Honeypot Coupled Machine Learning Model for Botnet Detection and Classification in IoT Smart Factory – An Investigation

Smart factories use a diversity of IoT equipment which on the other hand exposes to the high security risk such as botnet attacks. The current botnet detection models using honeypot are limited in accuracy and detection time. Therefore, honeypot combining ML is able to improve model performance.

The purpose of the present study was to present a model for botnet detection and rapid botnet classification in smart factories by combining honeypot and machine learning together. With the dataset collected in the log file, this proposed model is expected to efficiently minimise failure of botnet detection and information leakage in smart factory. In addition, using two ML classification techniques (R-studio and Weka), results were obtained for high accuracy, p-value, false positive of botnet tracking. Setup of the hardware configuration was very useful in this study to simulate the operation of smart factory and conduct the investigation of the honeypot combined machine learning model.
In future work, time taken build the model will be measured appropriately. Accuracy, false positive ratio and p-value are good measures to evaluate the results. It is suggested that more working parameters should be considered to include in order to increase the satisfaction and expectation of the capacity/scale of smart factory. Additionally, optimisation of this model is suggested to test in a real factory environment for a real-time evaluation.

Presentation Video

What do you think about the project? Leave your comment.

Taylor's InnoFest 2020

Where technology meets design and business

Contact Us

To learn more, don’t hesitate to get in touch

Copyright © 2020 Taylor's University.  All rights reserved.