Improving medical search tasks using learning to rank

Learning to Rank (LtR) is a new emerging research area in which machine learning techniques are used to solve the problem of ranking search results. In this paper, we consider applying LtR to the tasks performed by physicians in seeking out relevant information for providing better care to their pat...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB) pp. 1 - 8
Main Authors Alsulmi, Mohammad, Carterette, Ben
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2018
Subjects
Biological system modeling
clinical decision support
Computational modeling
Feature extraction
learning to rank
medical case search
Medical diagnostic imaging
Medical services
Task analysis
Training
Online AccessGet full text

Cover

More Information
Summary:Learning to Rank (LtR) is a new emerging research area in which machine learning techniques are used to solve the problem of ranking search results. In this paper, we consider applying LtR to the tasks performed by physicians in seeking out relevant information for providing better care to their patients. More specifically, we present a general approach for applying learning to rank (LtR), broadly applicable across many different algorithms, to clinical search tasks. Our approach consists of a set of learning features as well as a feature selection method which can help in learning effective models for retrieving relevant biomedical literature. In our experiments, we examine our approach using several state-of-the-art LtR algorithms and show that the proposed LtR ranking can effectively promote search results for the clinical domain resulting in a performance increase up to 28% compared to traditional ranking models.
DOI:10.1109/CIBCB.2018.8404965