Open AccessClassification of Concentration Levels in Adult-Early Phase using Brainwave Signals by Applying K-Nearest Neighbor
Author(s) -
Ahmad Azhari,
Fathia Irbati Ammatulloh
Publication year - 2019
Publication title -
signal and image processing letters
Language(s) - English
Resource type - Journals
eISSN - 2714-6677
pISSN - 2714-6669
DOI - 10.31763/simple.v1i1.170
Subject(s) - electroencephalography , respondent , brain activity and meditation , cognition , psychology , pattern recognition (psychology) , artificial intelligence , computer science , stimulus (psychology) , preprocessor , audiology , developmental psychology , speech recognition , cognitive psychology , neuroscience , medicine , political science , law
The brain controls the center of human life. Through the brain, all activities of living can be done. One of them is cognitive activity. Brain performance is influenced by mental conditions, lifestyle, and age. Cognitive activity is an observation of mental action, so it includes psychological symptoms that involve memory in the brain's memory, information processing, and future planning. In this study, the concentration level was measured at the age of the adult-early phase (18-30 years) because in this phase, the brain thinks more abstractly and mental conditions influence it. The purpose of this study was to see the level of concentration in the adult-early phase with a stimulus in the form of cognitive activity using IQ tests with the type of Standard Progressive Matrices (SPM) tests. To find out the IQ test results require a long time, so in this study, a recording was done to get brain waves so that the results of the concentration level can be obtained quickly.EEG data was taken using an Electroencephalogram (EEG) by applying the SPM test as a stimulus. The acquisition takes three times for each respondent, with a total of 10 respondents. The method implemented in this study is a classification with the k-Nearest Neighbor (kNN) algorithm. Before using this method, preprocessing is done first by reducing the signal and filtering the beta signal (13-30 Hz).The results of the data taken will be extracted first to get the right features, feature extraction in this study using first-order statistical characteristics that aim to find out the typical information from the signals obtained. The results of this study are the classification of concentration levels in the categories of high, medium, and low. Finally, the results of this study show an accuracy rate of 70%.