Artificial Immune System-Negative Selection algorithm | SciTechnol

Journal of Computer Engineering & Information Technology.ISSN : 2324-9307

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Commentary,  J Comput Eng Inf Technol Vol: 10 Issue: 3

Artificial Immune System-Negative Selection algorithm

Ramineni Shekar

Department of Computer Science,Osmania University, Hyderabad, India. 

*Corresponding Author:
Ramineni Shekar
Department of Computer Science,Osmania University, Hyderabad, India. 
Tel: +919955616362, E.Mail: Shekarramineni91@

Received: March 3rd, 2021 Accepted: March 17th, 2021 Published: March 24th, 2021

Citation: Ramineni Shekar, (2021) Artificial Immune SystemNegative Selection algorithm J Comput Eng Inf Technol 10:3 (257)


The negative choice formula (NSA) is one in all the essential algorithms of the bogus system. The formula divides the method of detector generation into 3 steps: the primary step is to calculate the density of the substances by victimization the tactic of antigen density clump to pick none self-clusters. Artificial systems (AIS) square measure intelligent algorithms derived from the principles galvanized by the human immune system.Electroencephalography (EEG) signals for four distinct motor movements of human limbs square measure detected and classified employing a negative choice classification formula (NSCA).

Keywords: Artificial Immune System; Negative Selection algorithm; Electroencephalography


For this study, wide studied open supply graph signal information (BCI IV–Graz dataset 2a, comprising 9 subjects) has been used. Mel frequency cepstral coefficients (MFCCs) square measure extracted as chosen options from recorded graph signals. spatiality reduction of information is applied by applying 2 hidden stratified stacked auto-encoder. Genetic formula (GA) optimized detectors (artificial lymphocytes) square measure trained victimization negative choice formula (NSA) for detection and classification of 4 motor movements. The trained detectors carries with it four sets of detectors; every set is trained for detection and classification of 1 of the four movements from the opposite 3 movements. The optimized radius of detector is little enough to not mis-detect the sample. Geometrician distance each detector with every coaching dataset sample is taken and compared with the optimized radius of the detector as a oneself detector. Our projected approach achieved a mean classification accuracy of eighty six.39% for limb movements over 9 subjects with a most individual subject classification accuracy of ninety seven.5% for subject variety eight.
The exaggerated nature of email spam with the employment of urge mailing tools prompt the necessity for detector generation to counter the menace of uninvited email. Detector generation galvanized by the human system implements particle swarm improvement (PSO) to get detector in negative choice formula (NSA). Outlier detectors square measure distinctive options generated by native outlier issue (LOF). The native outlier issue is enforced as fitness operate to see the native best (Pbest) of every candidate detector. rate and position of particle swarm improvement is used to support the movement and new particle position of every outlier detector. The particle swarm improvement (PSO) is enforced to enhance detector generation in negative choice formula instead of the random generation of detectors. The model is termed swarm negative choice formula (SNSA). The experimental result show that the projected SNSA model performs higher than the quality United States intelligence agency. A brain computer interface (BCI) provides a communication channel between the human brain and external devices. This interface is generally composed of a signal processing device, a set of non-invasive sensors and an external device. The electroencephalogram (EEG) based BCI system is generally composed of three processes namely, signal preprocessing, feature extraction, and classification. The signal preprocessing step is performed to enhance the acquired signal by removing baseline noise and selecting a band frequency of interest. The feature extraction step is mainly employed to extract meaningful information in the form of features from raw EEG signals. The last step of classification is performed to encode the specific features into meaningful information (determining the class) in order to control the external devices. Since the BCI system is independent of muscles and peripheral nerves, it is mainly beneficial for patients who suffer from motor disorders with cognitive disabilities

Track Your Manuscript