A Symptom Selection Algorithm Based on Classification Errors
Akhram Kh. NishanovTashkent University of Information Technologies named after Muhammad al-Khwarizmi, Software Engineering Faculty, Tashkent, UzbekistanBakhtiyorjon AkbaralievTashkent University of Information Technologies named after Muhammad al-Khwarizmi, Software Engineering Faculty, Tashkent, UzbekistanG.P. DjurayevTashkent University of Information Technologies named after Muhammad al-Khwarizmi, Scientific and Innovation Center of Information and Communication Technologies, Tashkent, Uzbekistan
2020 International Conference on Information Science and Communications Technologies (ICISCT)conference2020en
ABI
Аннотация
In this paper the issues like preprocessing of ischemic heart disease data and optimization of feature space are discussed and solved. Here an algorithm for selection a set of information features based on classification error are proposed. Using this algorithm was obtain the set of most informative symptoms for three class of ischemic heart disease: “Progressive angina pectoris”, “Acute myocardial infarction” and “Arrhythmic form”.
Перевод пока недоступен