Efficient Prediction of Stroke Patients
Using Random Forest Algorithm in
Comparison to Support Vector Machine
Ritaban Mitra
a
and T. Rajendran
b,1
a
Research Scholar, Dept. Of CSE, Saveetha School of Engineering,
b
Asso. Prof., Dept. Of CSE, Saveetha School of Engineering,
a,b
SIMATS, Chennai.Tamil Nadu, India
Abstract. The work aims to make an efficient prediction of stroke in patients using
several Machine learning modeling techniques and evaluating their performance.
The two groups used in this paper are the Random Forest Algorithm (RFA) and the
Support Vector Machine(SVM) Algorithm. The dataset implemented and tested
consists of over 5000 records of patients' medical and personal records. They were
using N = 20 iterations for each algorithm. The G-Power test used is about 80%.
The results of our work have given us the mean accuracy of 94.61 on Random
Forest and 93.91 on Support Vector Machine Algorithms. The statistically
significant difference was obtained by generating independent sample t-tests at
0.015. This work is intended to implement innovative approaches to increase the
efficiency of stroke prediction algorithms and improve the accuracy of existing
algorithms. The results show that the Random Forest Model performs higher than
Support Vector Machines.
Keywords. Innovative Stroke Prediction, Machine learning, Data Science,
Random Forest Algorithm, Support Vector Machine Algorithm, Statistical
Analysis.
1. Introduction
Stroke is the second biggest reason of mortality globally, as per the WHO report,
accounting for 11% of fatalities yearly. A stroke is a medical emergency that causes
damage to the brain due to a shortage of blood supply, causing brain cells to die. This
research paper will explore stroke conditions and use a Machine learning approach to
solve this problem and develop an Innovative Stroke Prediction technique in
patients [1]. Over the years, as computers have become more powerful, their ability to
support research work in the medical domain has also increased. This is a massive
benefit to the world as it can combine the power of human intelligence with the
potential of computers and gain insights into patterns from Statistical Analysis [2]. This
analysis is done using a Data Science driven approach [3]. Applications of the research
include clinical prognosis and drug development [2]. The Prediction and classification
of heart failure have been made using a Machine learning approach [4]. In paper [5], a
1
T. Rajendran
,
Department of Computer Science and Engineering Saveetha School of Engineering
Saveetha Institute of Medical and Technical Sciences, Chennai, Tamil Nadu, India,
Email:rajendrant.sse@saveetha.com.
Advances in Parallel Computing Algorithms, Tools and Paradigms
D.J. Hemanth et al. (Eds.)
© 2022 The authors and IOS Press.
This article is published online with Open Access by IOS Press and distributed under the terms
of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
doi:10.3233/APC220075
530