Extracting the structure of the explanation for the biomedical domain

Organization
Arvutiteaduse instituut
Abstract
The scientific explanation is one of the most critical aspects of scientific practice.
This thesis aims to find the textual structures by which the scientific explanation is realized in the biomedicine domain. A corpus of scientific articles related to biomedicine (but see also Nota bene) will be collected from PubMed. The thesis work will extend the seminal work of Overtone (2012), where he builds a small corpus of explanation patterns and classifies them according to an ontology.
The student will produce his/her formalization of the textual explanation structures in the chosen scientific domain. Some textual explanation patterns will be defined as a machine learning task (most likely binary or ternary relation extraction) and then attempt to learn them automatically.

Nota bene. The scientific domain in which the study will be performed can be changed from biomedicine to one where the student is competent.
Graduation Theses defence year
2022-2023
Supervisor
Eduard Barbu
Spoken language (s)
English
Requirements for candidates
Level
Masters
Keywords
#relation_extraction, #explanation, #biomedicine

Application of contact

 
Name
EDUARD BARBU
Phone
E-mail
eduard.barbu@ut.ee