Distilling Structured Knowledge into Embeddings for Explainable and Accurate Recommendation
This addresses the problem of opaque and sparse data in recommendation systems for users and developers, offering an incremental improvement by integrating knowledge distillation without extra overhead.
The paper tackles the lack of explainability and data sparsity in embedding-based recommendation models by proposing an end-to-end joint learning framework that distills structured knowledge from a differentiable path-based model, achieving state-of-the-art performance and providing interpretable recommendations.
Recently, the embedding-based recommendation models (e.g., matrix factorization and deep models) have been prevalent in both academia and industry due to their effectiveness and flexibility. However, they also have such intrinsic limitations as lacking explainability and suffering from data sparsity. In this paper, we propose an end-to-end joint learning framework to get around these limitations without introducing any extra overhead by distilling structured knowledge from a differentiable path-based recommendation model. Through extensive experiments, we show that our proposed framework can achieve state-of-the-art recommendation performance and meanwhile provide interpretable recommendation reasons.