Corporate System Users Identification by the Keyboard Handwriting based on Neural Networks

Abstract

The paper is devoted to practical information security aspects in the corporate system. Particular attention is paid for the ensuring access control problem for corporate confidential information. Solving this problem involves identifying a person trying to gain unauthorized access to the corporate system and identification of a authorized user committing illegal actions. The paper shows that a person's keyboard handwriting is determined by behavioural characteristic, which is very difficult to imitate. This means that the person identification based on the keyboard handwriting analysis is the most reliable. It is necessary to consider the possibility of accidental uncontrolled change user’s keyboard handwriting settings when it works. Such changes can be caused by changes in physiological or emotional state. Using neural networks enables to perform analysis of handwriting keyboard considering these changes at the real time. Thus, the artificial neural networks using makes it possible considerably improves the security of corporate confidential information protection from authorized user committing illegal actions. Existing methods and algorithms for user identifying by his keyboard handwriting are focused on using a multilayer perceptron. However, FAM architecture and capabilities analysis of adaptive resonance theory has identified advantages such as the ability to form associative pairs and map the rules of fuzzy inferences to clusters of neural networks in this category. Considering these characteristics was decisive in choosing the neural network intelligent system for users identification by his keyboard handwriting style. As a result of the research proposed generalized architecture Corporate System Users Identification by the Keyboard Handwriting, based on the use of Cascade ARTMAP. The scheme of rules formation in Cascade ARTMAP and the knowledge base formation scheme of Users Identification Intelligent System by Keyboard Handwriting are proposed.

The Use of Neuro-Fuzzy Models in Expert Support Systems for Forensic Building-Technical Expertise

“…The knowledge base of a neuro-fuzzy system is implicitly set by the ANN structure. At the same time [12,14]: -input nodes perform the function of phasing; -hidden nodes reflect the rules; -the output node performs the function of defuzzification. To preserve the fuzzy rule, reflects the heuristic of associative thinking of an expert, the neuro-fuzzy network Kosko Fuzzy Associate Memory (FAM) can be used.…”

“…In [14], a scheme for acquiring knowledge by reflecting a priori expert knowledge on an integrated Cascade ARTMAP memory card is shown.…”