Social robot detection method based on variational self-coding and K neighbor combination

Abstract

A social robot detection method based on variational self-encoding and K-nearest neighbor combination belongs to the technical field of anomaly detection, and comprises the following steps: acquiring public data of a social robot through a network, extracting features through preprocessing, training by adopting the data, and encoding and decoding by using variational self-encoding. Normal sample features are more similar to initial features after being decoded, abnormal samples are greatly different from the initial features, the original features and the decoded features are fused, and then anomaly detection is performed by using an anomaly detection method K-nearest neighbor. In a social network environment, the number of abnormal user groups is smaller than that of normal user groups, so that collection of abnormal users is relatively troublesome in a data collection process. According to the method provided by the invention, the defects of high-cost labeling and unbalanced positive and negative samples in the existing social robot detection method are overcome, and efficient detection of social network machine users is realized by reducing abnormal sample participation model training.

Description

technical field [0001] The invention belongs to the technical field of anomaly detection, and in particular relates to a social robot detection based on differential autoencoding. Background technique [0002] With the great popularity and development of the Internet, a large amount of real online user behavior data is provided for the study of human behavior. As of December 2020, the number of netizens in my country has reached 989 million, and the daily active users of Twitter have reached 192 million. As of September 2020, the monthly active users of Weibo are 511 million, and the average number of daily active users is 224 million. Such a huge The number of users generates terabytes of data every day, which records the rich online behavior of thousands of users. Social media has become an integral part of people's lives for obtaining and sharing information. In general, social media sites such as Twitter and Weibo bring us unprecedented opportunities to study whether us...

Images