Determination of Some Heart Disease Indicators with Multiple Correspondence Analysis

Öz

Various statistical techniques are used by researchers to help diagnose diseases. One of these, the detection of the presence of heart disease, it is important to develop rapid and effective techniques. Multiple Correspondence analysis can also be used to determine variables associated with some diseases. In this study, it is aimed to determine some variables that may cause heart diseases by multiple correspondence analysis. In this study, multiple correspondence analysis was applied to the data set of 303 patients presenting with heart disease. Multiple correspondence analysis is an analysis method that presents the relationships between categorical variables in two-dimensional space. The statistical study was conducted in June-September 2019 in Van. The application material for this study was obtained from the free access data site Kaggle.1,2 This is a retrospective study. In this study; the relationship of the variables between the “presence of Heart Disease” and “some heart disease indicators” were investigated. According to “the transformed correlation coefficients for the presence of heart disease”; The variables associated with the presence of heart disease are “exercise-related angina, gender, heart rate, age, electrocardiography, systolic blood pressure, fasting blood sugar”, respectively. In the study, some variables that may have an impact on heart diseases were determined by multiple correspondence analysis. It is hoped that the development of rapid and effective techniques for the detection of heart diseases will be important in terms of providing new perspectives to statistical decision-making processes.

Anahtar Kelimeler

• Multiple Correspondence Analysis
• Heart Disease
• Inertia
• Variance
• Dimension

References

1. Aha, D. and Kibler, D., (1988). Instance-based Prediction of Heart-disease Presence with the Cleveland Database. The University of California, 3(1):3-2.
2. Kaggle Datasets, (2019). (internet) (Access:10.07.2019). https://www.kaggle.com/ronitf/heart-disease-uci.
3. Detrano, R., Janosi, A., Steinbrunn, W., Pfisterer, M., Schmid, J., Sandhu, S., Guppy, K., Lee, S., and Froelicher, V., (1989). Int. Application of a New Probability Algorithm for the Diagnosis of Coronary Artery Disease. Am. J. Cardiology, 64:304-310.
4. Gennari, J.H., Langley, P., and Fisher, D., (1989). Models of Incremental Concept Formation. Artificial Intelligence, 40: 11-61.
5. Patel, J., Tejalpadhyay, D., and Patel, S., (2015). Heart Disease Prediction Using Machine Learning and Data Mining Technique. Heart Disease, 7(1):129-137.
6. Greenacre, M.J., (1981). Practical Correspondence Analysis. Looking at Multivariate Data, 81-107.
7. Clausen, S.E., (1988). Applied Correspondence Analysis, An Introduction. California, 1. Ed, Sage Publications.
8. Suner, A. ve Çelikoğlu, C.C., (2008). Uygunluk Analizinin Benzer Çok Değişkenli Analiz Yöntemleri Ile Karşılaştırılması. İstatistikçiler Dergisi: İstatistik ve Aktüerya, 1(1):9-15.

9. Onat, A., Şurdumavci, G., Şenocak, M., Örnek, E., Gözükara, Y., Karaaslan, Y. ve Özcan, R., (1991). Türkiye'de Erişkinlerde Kalp Hastalığı Ve Risk Faktörleri Sıklığı Taraması: 3. Kalp Hastalıkları Prevalansı. Türk Kardiyoloji Derneği Arşivi, 19(1):26-33.