Knowledge Graph-based Recommender System using Skip-gram Network

Document Type



Knowledge Graph-based Recommender System using Skip-gram Network proposes an approach to generate item sequences for training a recommender system using a Knowledge Graph Skip-Gram (KGSG) model. The KGSG model extracts data from multiple sources, converts them into a knowledge graph, and implements biased random walk to generate item sequences. By controlling the balance between exploration and exploitation, the biased random walk produces a sequence of nodes that is well-suited for a specific application or use case. These generated sequences are used to create both positive and negative training examples, which are utilized to train a skip-gram model. The resulting items' embeddings can then be used to make personalized recommendations for new items to users based on their preferences and past viewing history. The proposed KGSG model leverages the power of knowledge graphs and skip-gram network to build a highly accurate and efficient recommender system.

Publication Date



Thesis submitted to the Deanship of Graduate and Postdoctoral Studies

In partial fulfillment of the requirements for the M.Sc degree in Natural Language Processing

Advisors: Dr. Shangsong Liang, Dr. Muhammad Harris

Online access available for MBZUAI patrons

This document is currently not available here.