Affinity artificial immune based feature selection for colour visualization tasks using ensemble emphasis boosting classifier

Abstract

Certain people are unable to express their views due to severe newlineimpairments brought on by voice or muscle problems. This will enable the newlineimpaired person to communicate their opinions through brain signals. The newlineBrain-Machine Interface (BMI) is a device that extracts commands from newlinehuman brain signals and transmits them to the outside world. EEG is the newlinedevice that is used to record the multi-frequency, non-stationary brain signal newlinefrom the various localized channels. By creating a conduit between the brain newlineand external equipment, the BMI operates for those with brain illnesses taking newlineinto account the EEG signals produced when viewing different colors. With newlinethe help of color-visual activities, the BMI-based EEG can meet the needs of newlinepeople with physical disabilities by offering brain-activated control signals for newlinesystem communication. As a result, people with physical limitations can newlinecommunicate themselves through color using their visual signals. It is crucial newlineto choose the relevant features to carry out further categorization for the visual newlinestimulation of distinct colors. The features reduce the risk of losing crucial newlinesignal information. newlineIn EEG signal analysis, the process of obtaining the crucial signal newlineinformation, or features, from the signals is referred to as feature extraction or newlinefeature selection. Before the EEG signal analysis, the data needs to pass newlinethrough a series of signal processing steps known as preprocessing because the newlineraw EEG signal contains noise. With this intent, the proposed research work is newlineimplemented to perform effective classification for visual stimulation of newlinedistinct colors from the EEG signals through a series of signal processing newlinesteps, namely pre-processing, feature extraction, feature selection, and newlineclassification using essential techniques. newline