Recommender Systems Based on Knowledge Graphs

In the present situation COVID19 is one of challenging threats to the whole world. People in many countries are infected with an exponential rate in case of direct contact. Almost all countries have proposed their best measures and strict policies for breakdown the speed of COVID19. The whole world taking such restrictive steps like lockdown, border closing, partial, and full closing of all kinds of private and public institutes to control the coronavirus spreading. Now, every county is falling under an economic crisis for the corona case increasing. In my thesis, one of the parts we proposed a link prediction model based on knowledge graph embedding that can trace the contact of COVID19 positive patients and keeping them quarantined so that they are not spread it to others. On the other hand, nowadays, lots of exertion has been spent on look into collaboration recommendations because of the quick development of science and innovation, a lot of continually refreshed scientific accomplishments containing inventive information can be procured and used to take care of issues. The endorsement of the cross-disciplinary scientific alliance has also been demonstrated by scientific research and technological modernization in recent times. Moreover, enormous data have generated every second that can be difficult for representing traditional techniques. In this paper we used a state-of-art data representing technique named Knowledge Graph for building a cross disciplinarian author recommendation system.