Malware in Internet A study on Botnets

Abstract

In the present information age of internet, botnets are becoming a challenge for the community of internet. The internet users are troubled because of these malicious activities which cause a lot of problem for the users, administrators and the internet service providers. The present research work aims to investigate the presence of BOT in the system, different types of Botnets and their harmful effects suggest the suitable technique for safeguard against particular type of bot attack. By analyzing and comparing known malware and normal processes, the study exploited differences in their network activity behavior and produced accurate and effective malware detection with minimal false positives and false negatives. Result also indicates significant improvement in detecting Malware as after observation, O1, occurred the most in the malware samples with 49% followed by O2 with 21% and O4 with 18%.The study will contribute significantly to the future research as the researchers can use the current safeguard techniques and novel designing of the app for detection of the presence of botnets and bot both together. results successfully classified a diverse group of malware and Normal process with very high accuracy and minimal false positives and false negatives. Classification algorithms correctly detected newly introduced malware samples also with minimal false negatives and false positives. Most interestingly, our data set included 31 malware samples are not detected by any tools. These undetected malware were correctly identified using our analysis in classification algorithms with few exceptions. This provides strong evidence that our identified behaviors can be added to existing behavior-based bots and malware detection solutions to help stop zero-day attacks on a host machine newline