Recommendation with Deep Structured Semantic Model

Deep Structured Semantic Model (DSSM) [1] is a deep neural network for learning semantic representations of entities in a common continuous semantic space and measuring their semantic similarities. It is widely used in information retrieval area and is supremely suitable for top-n recommendation [39, 182]. DSSM projects different entities into a common low-dimensional space, and computes their similarities with cosine function. Basic DSSM is made up of MLP. Note that, more advanced neural layers such as convolution and max-pooling layers can also be easily integrated into DSSM.

reference
1.Learning Deep Structured Semantic Models for Web Search using Clickthrough Data(2013)
2.『 DSSM』A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems