Application Of Various Data Mining Techniques To Classify Heart Diseases

Ghosh, Saswata (2017) Application Of Various Data Mining Techniques To Classify Heart Diseases. Masters thesis, Dublin, National College of Ireland.

Preview

PDF (Master of Science) Download (1MB) | Preview

Abstract

Diseases are quite common in human beings but the most common disease that affects human beings are heart diseases which are the main cause of death in developed and developing countries. Detecting heart disease is a great challenge for medical practitioners with high level of accuracy so that a patients life is saved. Large amount of medical data are being generated which can be useful for the physicians, various data mining algorithm helps us to find meaningful insights from data which would otherwise remain undiscovered. My research aims to deploy a model that gives us the highest performance metric (accuracy, sensitivity,specificity and kappa) among the three algorithm Random Forest, Logistic regression and Artificial Neural Network so that a patient life is saved in places where there is lack of medical practitioners. I would be following the CRISPDM data mining methodology. Supervised classification algorithm is used to classify the heart disease in the UCI-ML data set which consists of 14 attribute and 303 observations. The effect of detecting heart disease is to render in the different data mining algorithm in terms of accuracy, sensitivity, specificity and kappa with the sole motive of obtaining more superior results than any other research method. The accuracy, sensitivity, specificity and kappa of ANN is 79%, 85%, 73% and 59%, whereas that of Random forest is 80%, 83%, 78% and 61% percent and Logistic regression is 85%, 89%, 80% and 70% percent,which clearly shows that Logistic regression outperforms the other two algorithm in terms of all the performance measures. Thus this research will enable the medical industry as a whole in timely treatment of heart disease without the intervention of medical practitioners and also this will help the general practitioners to decide the next step for treatment without the intervention of trained cardiologists in remote area.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science R Medicine > Healthcare Industry
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Caoimhe Ní Mhaicín
Date Deposited:	28 Aug 2018 10:51
Last Modified:	28 Aug 2018 10:51
URI:	https://norma.ncirl.ie/id/eprint/3082

Actions (login required)

View Item