Gayatri Ketepalli

The security of computer networks is critical for network intrusion detection systems (NIDS). However, concerns exist about the suitability and sustainable development of current approaches in light of modern networks. Such concerns are particularly related to increasing levels of human interaction required and decreased detection accuracy. These concerns are also highlighted. This post presents a modern intrusion prevention deep learning methodology. For unattended function instruction, we clarify our proposed Symmetric Deep Autoencoder (SDAE). Also, we are proposing our latest deep research classification model developed with stacked SDAEs. The classification proposed by the Network Security Laboratory-Knowledge Discovery in Databases (NSL-KDD) and Canadian Institute for Cybersecurity -Intrusion Detection System (CICIDS 2017) data sets was implemented in Tensor Flow, a Graphics Procedure Unit (GPU) enabled and evaluated. We implemented and tested our experiment with different batch sizes using Adam optimizer. Promising findings from our model have been achieved so far, which demonstrates improvements over current solutions and the subsequent improvement for use in advanced NIDS.

Cloud computing empowers the clients to get outsourced information from cloud storage with no equipment and programming administrations. For compelling the usage of secret information from Cloud Service Provider (CSP), the information owner encodes before storing in the cloud. To secure information in cloud, information protection is a testing assignment. In order to deal this issue, a proficient information security strategy utilizing cryptographic procedures and Pk-Anonymization method is introduced. Accordingly, the proposed technique scrambles the sensitive information, as well as distinguishes the exploitative gathering to get to the information utilizing consolidated hash capacities. Anonymization is fundamental concept for bringing about assurance on users secret information. Protection of Data is likely necessary because of the control lessening system and PK-Anonymization method. Pk-Anonymization and cryptography in Cloud computing to improve data security is introduced in the proposed method.