Breaking the Information Silo: Semantic Personas for Cross-Domain Recommendation

2026-06-01

AI Summary

This work addresses the cross-domain information silo problem arising from the absence of shared users or items across independent digital platforms. To this end, it proposes SPHERE, a novel framework that formulates cross-domain recommendation as behavior-based semantic alignment—eliminating the need for shared entities or graph structures. SPHERE leverages large language models to construct user semantic profiles and generates community-level source profiles by matching source-domain communities based on behavioral similarity. Integrated with a dual-tower architecture and a dynamic fusion gating mechanism, SPHERE enhances the cross-domain transferability of existing recommender models such as NCF and LightGCN. Experiments on Amazon Books, Goodreads, and Steam datasets demonstrate that SPHERE significantly outperforms state-of-the-art baselines, revealing that cross-domain performance hinges more on the structural density and predictive capacity of the target domain than on mere semantic similarity, while offering strong interpretability and modular flexibility.

Abstract

Digital platforms increasingly operate as isolated information silos, limiting their ability to construct comprehensive user representations across domains. Cross-domain recommender systems seek to overcome this limitation by transferring knowledge from a source domain to a target domain, yet most existing approaches depend on shared users, shared items, or structurally similar interaction graphs. These assumptions are often unrealistic across independent platforms. We propose SPHERE (Semantic Personas for Heterogeneous cross-domain Recommendation), a design artifact that enables recommendation knowledge transfer across strictly disjoint domains with no shared users or items. Rather than aligning domains through identity or graph structure, SPHERE uses large language models to induce a shared behavioral vocabulary, generate structured semantic personas for users, and retrieve behaviorally similar source-domain communities that form a Community Source Persona. This semantic signal is integrated with collaborative signals through a dual-tower architecture and dynamic fusion gate, allowing SPHERE to augment standard recommender backbones. Empirical evaluation across Amazon Books, Goodreads, and Steam demonstrates consistent improvements over NCF, SVD++, and LightGCN baselines under full-ranking evaluation. The results show that cross-domain transfer effectiveness is not determined solely by semantic proximity between domains; rather, it depends critically on the structural density and native predictive strength of the target domain. The study contributes to information systems research by reframing cross-domain personalization as behavior-based semantic alignment, offering a practical mechanism for overcoming information silos while preserving interpretability and modularity.

Problem

• information silos
• cross-domain recommendation
• user representation
• knowledge transfer
• disjoint domains

Innovation

• cross-domain recommendation
• semantic personas
• large language models
• information silos
• behavioral alignment