Using KG for Recommendation Systems
In recent years, recommender systems have played an increasingly significant role in helping users discover items of interest from numerous resource collections in various online services. Recommendation methods Include collaborative filtering (CF), a content-based approach, and a hybrid of both methods.
With the constant emergence of knowledge graph (KG), a growing number of researchers have leveraged the knowledge base such as YAGO, NELL, DBpedia, and DeepDive with rich knowledge to improve the recommendation performance. KGs are built as semantic networks, which means we could calculate the semantic similarities between the entities and are able to fuse various recommendation models and can handle the issues of data sparsity and cold start.
As an example in the movie domain, a recommender system can offer you some unwatched movies which you probably like, based on a KG that includes collaborative information, users, items, and their attributes. (the above picture)