Internet malicious entity detection method and storage medium

• Main Authors: ZHANG KUNSAN, LIN XIAOHUANG, ZOU WEIFU, LIN NAN, LIN ZHIHANG, CHEN GENGEN, LIAO SHENGYANG, LIAN YUHAN, CHENG BINGJIE, PENG WEIWEN
• Format: Patent
• Language: Chinese; English
• Published: 05.04.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRIC DIGITAL DATA PROCESSING; ELECTRICITY; PHYSICS; TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHICCOMMUNICATION

The invention discloses a method for detecting internet malicious entities, which comprises the following steps of: performing feature extraction on different malicious entities to obtain feature data in a word stream form; for different malicious entities, according to the number of the feature data extracted by the malicious entities, feature aggregation is carried out in combination with an N-Gram method, feature vectors with consistent dimensions are generated, and a feature data set is obtained; training a CNN model through the feature data set to obtain a malicious entity detection model, and performing malicious entity detection through the malicious entity detection model; feature extraction is carried out on different malicious entities, feature aggregation is carried out on the basis of the number of extracted feature data in combination with an N-Gram method, feature vector representation forms of the different malicious entities are consistent, a CNN model is trained through training data of the d