| Abstract |
The growing number of scientific published papers has resulted in information overload, which also raises the problem that researchers have difficulty in finding papers relevant to their interests. Recommendation system is a good solution to address this problem. Most of the previous works have focused on the accuracy of recommendations. However, several works argue that there are important aspects other than accuracy, such as serendipity and explainability. Knowledge Graphs contain a large number of semantic associations and may provide great values in revealing interesting connections between different terms/concepts/topics extracted from research articles. Therefore, my work is focused on how to leverage Knowledge Graphs to represent researcher profiles as well as candidate papers to be recommended so as to generate recommendations that will not only surprise users but also provide users with explanations on why these articles are recommended. |