Detection of Masquerade Attack by Data Driven Semi Global Alignment Approach

Masquerade attackers behave like a authorized user to utilize user requirements. The semi-global alignment algorithm (SGA) is one of the most optimize and unique techniques to find out these attack but it has not extend the correctness and executions required by large scope, multiuser systems. To increase all the accuracy and the execution of this algorithm, we recommend the Data-Driven Semi-Global Alignment, DDSGA approach. For security purpose, DDSGA improve the scoring systems by altering various alignment arguments for each user. like wise, it allow small replacement in user command series by assinging small suitable different in the low-level showing of the command to ability to perform a task . It seems to make appropriate changes in the client using technique by updating the pattern of the a user as per to its current using technique. To fix the runtime located, DDSGA to make as little the alignment context and parallelizes the search out and to update. After showing the DDSGA phases, we show the experimental outputs. This output is to represent that DDSGA get the high hit ratio of 88.4% with low wrong positive rate. It improves the hit ratio of advanced SGA and minimizes Maxion-Townsend cost. So, DDSGA results in improving all the hit ratio and false positive rates with a capable calculation context.