In risky pregnancy, various diseases such as heart, lung, kidney, high blood pressure, diabetes and liver that pregnant women have before may aggravate the expectant mother's condition during pregnancy. By analyzing medical parameters such as maternal age, heart rate, blood oxygen level, blood pressure, body temperature, and the values corresponding to these parameters, information on risk intensity can be estimated for some patients. It is possible to reduce such pregnancy-related complications by classifying risk factors early in symptoms. It is possible to benefit from machine learning methods in determining maternal risk health. Therefore, in this study, six different machine learning methods were used to determine maternal risk health. The results obtained in these methods were compared with each other and it was observed that the most successful method in estimating maternal risk health was Decision Tree. The accuracy value obtained in the Decision Tree method was 89.16%. The lowest accuracy rate among the methods used in the paper was obtained in the k-nearest neighbors (KNN) method with 68.47%.
Artificial Intelligence Classifiers KNN Machine Learning Maternal Health
Birincil Dil | İngilizce |
---|---|
Konular | Yazılım Mühendisliği (Diğer) |
Bölüm | Research Articles |
Yazarlar | |
Yayımlanma Tarihi | 30 Haziran 2023 |
Gönderilme Tarihi | 5 Mayıs 2023 |
Kabul Tarihi | 1 Haziran 2023 |
Yayımlandığı Sayı | Yıl 2023 |