ORCID

https://orcid.org/0000-0002-9213-3825

Date of Award

Spring 5-14-2023

Author's School

McKelvey School of Engineering

Author's Department

Computer Science & Engineering

Degree Name

Master of Science (MS)

Degree Type

Thesis

Abstract

Survey data collected from human subjects can contain a high number of features while having a comparatively low quantity of examples. Machine learning models that attempt to predict outcomes from survey data under these conditions can overfit and result in poor generalizability. One remedy to this issue is feature selection, which attempts to select an optimal subset of features to learn upon. A relatively unexplored source of information in the feature selection process is the usage of textual names of features, which may be semantically indicative of which features are relevant to a target outcome. The relationships between feature names and target names can be evaluated using large language models (LLMs) such as ClinicalBERT to produce semantic textual similarity (STS) scores, which can then be used to select features. This thesis introduces two new variations upon the minimal-redundancy-maximal-relevance (mRMR) algorithm that integrate semantic textual similarity (STS) into selection. The performance of STS as a feature selection metric is evaluated against preliminary survey data collected as a part of a clinical study on persistent post-surgical pain (PPSP). The results suggest that features selected with STS can result in higher performance models compared to those with the baseline mRMR algorithm.

Language

English (en)

Chair

Chenyang Lu

Committee Members

Simon Haroutounian Thomas Kannampallil Cynthia Ma

Share

COinS