用于冷启动推荐的异质信息网络对比元学习

在推荐系统中,冷启动推荐由于缺乏用户和物品交互信息而具有很大的挑战性.该问题可以由数据层和模型层的策略进行缓解.传统的数据层方法利用如特征信息的辅助信息来增强用户和物品表示的学习.最近,异质信息网络被整合于推荐系统中.它可以提供更丰富的辅助信息和更有意义的语义信息.但是,这些模型无法充分利用结构和语义信息,并且忽视了网络中的无标签信息.模型层的方法应用了元学习框架,该框架通过学习相似任务的先验知识然后利用很少的标签信息适应新任务,与冷启动问题相似.综上,我们提出了一个基于异质信息网络的对比元学习框架CM-HIN,同时在数据层和模型层解决冷启动问题.具体的,利用元路径和网络模式两个视图分别刻画异质信息网络的高阶以及本地结构信息.在元路径和网络模式视图中,采用对比学习挖掘异质信息网络的无标签信息并整合两个视图.在三个基准数据集上的三个冷启动推荐场景的大量实验中,CM-HIN超越了所有先进的基线模型.

Contrastive Meta-learning on Heterogeneous Information Networks for Cold-start Recommendation

In recommender system, cold-start issue is challenging due to the lack of interactions between new users or new items. Such issue could be alleviated via data-level and model-level strategies. Traditional data-level methods employ side information like feature information to enhance the learning of user and item embeddings. Recently, heterogeneous information networks (HINs) have been incorporated into the recommender system as they provide more fruitful auxiliary information and meaningful semantics. However, these models are unable to capture the structural and semantic information comprehensively and neglect the unlabeled information of HINs during training. Model-level methods propose to apply the meta-learning framework which naturally fits into the cold-start issue, as it learns the prior knowledge from similar tasks and adapt to new tasks quickly with few labeled samples. Therefore, we propose a contrastive meta-learning framework on HINs named CM-HIN, which addresses the cold-start issue in both data level and model level. In specific, we explore metapath and network schema views to describe the higher-order and local structural information of HINs. Within metapath and network schema views, contrastive learning is adopted to mine the unlabeled information of HINs and incorporate these two views. Extensive experiments on three benchmark datasets demonstrate that CM-HIN outperforms all state-of-the-art baselines in three cold-start scenarios.