Abstract:
In this paper, we study an automatic hypothesis generation (HG) problem, which involves finding meaningful implicit connections between scientific terms like diseases, chemicals, drugs, and genes from biomedical publication databases. Most previous studies used static term information and ignored the temporal dynamics of scientific term relations. In recent studies that considered dynamics, they learned scientific term representations rather than term-pair relations. Since the HG problem is to predict term-pair connections, it is more important to know how the connections were formed (in a dynamic process) than who the terms are connected to. This HG problem is a dynamic attributed graph connectivity prediction. Capturing node-pair (term-pair) temporal evolution is crucial. T-PAIR, an inductive edge (node-pair) embedding method, encodes the temporal node-pair relationship using the graphical structure and node attribute. The proposed model is tested on three graphs of Pubmed papers from Neurology, Immunotherapy, and Virology published until 2019. Predicting future term-pair relations between millions of seen terms (transductive) and unseen terms (inductive) were evaluated. Experiments and case studies prove the model works.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
