Symptom-Treatment Knowledge Graphs for Contagious and Chronic Disease Profiling

Kukkala, Muralidhar (2025) Symptom-Treatment Knowledge Graphs for Contagious and Chronic Disease Profiling. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (830kB) | Preview
Preview	PDF (Configuration Manual) Download (684kB) | Preview

Abstract

This paper provides an elucidated, KG, model of disease classification in accordance with the level of chronicity and contagiousness, through the integration of symptom and treatment relationship as obtained through biomedical texts. With the recent research on biomedical NLP and graph learning, the proposed framework addresses the problem of the unaddressed graph structure and explainability problems, using entity canonicalization that is conducted with the help of the SciSpaCy model, TF-IDF and BioBERT embeddings, community extraction, learning of the proposed model, and supervised learning, and GNN. Empirical evaluations have indicated that relational learning usage has a significant effect of improvement of efficacy in disease prediction since XGBoost provides a prediction score of up to 0.976 and GNN provides a prediction score of up to 0.890 when predicting disease according to the extent of chronicity. Knowledge graph analytics are used to find insights into structured communities of diseases and gaps in their treatment and deploy model-level and graph-level explainers, SHAP and GNNExplainer, to the comprehensible representation of model decisions. These results indicate the importance of an efficient integration of meaning based embeddings and application of graph-based learning towards the achievement of strong predictive and biomedical interpretability.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Subhnil, Shubham UNSPECIFIED
Uncontrolled Keywords:	Biomedical NLP; Knowledge Graphs; Disease Classification; Graph Neural Networks; Explainable AI; BioBERT; SHAP; GNNExplainer
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing R Medicine > Diseases R Medicine > Healthcare Industry
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	08 Sep 2026 08:53
Last Modified:	08 Sep 2026 08:53
URI:	https://norma.ncirl.ie/id/eprint/9874

Actions (login required)

View Item