APT malicious software classification method and device based on prediction guidance prototype

The invention discloses an APT malicious software classification method and device based on a prediction guidance prototype. The method comprises the following steps: acquiring a global feature code and a numeralization code sequence of an API calling sequence of malicious software; after feature ex...

Full description

Saved in:
Bibliographic Details
Main Authors LIU FENG, WANG WEN, BAO HUAIFENG, TANG ZIXIAN
Format Patent
LanguageChinese
English
Published 23.06.2023
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:The invention discloses an APT malicious software classification method and device based on a prediction guidance prototype. The method comprises the following steps: acquiring a global feature code and a numeralization code sequence of an API calling sequence of malicious software; after feature extraction of the auxiliary function classification task and the main organization classification task is carried out on the numerical coding sequence, feature vector representations of the auxiliary function classification task and the main organization classification task are obtained in combination with global feature coding; using the auxiliary function classification task feature vector to represent the function category Gaussian distribution in the fitting feature space to obtain the prediction probability on the function category; in combination with the prediction probability on the function category, enabling the feature vector of the main organization classification task to represent the organization catego
Bibliography:Application Number: CN202310165123