Paper Title
Pain Classification Using Weighted K -Nearest Neighbour

Abstract
Automated pain evaluation is essential for anyone working in medicine. Instead as it depends so much on the patient's response, physical pain assessment today is imprecise. The aim of the pain identification method we provide in this work is based on biological signals, and it may be used to extract new characteristics from data on electromyography (EMG), electrodermal activity (EDA), and electrocardiogram (ECG) that have not been used for pain identification before. The recommended weighted k -nearest neighbour based method executes much better than the previous methods shown in the research for both the electrodermal activity (EDA), with average performances of 82.16% for the binary classification. The experiment that discriminates within the baseline and the pain tolerance level (T0 vs. T4). Assessment and validation in clinical settings are conducted using the Bio-Vid Heat Pain Database (Part A). Keywords - Pain, Feature, Intensity, Classification.