Longitudinal federated learning-based social network cross-platform malicious user detection method

Abstract

The invention discloses a social network cross-platform malicious user detection method based on longitudinal federated learning. The method comprises the following steps: step 1, constructing a social network cross-platform malicious user detection hierarchical architecture based on longitudinal federated learning; 2, dividing participants into active parties and passive parties, and performing preprocessing operation on sample data of the active parties and the passive parties in a data preprocessing layer to obtain structured data; step 3, mapping common sample data of the active party and the passive party by the structured data processed by the data preprocessing layer; 4, cooperatively training a global model under the definition of machine learning, and encrypting and decrypting data of an active party and a passive party by using homomorphic encryption to complete federal learning layer training; step 5, enabling the active party and the passive party to update own local model training parameters and output prediction results; and step 6, transmitting a prediction result obtained by the federal learning layer back to each participant in the data application layer, and realizing a high-quality malicious user detection effect.

Description

technical field [0001] The invention belongs to the technical field of the Internet, in particular to a detection method for cross-platform malicious users of a social network based on longitudinal federated learning. Background technique [0002] With the rapid development of Online Social Networks (OSNs), as of March 2020, the 45th "China Internet Development Statistical Report" shows that the number of OSNs users has reached 904 million, and the Internet penetration rate has reached 64.5%. Therefore, , while OSNs help people build social network application services, they have gradually become the primary target of malicious users trying to perform illegal activities and malicious harm. These malicious behaviors have caused adverse effects and huge harm to today's society. [0003] At present, traditional machine learning methods, such as semi-supervised clustering and support vector machine classifiers, rely on big data to extract and train malicious user behavior charac...

AI Technical Summary

Problems solved by technology

[0006] However, the successful applications of the above solutions are all based on social big data. In actual application scenarios, malicious users have the characteristics of dispersion, latency, and complexity. It is difficult for unilateral data to meet the detection requirements. The data of two parties or even multiple parties are jointly trained to achieve satisfactory detection results; secondly, with the improvement of laws and regulations, it has become a worldwide recognized trend to pay attention to user privacy and data security, such as the "General Data Protection Regulation" promulgated by the European Union. "(General Data Protection Regulation, GDPR) stipulates that without the consent of the user, it has been expressly prohibited to collect user data from all parties in one place.

Images