Research Article
A Real-Time Detection Algorithm for Abnormal Users in Multi Relationship Social Networks Based on Deep Neural Network
@INPROCEEDINGS{10.1007/978-3-030-94551-0_15, author={Ai-ping Zhang and Ying Chen}, title={A Real-Time Detection Algorithm for Abnormal Users in Multi Relationship Social Networks Based on Deep Neural Network}, proceedings={Advanced Hybrid Information Processing. 5th EAI International Conference, ADHIP 2021, Virtual Event, October 22-24, 2021, Proceedings, Part I}, proceedings_a={ADHIP}, year={2022}, month={1}, keywords={Deep neural network Multi relationship social network Abnormal user Real time detection}, doi={10.1007/978-3-030-94551-0_15} }- Ai-ping Zhang
Ying Chen
Year: 2022
A Real-Time Detection Algorithm for Abnormal Users in Multi Relationship Social Networks Based on Deep Neural Network
ADHIP
Springer
DOI: 10.1007/978-3-030-94551-0_15
Abstract
In view of the imbalance of abnormal data in social networks, the bandwidth value of detection algorithm is high. To solve this problem, a real-time anomaly detection algorithm based on deep neural network is designed. A multi relationship social gathering model was set up, and random forest was used to process tag data. The deep neural network is used to create a set of suspicious network abnormal nodes, and the time-varying component of abnormal data is set. The wavelet transform square integral is used to deal with the abnormal data acquisition process, and the real-time detection algorithm is finally constructed. Prepare the environment parameters required by the algorithm, build the algorithm running environment, prepare two kinds of traditional detection algorithm and design detection algorithm for experiments, the results show that the designed detection algorithm has the largest bandwidth and the best performance.
