engleski

Application of neural networks in the identification if types of personality

The algorithm Triatlon for neural network cluster analysis was applied in the personality study of 666 male subjects, 18 years old, randomly selected from the population of clinically healthy and literate young males. The sample was described over a set of eleven personality variables selected under extension of cybernetic model of regulatory functions. The essence of the applied clustering algorithm is a taxonomic neural network based on adaptive multilayer perceptron as a core engine working on the basis of starting classification obtained by method of fuzzy clustering of variables, an then of fuzzy clustering of objects described by fuzzy clusters of variables. An excellent classification was obtained: coefficient of efficacy of neural network was 0.994. The identification of types of personality on the basis of all identification structures was very simple due to clear pattern of centroid vectors and pattern and structure of discriminant functions. Type 1 (31.68% of subjects) was defined by very low control of aggressive impulses, weak control of excitation and high frequency of manifest deviant behavior, so it was clear that subjects belonging to this type are those with antisocial personality disorder. Type 2 (32.43% of subjects) was defined by very good control of all neural regulatory systems and acceptable level of activity, therefore, recognised as type with normal level of conative functions. Type 3 (35.89% of subjects) was defined by very low coordination and integration of regulatory functions, including low control of elementary biological systems, low level of activity and unsatifactory control of aggression, and was recognised as the cathegory of neurotics with diffuse neurotic symptoms.

personality, neural networks, cluster analysis

nije evidentirano