Proceedings of the 1st International Conference on Artificial Intelligence, Communication, IoT, Data Engineering and Security, IACIDS 2023, 23-25 November 2023, Lavasa, Pune, India

Research Article

Improving the Prediction Accuracy of Onset of Cardiovascular Diseases, using Ensemble Learning

Download49 downloads
  • @INPROCEEDINGS{10.4108/eai.23-11-2023.2343340,
        author={Muralidharan  Jayaraman and Shanmugavadivu  Pichai},
        title={Improving the Prediction Accuracy of Onset of Cardiovascular Diseases, using Ensemble Learning},
        proceedings={Proceedings of the 1st International Conference on Artificial Intelligence, Communication, IoT, Data Engineering and Security, IACIDS 2023, 23-25 November 2023, Lavasa, Pune, India},
        publisher={EAI},
        proceedings_a={IACIDS},
        year={2024},
        month={3},
        keywords={ensemble learning cardiovascular diseases prediction},
        doi={10.4108/eai.23-11-2023.2343340}
    }
    
  • Muralidharan Jayaraman
    Shanmugavadivu Pichai
    Year: 2024
    Improving the Prediction Accuracy of Onset of Cardiovascular Diseases, using Ensemble Learning
    IACIDS
    EAI
    DOI: 10.4108/eai.23-11-2023.2343340
Muralidharan Jayaraman1,*, Shanmugavadivu Pichai1
  • 1: The Gandhigram Rural Institute – Deemeed to be University, Tamil Nadu, India
*Contact email: jaymurali@gmail.com

Abstract

This work presents an approach for classifying cardiac and non-cardiac data extracted from a dataset comprising of 70,000 records. The methodology begins with preprocessing, eliminating noisy and inconsistent data points using the box plot-based outlier removal technique. Subsequently, training and testing sets are taken out of the cleansed dataset, for model evaluation using a variety of base classifiers, such as support vector machines, decision trees, and random forests, within the ensemble framework. The experimental result of proposed method reveals the accuracy of the ensemble classifier model in classifying cardiac and non-cardiac data with an accuracy of 88.39%, with a focus on minimizing both false positives and false negatives