Title: Research on anomaly detection method for hybrid big data subarea based on ant colony algorithm
Authors: Shu Xu
Addresses: College of Information and Electronic Engineering, Hunan City University, Yiyang, Hunan, 413000, China
Abstract: Due to the problems of low accuracy and poor degree of freedom of the existing big data anomaly detection methods, a mixed big data partition anomaly detection method based on ant colony algorithm is proposed. The number of common neighbourhood between nodes in weighted network is redefined and the mixed big data sub-region is realised. Combining the operation, vulnerability and threat of the database, the security situation value is substituted into the abnormal location part to form the coordinate matrix. The pheromone concentration of each region was calculated, and the region where the concentration was reduced was defined as the abnormal region to complete the big data anomaly detection. Experimental results show that this method has high accuracy, freedom of anomaly location and good accuracy performance, which is a great progress of big data anomaly detection technology. In the future, an effective method to repair abnormal data and improve the specific application scope of this method should be developed on the basis of this method.
Keywords: ant colony algorithm; mixed type; big data; subregion; abnormal detection; weighted network nodes; coordinate matrix.
International Journal of Information and Communication Technology, 2020 Vol.17 No.2, pp.164 - 177
Received: 26 Apr 2019
Accepted: 14 Jul 2019
Published online: 09 Jul 2020 *