An Enhanced Neural Graph based Collaborative Filtering with Item Knowledge Graph

Recommendation system is a process of filtering information to retain buyers on e-commerce sites or applications. It used all sites, social media platform and multimedia platform. This recommendation based their own experience between users. In recent days, the graph-based techniques are for improve suggestions easy analysing. Neural graph collaborative also one system. implemented benchmark datasets like Yelp, Gowalla Amazon books. technique can suggest better recommendations as compared existing convolutional networks. However, it requires higher processing time neural network performing limited suggestions. Hence, in this paper, an improved proposed. Here, content-based performed before process. Then, embedding layer will both provide order relation users items. As suggestion hybrid recommendation, Convolutional reduced by reducing number epochs. Due this, final not affected smaller epochs able reduce its computational time. The whole realized Python 3.6 under windows 10 environment Go Walla Based comparison recall NDCG metric, proposed outperforms convolution network.