Software Behavior Trustworthiness Detection Based on PrefixSpan Method

Current study of Trusted Computing is mainly about TCP and it is not effective to check trustworthiness of software behavior. With analysis of current detection technique, a static detection method towards software behavior trustworthiness is proposed and its key technique is discussed. First, Malware code is disassembled and preprocessed into sequential data, second, a Prefix Span-based algorithm is used to discover frequent sequential pattern and remove normal pattern, at last, the result pattern set can be used to detect target software and give results about its behavior trustworthiness. Test result indicates that the method has high accuracy rate and low false positive rate.