Contextual Embedding-based Clustering to Identify Topics for Healthcare Service Improvement

Document Type

Conference Proceeding

Publication Date

2025

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Source Publication

IEEE 49th Annual Computers, Software, and Applications Conference (COMPSAC)

Source ISSN

2836-3795

Original Item ID

DOI: 10.1109/COMPSAC65507.2025.00106

Abstract

Understanding patient feedback is crucial for improving healthcare services, yet analyzing unlabeled short-text feedback presents challenges due to limited data and domainspecific nuances. Traditional supervised approaches require extensive labeled datasets, making unsupervised methods more practical for extracting insights. This study applies unsupervised techniques to analyze 439 survey responses from a healthcare system in Wisconsin, USA. A keyword-based filter was used to isolate complaint-related feedback using a domain-specific lexicon. To identify dominant themes, we evaluated traditional topic models such as Latent Dirichlet Allocation (LDA) and Gibbs Sampling Dirichlet Multinomial Mixture (GSDMM) - alongside BERTopic, a neural embedding-based clustering method. To improve coherence and interpretability in sparse, short-text data, we propose kBERT, which integrates BERT embeddings with k-means clustering. Model performance was assessed using coherence scores (Cv) and average Inverted Rank-Biased Overlap (IRBOavg). kBERT achieved the highest coherence (Cv = 0.53) and topic separation (IRBOavg = 1.00), outperforming all other models. These findings highlight the value of embedding-based, context-aware models in healthcare analytics.

Comments

Published as part of the proceedings of the IEEE 49th Annual Computers, Software, and Applications Conference (COMPSAC) 2025: 794-799. DOI.

Link to Full Text

Share

COinS