نوع مقاله : مقاله علمی
نویسندگان
1 کارشناس ارشد مهندسی نرم افزار علوم و تحقیقات لرستان
2 کارشناس ارشد مدیریت فناوری اطلاعات دانشگاه علوم و تحقیقات تهران
3 کارشناس ارشد اطلاعات و حفاظت اطلاعات و عضو هیئت علمی دانشکده علوم و فنون فارابی
4 مهندسی کامپیوتر علوم و تحقیقات لرستان
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Due to the growth of computer networks, network security has also been raised as a major challenge. Intrusion detection systems have been developed to ensure secure data processing and storage on the network; they are also considered as major components of network security. Because traditional intrusion detection systems do not respond to new attacks, therefore, data-mining penetration detection systems are suggested in this study.Data mining techniques is Used to increase the accuracy of intrusion detection systems, and it is typically increases the network security.Due to the wide range of data mining techniques, self-organized neural networks and backup machines have been used to detect and predict the influence of this study;Therefore, the combination of the two above mentioned technologies ,to some extent ,improves the anomaly detection. However, using self-organized neural networks and backup carriers, normal behavior and network traffic can be categorized in one category, and heterogeneous attacks or heterogeneous neurons are classified in the other categories.The KDD CUP 99 dataset is used to train and evaluate the proposed method. In other words, because of the use of competitive learning for training in this study this method was used;Therefore, compares the accuracy of the proposed method with the self-organized neural network learning and decision making tree learning.The results show that the proposed system has a high accuracy for intrusion detection, and also less time is needed in comparison with the other techniques .
کلیدواژهها [English]
)