Study of Artificial Neural Networks in Information Security Risk Assessment

Abstract

Objective: The principle of a more comprehensive information capital and organization in the information age and maintain it is very important in this research We decided to use the themes of artificialintelligence algorithms to identify information security risks and impact of Planning for security Reduce risk. Methodology: For this purpose we examined the IT organization referred to security measures based onthe Chkl sets the standard view and review the information we collect to attack the sources of information were both Clustering and classification of the information packet and a peak of the nightsaw the need of each one have for the probability of error And To Won According to the results of security measures and reduce the likelihood of attacks was effective in reducing dangerous attacks of the clustering of similar results, and they used the systemshows the data integrity of the system may be Drnha the data on the different organizational and (c) of Adhsaz the impact of security measures to assess. Results: The results of this research can help managers in assessing and rating the risk to information security in the organization of the system designed in this study may be Impact of security measures on the reduction of information security risksshow. Conclusion: after presentation of the organization and review the data and information available and the resources to implement the research model of this study in order to improve results and compliance with the the coin real information below to copy the proposed change found Login to have consisted of real attacks Statistics information to the relevant organization in order to carry out the investigation after the observation and study of the statistics Vvyzhgy attacks the nervous system sorted them by teaching the necessary research to cthe night of the the security risk information by using the algorithm of the themes of artificial intelligence 68 M to be in this way given the number 87789 in the neural network C of the conjugate graded 69 to 10 layers, and 5 percent were trained in control and test data.