Open AccessPredictive Model for Heart Disease Diagnosis Based on Multinomial Logistic Regression
Author(s) -
Munandar Tb Ai,
Sumiati Sumiati,
Vidila Rosalina
Publication year - 2021
Publication title -
informacinės technologijos ir valdymas
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.286
H-Index - 19
eISSN - 2335-884X
pISSN - 1392-124X
DOI - 10.5755/j01.itc.50.2.27672
Subject(s) - categorical variable , multinomial logistic regression , logistic regression , computer science , regression analysis , statistics , predictive modelling , regression , heart disease , data mining , artificial intelligence , machine learning , medicine , mathematics
Many computational approaches are used to assist the analysis of influencing factors, as well as for the need forprediction and even classification of certain types of disease. In the case of disease classification, the data usedare often categorical data, both for dependent variables and for independent variables, which are the results ofconversion from numeric data. In other words, the data used are already unnatural. Conversion processes oftendo not have standard rules, thus affecting the accuracy of the classification results. This research was conductedto form a predictive model for heart disease diagnosis based on the natural data from the patients' medicalrecords, using the multinomial logistic regression approach. The medical record data were taken based on thepatients’ electrocardiogram information whose data had been cleansed first. Other models were also tested tosee the accuracy of the heart disease diagnosis against the same data. The results showed that multinomiallogistic regression had the highest level of accuracy compared to other computational techniques, amountingto 75.60%. The highest level of accuracy is obtained by involving all variables (based on the results of the firstexperiment). This research also produced seven regression equations to predict the heart disease diagnosisbased on the patients’ electrocardiogram data.