Abstract
A recommendation system uses a graph mining based approach to cluster web pages with similar contents and provide related content recommendations. The system receives a request from a user to access a first web page. The system then retrieves a list of secondary web pages that are associated with the first web page in an entity graph. The system calculates a similarity score between a secondary web page and the first web page. The similarity score between the first and a secondary web pages is calculated based on overlapping entities between the first web page and the secondary web page. The system then ranks the secondary web pages based on ascending similarity scores. Further, the system provides the list of secondary web pages, based on their similarity scores, to the user as an recommendation. PROBLEM STATEMENT
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Zhang, Wei and Gong, Xiaohong, "PERSONALIZED CONTENT RECOMMENDATIONS", Technical Disclosure Commons, (January 27, 2016)
https://www.tdcommons.org/dpubs_series/128