Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis
Regression analysis is one of the common machine learning method to model the relationship between dependent and independent variables. In this study, we aim to tackle two crucial elements that affect the performance of regression models, which are the type of crossing-over method used for model eva...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | en en |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/38467/1/ABSTRACT.pdf https://eprints.ums.edu.my/id/eprint/38467/2/FULLTEXT.pdf https://eprints.ums.edu.my/id/eprint/38467/ https://ieeexplore.ieee.org/document/9918722 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1831796041478832128 |
|---|---|
| author | Ong, Song Quan Pradeep Isawasan Khairulliza Ahmad Salleh |
| author_facet | Ong, Song Quan Pradeep Isawasan Khairulliza Ahmad Salleh |
| author_sort | Ong, Song Quan |
| building | UMS Library |
| collection | Institutional Repository |
| content_provider | Universiti Malaysia Sabah |
| content_source | UMS Institutional Repository |
| continent | Asia |
| country | Malaysia |
| description | Regression analysis is one of the common machine learning method to model the relationship between dependent and independent variables. In this study, we aim to tackle two crucial elements that affect the performance of regression models, which are the type of crossing-over method used for model evaluation and multivariate analysis with the number of predictors. We used the classic thyroid disease dataset from the UCI machine learning repository and compare the crossing-over approaches of k-fold with different folds, bootstrap, Leave One Out Cross-Validation (LOOCV), and repeated k-fold on linear and logistics regression. For multivariate analysis, we compare the performance of the models by using the different combinations of bi-predictors and multi-predictors. Our result shows that models that use kfold cross-validation have greater performance, and a higher number of k does not improve the model performance. For the multivariate analysis, we found that the number of variable is not the key element to determine the performance of a model, rather than a suitable combination of strong predictors. Future studies could explore the effects of cross-validation and multivariate analysis on other machine learning algorithms. |
| format | Conference or Workshop Item |
| id | my.ums.eprints-38467 |
| institution | Universiti Malaysia Sabah |
| language | en en |
| publishDate | 2022 |
| record_format | eprints |
| spelling | my.ums.eprints-384672024-05-02T08:40:44Z https://eprints.ums.edu.my/id/eprint/38467/ Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis Ong, Song Quan Pradeep Isawasan Khairulliza Ahmad Salleh QA299.6-433 Analysis Regression analysis is one of the common machine learning method to model the relationship between dependent and independent variables. In this study, we aim to tackle two crucial elements that affect the performance of regression models, which are the type of crossing-over method used for model evaluation and multivariate analysis with the number of predictors. We used the classic thyroid disease dataset from the UCI machine learning repository and compare the crossing-over approaches of k-fold with different folds, bootstrap, Leave One Out Cross-Validation (LOOCV), and repeated k-fold on linear and logistics regression. For multivariate analysis, we compare the performance of the models by using the different combinations of bi-predictors and multi-predictors. Our result shows that models that use kfold cross-validation have greater performance, and a higher number of k does not improve the model performance. For the multivariate analysis, we found that the number of variable is not the key element to determine the performance of a model, rather than a suitable combination of strong predictors. Future studies could explore the effects of cross-validation and multivariate analysis on other machine learning algorithms. 2022-10-26 Conference or Workshop Item PeerReviewed text en https://eprints.ums.edu.my/id/eprint/38467/1/ABSTRACT.pdf text en https://eprints.ums.edu.my/id/eprint/38467/2/FULLTEXT.pdf Ong, Song Quan and Pradeep Isawasan and Khairulliza Ahmad Salleh (2022) Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis. In: 2022 3rd International Conference on Artificial Intelligence and Data Sciences (AiDAS), 07-08 September 2022, IPOH, Malaysia. https://ieeexplore.ieee.org/document/9918722 |
| spellingShingle | QA299.6-433 Analysis Ong, Song Quan Pradeep Isawasan Khairulliza Ahmad Salleh Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis |
| title | Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis |
| title_full | Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis |
| title_fullStr | Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis |
| title_full_unstemmed | Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis |
| title_short | Regression study for thyroid disease prediction Comparison of crossing-over approaches and multivariate analysis |
| title_sort | regression study for thyroid disease prediction comparison of crossing-over approaches and multivariate analysis |
| topic | QA299.6-433 Analysis |
| url | https://eprints.ums.edu.my/id/eprint/38467/1/ABSTRACT.pdf https://eprints.ums.edu.my/id/eprint/38467/2/FULLTEXT.pdf https://eprints.ums.edu.my/id/eprint/38467/ https://ieeexplore.ieee.org/document/9918722 |
| url_provider | http://eprints.ums.edu.my/ |