Among different recommendation techniques, collaborative filtering usually suffer from limited performance due to the sparsity of user-item interactions. To address the issues, auxiliary information is usually used to boost the performance. Due to the rapid collection of information on the web, the knowledge base provides heterogeneous information including both structured and unstructured data with different semantics, which can be consumed by various applications. In this paper, researchers investigate how to leveragethe heterogeneous information in a knowledge base to improve thequality of recommender systems. First, by exploiting the knowledge base, they design three components to extract items’ semantic representations from structural content, textual content and visual content, respectively. To be specific, they adopt a heterogeneous network embedding method, termed as TransR, to extract items’ structural representations by considering the heterogeneity of both nodes and relationships. Tehy apply stacked denoising auto-encoders and stacked convolutional auto-encoders, which are two types of deep learning based embedding techniques, to extract items’ textual representations and visual representations, respectively. Finally, they propose our final integrated framework, which is termed as Collaborative Knowledge Base Embedding (CKE), to jointly learn the latent representations in collaborative filtering as well as items’ semantic representations from the knowledge base. To evaluate the performance of each embedding component as well as the whole system, they conduct extensive experiments with two real- world datasets from different scenarios. The results reveal that our approaches outperform several widely adopted state-of-the-art recommendation methods.