Linguistic Steganalysis Based on Meta Featuresand Immune Mechanism

Linguistic steganalysis depends on efficientdetection features due to the diversity of syntax andthe polysemia of semantics in natural language processing.This paper presents a novel linguistics steganalysisapproach based on meta features and immune clone mechanism.Firstly, meta features are used to represent texts.Then immune clone mechanism is exploited to select appropriatefeatures so as to constitute effective detectors.Our approach employed meta features as detection features,which is an opposite view from the previous literatures.Moreover, the immune training process consistsof two phases which can identify respectively two kindsof stego texts. The constituted detectors have the capableof blind steganalysis to a certain extent. Experimentsshow that the proposed approach gets better performancethan typical existing methods, especially in detecting shorttexts. When sizes of texts are confined to 3kB, detectionaccuracies have exceeded 95%.