Date of Submission
2024
Supervisor
Dr. Sajjad Haider, Professor, Department of Computer Science, Institute of Business Administration, Karachi
Committee Member 1
Dr. Sajjad Haider, Supervisor, Department of Computer Science, Institute of Business Administration, Karachi
Committee Member 2
Dr. Quratulain Rajput, Examiner – I, Department of Computer Science, Institute of Business Administration, Karachi
Committee Member 3
Dr. Tariq Mahmood, Examiner – II, Department of Computer Science, Institute of Business Administration, Karachi
Degree
Master of Science in Computer Science
Department
Department of Computer Science
School
School of Mathematics and Computer Science (SMCS)
Keywords
Ontology knowledge, Natural language processing, Pattern matching rules, Feature extraction, Clinical notes
Abstract
Feature extraction from clinical narratives involves the application of natural language processing (NLP) techniques and knowledge extraction. However, it has been observed that integrating NLP techniques with an ontology-guided approach in clinical notes has not been widely used in the feature extraction pipeline. Consequently, the utilization of ontology knowledge in feature extraction has not been extensively explored.
This thesis demonstrates the feasibility of integrating ontology knowledge into feature extraction alongside NLP techniques. The approach involves building a system that incorporates ontology knowledge using owlready2 on the Harvard medical dataset. To enable pattern matching, rules are formulated using regular expressions.
The findings reveal that the extracted features, while sufficient for demonstrating feasibility, can be effectively employed in any analytical model. The pattern matching rules are specifically tailored to the Harvard medical dataset, suggesting potential modifications may be required to apply this system to other datasets. Although this thesis focuses on a subset of medical ontologies, it establishes that the system is adaptable to incorporating additional medical ontologies.
Document Type
Restricted Access
Submission Type
Thesis
Recommended Citation
Khatri, M. (2024). Using Natural Learning and Ontology guided techniques for feature extraction from Clinical Narratives: Hybrid Approach (Unpublished unpublished graduate thesis). Retrieved from https://ir.iba.edu.pk/etd-ms-cs/1
