Jitao Sang

TL;DR: A multimodal multitask deep belief network to learn joint image-ingredient representation regularized by different attributes for recipe retrieval and exploration and incorporates multitask learning to make different attributes collaborate each other.

TL;DR: A novel Explainable Interaction-driven User Modeling (EIUM) algorithm to exploit Knowledge Graph for constructing an effective and explainable sequential recommender and captures the interaction-level user dynamic preferences by modeling the sequential interactions.

TL;DR: A unified Contextual Self-Attention Network (CSAN) is proposed to address the three properties of heterogeneous user behaviors, which are projected into a common latent semantic space and fed into the feature-wise self-attention network to capture the polysemy of user behaviors.

TL;DR: This paper considers the problem of automatically constructing a folksonomy-based visual ontology (FBVO) from the user-generated annotated images and proposes a systematic framework consisting of three stages as concept discovery, concept relationship extraction, and concept hierarchy construction.

TL;DR: This article proposes a unified YouTube video recommendation solution by transferring and integrating users’ rich social and content information in Twitter network and shows that the proposed cross-network collaborative solution achieves superior performance not only in terms of accuracy, but also in improving the diversity and novelty of the recommended videos.