Content-Based Collaborative Generation for Recommender Systems

TL;DR

This paper proposes a sequence-to-sequence framework which is tailored for directly generating the recommended item identifier, namely ColaRec, and proposes an item indexing task to conduct the alignment between the content-based semantic space and the interaction-based collaborative space.

Abstract

Generative models have emerged as a promising utility to enhance recommender systems. It is essential to model both item content and user-item collaborative interactions in a unified generative framework for better recommendation. Although some existing large language model (LLM)-based methods contribute to fusing content information and collaborative signals, they fundamentally rely on textual language generation, which is not fully aligned with the recommendation task. How to integrate content knowledge and collaborative interaction signals in a generative framework tailored for item recommendation is still an open research challenge. In this paper, we propose content-based collaborative generation for recommender systems, namely ColaRec. ColaRec is a sequence-to-sequence framework which is tailored for directly generating the recommended item identifier. Precisely, the input sequence comprises data pertaining to the user's interacted items, and the output sequence represents the generative identifier (GID) for the suggested item. To model collaborative signals, the GIDs are constructed from a pretrained collaborative filtering model, and the user is represented as the content aggregation of interacted items. To this end, ColaRec captures both collaborative signals and content information in a unified framework. Then an item indexing task is proposed to conduct the alignment between the content-based semantic space and the interaction-based collaborative space. Besides, a contrastive loss is further introduced to ensure that items with similar collaborative GIDs have similar content representations. To verify the effectiveness of ColaRec, we conduct experiments on four benchmark datasets. Empirical results demonstrate the superior performance of ColaRec.

Figures (5)

• Figure 1: Illustration of collaborative signals and content information. Collaborative signals refer to the knowledge contained in user-item interactions while content information refers to the textual description of items.
• Figure 2: Comparison between conventional itemIDs and GIDs. GIDs contain more concrete correlations.
• Figure 3: Overview of ColaRec. ColaRec assigns each item with a GID obtained from a GNN-based CF model. ColaRec consists two tasks. User-Item Recommendation aims to map the user’s interacted items with textual content into the GID of the recommended item, i.e., $\mathcal{L}_{\text{rec}}$. Item-Item Indexing targets on the mapping from item side information into the item’s GID, i.e., $\mathcal{L}_{\text{index}}$. Besides, a ranking loss $\mathcal{L}_{\text{bpr}}$ and a contrastive loss $\mathcal{L}_{\text{c}}$ are also introduced.
• Figure 4: Impact of the length of GIDs.
• Figure 5: Impact of the number of clusters.