Bogdanov M.R.1,2, Dumchikov A.A.3, Akhmerova A.Z3, Nasyrov D.I1
1. Ufa State Aviation Technical University, Ufa City, Russia.
2. M.Akmullah named after Bashkir State Pedagogical University, Ufa City, Russia.
3. B.N.Elcin Ural Federal University, Ekaterinburg, Russia.
Abstract: Under the research, the effectiveness of 14 machine learning algorithms for the diagnosis of cardiovascular diseases was evaluated. A PTB database of digitized electrocardiograms was used. It was found that the most preferred are a Label Propagation classifier (accuracy of recognition is 0.94), An extremely randomized tree classifier (accuracy is 0.92) and a Classifier implementing the k-nearest neighbors vote (accuracy is 0.90).
Keywords: Heart diseases, Diagnosis, Machine Learning, Electrocardiogram.
Pages: 81 – 84 | Full PDF Paper