Experiment Plan Discovery of Novel Conceptual Relations Based on Biomedical Literature

Abstract

With the large-scale knowledge bases provided by artificial intelligence, I have been thinking about whether AI can be used to explore new ideas and viewpoints. In this paper, I attempt to discover new possible connections between concepts. I screen concept groups that appear infrequently in existing papers but are semantically plausible, and use these concept groups together with the relevant paper sentences as evidence supporting the relations between them to learn new relations. Finally, I use the learned new relations to generate new concept groups and examine the plausibility of the newly generated concept groups.

Keywords