Publication Type : Journal Article
Publisher : IEEE
Url : https://doi.org/10.1109/SILCON59133.2023.10404380
Keywords : Machine learning algorithms; Predictive models; Breast cancer; MCDM; Random forests; Optimization; Medical diagnostic imaging; Machine Learning; Regression; MCDM; MOORA MCDM
Campus : Faridabad
Year : 2023
Abstract : Breast cancer is a serious health issue that affects women all over the globe. For improved treatment outcomes, early breast cancer identification is essential. However, the present diagnostic procedure relies on time-consuming examination, providing difficulties for effective diagnosis.In order to analyze early identification of breast cancer; such a system would be created to make use of modern innovations like artificial intelligence and machine learning algorithms. Eight regression approaches are used in this work to anticipate breast cancer utilizing essential tumor cell data, including MLP, Linear Regression, LGBM Regressor, XGB Regressor, Kernel Ridge, Bayesian Ridge, SVR, and Random Forest Regression. The objective of the study is to choose the best regression model for making precise predictions. The MOORA (Multi Objective Optimization by Ratio Analysis) MCDM (Multi Criteria Decision Making) approach has been used to find out the best optimal ML regression for predicting the incidence of breast cancer. Finally, using the top-ranked two ML regressors, the fitting analysis and prediction accuracy was explored. It has been observed that the XGB Regressor is the best suited for the prediction of breast cancer having an RMSE value of 0.0483. The second most effective ML Regressor is MLP with a RMSE value of 0.0322. The suggested methodology in breast cancer prediction demonstrates the possibility for creating effective self-sustaining diagnostic systems.
Cite this Research Publication : Apurba Debnath, Anirban Tarafdar, Paritosh Bhattacharya, Azharuddin Shaikh, MOORA MCDM based Optimal Machine Learning Regression Techniques for Breast Cancer Prediction, [source], IEEE, 2023, https://doi.org/10.1109/SILCON59133.2023.10404380