抽象的な

FEATURE DETECTION APPROACH FROM VIRUSES THROUGH MINING

Raviraj Choudhary, Ravi Saharan

Anti-virus systems traditionally use signatures to detect malicious executables, but signatures are over fitted features that are of little use in machine learning. Other methods seek to utilize more general features, with some degree of success. Through this paper we present a new approach that conducts an exhaustive feature search on a set of computer viruses. This method detects mnemonics patterns in large amounts of data, and uses these patterns to detect future instances in similar data. We use apriori algorithm for select features to detect malicious executables. Through those features we make a rule set or detection model for trained over a given set of training data.