Pryliepov Y. The informational technology for abnormal data detection in the Internet of Things based on the cluster analysis

Українська версія

Thesis for the degree of Candidate of Sciences (CSc)

State registration number

0419U000596

Applicant for

Specialization

  • 05.13.06 - Інформаційні технології

28-02-2019

Specialized Academic Board

Д 26.861.05

State University of Telecommunications

Essay

The dissertation is dedicated to increase an efficiency of abnormal data detection in the Internet of things based on the cluster analysis. The current scientific task was formulated and solved, the goal was to to improve methods for abnormal data detection in the IoT. The data clustering method based on the k-means algorithm has been improved. The method of evaluation of classifier and reliability of the output clustered data of was improved. For the first time the new informational technology for abnormal data detection in the Internet of Things based on the cluster analysis was developed. The proposed information technology allows to detect the intrusions with a probability of 98.6%, which is confirmed by the relevant indicators of the effectiveness of detecting abnormal data on the Internet of Things.

Files

Similar theses