Tools
Tools
Wedage, Anjana Sasanka (2024) Comparative Evaluation of SMOTE Algorithms in Predictive Modelling of Cirrhosis. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (991kB) | Preview |
Abstract
Cirrhosis has become a major medical issue in the world and people suffer from it in every corner of the world. The medical world is working on many treatments and procedures to save people from loosing lives. In the modern world, machine learning has given hope by developing models to detect the disease early. This paper focuses on using Synthetic Minority Over-sampling Technique (SMOTE) to improve predictions from the machine learning models. Mostly medical data has imbalanced data in the datasets, and it affects the prediction ability and SMOTE helps to get an even balanced dataset for the analysis. The study focuses on 9 SMOTE techniques and evaluated using 5 machine learning models. The results showed that Adaptive SMOTE technique showed better performance than other SMOTE techniques and specifically Adaptive SMOTE with K-Nearest Neighbors showed the better metrics accuracy and the recall of the stages combined.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Agarwal, Bharat UNSPECIFIED |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning Q Science > Life sciences > Medical sciences |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Ciara O'Brien |
| Date Deposited: | 26 Aug 2025 12:24 |
| Last Modified: | 26 Aug 2025 12:24 |
| URI: | https://norma.ncirl.ie/id/eprint/8649 |
Actions (login required)
 |
View Item |