A book recommendation system based on named entities
Recommendation systems are extensively used for suggesting new items to users and play an important role in the discovery of relevant new items, be it books, movies or music. An effective recommendation system should provide heterogeneous results and should not be biased towards only the most popular items. Books are particularly well-suited to content-based filtering as they are now widely available in digital formats which can allow various text mining approaches to dig out content related information. This paper presents a framework to develop a content-based recommendation system for books which can further be integrated with a collaborative filtering model. The proposed content-based recommender will use the Named Entities as the basic criteria to rank books and give recommendations.
Natural Language Processing, Recommendation Systems, Named Entity Extraction, content-based, similarity metrics, Text Mining
Full Text: PDF (downloaded 147 times)