Price-guided user attention in large-scale E-commerce group recommendation

TL;DR

This work tackles how item price shapes which users influence group recommendations in large-scale E-commerce. It introduces PGUsA, an adaptive sigmoid-based attention module that modulates user weights using item price and purchase frequency, and demonstrates that it can be plugged into existing models such as AGREE or ConsRec. Through experiments on Epinion and Ichiba (including a denser variant), PGUsA improves metrics like HR, NDCG, and MSE, and also boosts GMV while maintaining efficiency. The findings reveal that frequent buyers are more influential for low-priced items, and the price-aware weighting adapts as price changes to yield practical gains in price-sensitive group recommendations.

Abstract

Existing group recommender systems utilize attention mechanisms to identify critical users who influence group decisions the most. We analyzed user attention scores from a widely-used group recommendation model on a real-world E-commerce dataset and found that item price and user interaction history significantly influence the selection of critical users. When item prices are low, users with extensive interaction histories are more influential in group decision-making. Conversely, their influence diminishes with higher item prices. Based on these observations, we propose a novel group recommendation approach that incorporates item price as a guiding factor for user aggregation. Our model employs an adaptive sigmoid function to adjust output logits based on item prices, enhancing the accuracy of user aggregation. Our model can be plugged into any attention-based group recommender system if the price information is available. We evaluate our model's performance on a public benchmark and a real-world dataset. We compare it with other state-of-the-art group recommendation methods. Our results demonstrate that our price-guided user attention approach outperforms the state-of-the-art methods in terms of hit ratio and mean square error.

Figures (9)

• Figure 1: User preference aggregation in AGREE. $u_p$ and $u_q$ are user representations in group $l$, $i_o$ is item $o$'s representation, $g_{lo}$ is group $l$'s representation for item $o$, $w_p$ and $w_q$ are the user aggregation weights.
• Figure 2: Item distribution segmented by item price. Set A is a set of items where the most influential user is a frequent buyer. Set B contains the rest of the items.
• Figure 3: Example of adaptive sigmoid functions. The output weight changes based on item price-related parameter $\alpha_o$ and user purchase frequency-related parameter $p_t$.
• Figure 4: Price-guided user attention (PGUsA) module. We utilize user shopping frequency and item price information to adjust the user aggregation weight.
• Figure 5: AGREE model with PGUsA group embedding $g_{lo}$ (shown in purple).