A method for discovering spammers in social networks

Abstract

The invention provides a method for discovering junk mail senders in a social network, which comprises the following steps of: step 1, performing data preprocessing, and processing mail data in a format required by a junk mail sender discovering model in the invention; step 2, performing user behavior feature extraction, wherein the extraction mode is to extract user behavior features by using a semi-supervised training graph convolutional network; step 3, performing user classification: judging whether each user node is a junk mail sender or not according to the user behavior attribute outputby a user behavior feature extractor; step 4, constructing a decision objective function, and guiding an optimization direction of the junk mail sender discovery model, so that the user behavior feature extractor and a user classifier in the junk mail sender discovery model can jointly learn parameters in the graph convolutional network and optimize a classification plane of the single-class support vector machine; and step 5, training the junk mail sender discovery model in the social network.

Description

technical field [0001] The invention designs an intelligent detection method, and mainly relates to a method for discovering spam senders in a social network. Background technique [0002] Since the 21st century, with the vigorous development of information technology, the use of the Internet has spread rapidly on a large scale, and more and more people use e-mail to communicate in life, work or business. E-mail has greatly facilitated the communication between people. communication efficiency. However, e-mail, as an important channel for network communication, not only facilitates normal users, but also facilitates those spammers. Especially in recent years, spam has troubled most people. According to the survey data, 93% of the respondents are very dissatisfied with the large amount of spam they receive. [0003] With the continuous development of the Internet, spam mails are increasing rapidly, and now people receive overwhelming spam mails every day. Initially, spam w...

AI Technical Summary

Problems solved by technology

However, these existing methods tend to isolate normal mail exchanges during use, or cannot effectively prevent spam

Images