Neural ODEs

TL;DR

Time Matters addresses two core challenges in sequential recommendation: irregular user interests and highly uneven item distributions over time. The authors propose TGODE, a framework built on two graphs (a user time graph and an item evolution graph), a time-guided diffusion generator to augment sparse temporal data, a user preference inference module to balance temporal pivots, and a generalized graph neural ODE to align the evolution of user and item dynamics. Through iterative training that couples diffusion augmentation with continuous-time graph evolution, TGODE achieves substantial improvements over state-of-the-art baselines across five real-world datasets (roughly 10% to 46% gains). This work advances time-aware SR by effectively handling temporal sparsity and exogenous promotion effects, enabling more accurate and temporally consistent recommendations in real-world platforms.

Abstract

Sequential recommendation (SR) is widely deployed in e-commerce platforms, streaming services, etc., revealing significant potential to enhance user experience. However, existing methods often overlook two critical factors: irregular user interests between interactions and highly uneven item distributions over time. The former factor implies that actual user preferences are not always continuous, and long-term historical interactions may not be relevant to current purchasing behavior. Therefore, relying only on these historical interactions for recommendations may result in a lack of user interest at the target time. The latter factor, characterized by peaks and valleys in interaction frequency, may result from seasonal trends, special events, or promotions. These externally driven distributions may not align with individual user interests, leading to inaccurate recommendations. To address these deficiencies, we propose TGODE to both enhance and capture the long-term historical interactions. Specifically, we first construct a user time graph and item evolution graph, which utilize user personalized preferences and global item distribution information, respectively. To tackle the temporal sparsity caused by irregular user interactions, we design a time-guided diffusion generator to automatically obtain an augmented time-aware user graph. Additionally, we devise a user interest truncation factor to efficiently identify sparse time intervals and achieve balanced preference inference. After that, the augmented user graph and item graph are fed into a generalized graph neural ordinary differential equation (ODE) to align with the evolution of user preferences and item distributions. This allows two patterns of information evolution to be matched over time. Experimental results demonstrate that TGODE outperforms baseline methods across five datasets, with improvements ranging from 10% to 46%.

Figures (8)

• Figure 1: Illustration example of sequential recommendation with (a) irregular user interests and (b) highly uneven item distributions.
• Figure 2: Data analysis regarding the Beauty dataset.
• Figure 3: The overall architecture of our proposed method.
• Figure 4: Comparison with different sequence decoders.
• Figure 5: Sequence length within <10, 10-20, >20.