A comprehensive study of machine learning for predicting cardiovascular disease using Weka and Statistical Package for Social Sciences tools

Belal Abuhaija, Aladeen Alloubani, Mohammad Almatari, Ghaith Jaradat, Hemn Abdallah, Abdallah M. Abualkishik, Mutasem K. Alsmadi

Abstract


Artificial intelligence (AI) is simulating human intelligence processes by machines and software simulators to help humans in making accurate, informed, and fast decisions based on data analysis. The medical field can make use of such AI simulators because medical data records are enormous with many overlapping parameters. Using in-depth classification techniques and data analysis can be the first step in identifying and reducing the risk factors. In this research, we are evaluating a dataset of cardiovascular abnormalities affecting a group of potential patients. We aim to employ the help of AI simulators such as Weka to understand the effect of each parameter on the risk of suffering from cardiovascular disease (CVD). We are utilizing seven classes, such as baseline accuracy, naïve Bayes, k-nearest neighbor, decision tree, support vector machine, linear regression, and artificial neural network multilayer perceptron. The classifiers are assisted by a correlation-based filter to select the most influential attributes that may have an impact on obtaining a higher classification accuracy. Analysis of the results based on sensitivity, specificity, accuracy, and precision results from Weka and Statistical Package for Social Sciences (SPSS) is illustrated. A decision tree method (J48) demonstrated its ability to classify CVD cases with high accuracy 95.76%.


Keywords


attribute selection; cardiovascular diseases; classification; machine learning; statistical package for social sciences; Weka;

Full Text:

PDF


DOI: http://doi.org/10.11591/ijece.v13i2.pp1891-1902

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

International Journal of Electrical and Computer Engineering (IJECE)
p-ISSN 2088-8708, e-ISSN 2722-2578