The　detection　of　unknown　malicious　executables　is　beyond　the　capability　of　many　existing　detection　approaches.　Machine　learning　or　data　mining　method　can　identify　new　or　unknown　malicious　executables　with　some　degree　of　success.　Bayes　or　improved　Bayes　algorithm　has　the　detection　capability　of　unknown　malicious　executables;　however,　it　takes　more　time　to　study.　A　new　improved　algorithm　is　proposed　in　this　paper.　The　new　classifier　based　on　strings　achieve　has　high　detection　rates　and　can　be　expected　to　perform　as　well　in　real-world　conditions.