An Improvement Of Software Vulnerability And Classification Model Using Deep Neural Network Based On Big Data

  • Dr. N. Yuvaraj, Dr.K.Munusamy,Dr. G. Saravanan, Mr.P.Mohanraj, Mr.M.Premkumar

Abstract

Software vulnerabilities are raising the security dangers. In the event that any helplessness is abused because of a Malignant Assault, it will bargain the framework's wellbeing. What's more, it might make cataclysmic misfortunes. Along these lines, programmed characterization techniques are required to oversee defenselessness in programming, and afterward security execution of the framework will be improved. It will likewise moderate the danger of framework being assaulted and harmed. In this venture, another model has been proposed with name programmed powerlessness arrangement model (IGTF-DNN) Information Gain dependent on Term Frequency - Deep Neural Network. The model is produced utilizing data gain (IG) which depends on recurrence backwards record recurrence (TF-IDF), and profound neural system (DNN): TF-IDF is utilized to figure recurrence/weight of words taken from defenselessness portrayal; IG is utilized to choose highlights to accumulate ideal arrangement of highlight words. At that point neural system model is utilized to develop a programmed powerlessness classifier to accomplish viable helplessness characterization. The National Vulnerability Database of the United States has been taken to test this new model's viability. By contrasting and KNN, this TFI-DNN model has accomplished better execution in assessment. The venture is planned utilizing R Studio 1.0 and language R 3.4.4.

Published
2020-06-01
How to Cite
Dr. N. Yuvaraj, Dr.K.Munusamy,Dr. G. Saravanan, Mr.P.Mohanraj, Mr.M.Premkumar. (2020). An Improvement Of Software Vulnerability And Classification Model Using Deep Neural Network Based On Big Data. International Journal of Advanced Science and Technology, 29(7), 5126-5134. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/23590
Section
Articles