Scholarlink

Objective <p>This study aims to investigate the efficacy of artificial intelligence (AI) techniques in predicting diabetes, with a primary focus on achieving high accuracy. The central research question addresses the performance assessment of various predictive models in the context of diabetes prediction using AI methodologies.</p> Methods <p>To accomplish this, the study employs three distinct predictive models: Artificial Neural Networks (ANN), Support Vector Machines (SVM), and a hybrid optimization model that combines Artificial Bee Colony (ABC) optimization with SVM. These models were selected for their diverse strengths in handling complex data and pattern recognition, with the hybrid ABC-SVM model standing out as an innovative approach that combines optimization with predictive capabilities.</p> Results <p>The results of the evaluation, measured through various performance metrics such as accuracy, sensitivity, specificity, precision, the Matthews correlation coefficient (MCC), and the area under the receiver operating characteristic curve (AUC-ROC) reveal that the hybrid ABC-SVM model outperforms both the ABC and SVM models in predicting diabetes. Remarkably, the hybrid model achieves an exceptional accuracy rate of 92.47%, indicating its potential as a valuable tool for early detection.</p> Conclusion <p>This study highlights the superiority of the hybrid optimization model ABC-SVM in predicting diabetes compared to conventional models. The high accuracy rate achieved by the hybrid model emphasizes its potential contribution to early diabetes identification, carrying significant implications for the healthcare field and underscoring the importance of timely diagnosis in improving patient outcomes.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of diabetes using hybrid support vector machines with Artificial Bee Colony

  • 期刊论文

  • 发表日期：

    2025年4月7日

查看全文
  • Samira Tared,
  • Mohamed Roubehie Fissa,
  • Latifa Khaouane,
  • Salah Hanini

摘要

Objective

This study aims to investigate the efficacy of artificial intelligence (AI) techniques in predicting diabetes, with a primary focus on achieving high accuracy. The central research question addresses the performance assessment of various predictive models in the context of diabetes prediction using AI methodologies.

Methods

To accomplish this, the study employs three distinct predictive models: Artificial Neural Networks (ANN), Support Vector Machines (SVM), and a hybrid optimization model that combines Artificial Bee Colony (ABC) optimization with SVM. These models were selected for their diverse strengths in handling complex data and pattern recognition, with the hybrid ABC-SVM model standing out as an innovative approach that combines optimization with predictive capabilities.

Results

The results of the evaluation, measured through various performance metrics such as accuracy, sensitivity, specificity, precision, the Matthews correlation coefficient (MCC), and the area under the receiver operating characteristic curve (AUC-ROC) reveal that the hybrid ABC-SVM model outperforms both the ABC and SVM models in predicting diabetes. Remarkably, the hybrid model achieves an exceptional accuracy rate of 92.47%, indicating its potential as a valuable tool for early detection.

Conclusion

This study highlights the superiority of the hybrid optimization model ABC-SVM in predicting diabetes compared to conventional models. The high accuracy rate achieved by the hybrid model emphasizes its potential contribution to early diabetes identification, carrying significant implications for the healthcare field and underscoring the importance of timely diagnosis in improving patient outcomes.