Thesis Title: Employing Blockchain and Machine Learning to Detect Malicious Nodes in the Internet of Sensor Things
Abstract: A Blockchain Ensemble stacked Machine Learning (BEML) approach has been proposed. The BEML approach consists of three modules: blockchain, InterPlanetary File System (IPFS) and attack detection. The blockchain module registers network nodes, authenticates data analysts, revokes network nodes, and securely stores data hashes and node credentials. The IPFS module stores the data and generates a unique hash for it, which is later used to access and download the data from IPFS. In the third module, the raw data is processed, normalized and balanced using the MinMax scalar and Synthetic Minority Oversampling Technique (SMOTE). Moreover, the single learner ML algorithms: linear discriminant analysis, decision tree, perceptron and ridge, are combined in a stacked fashion to compensate for the weaknesses of each individual algorithm. The resulting model demonstrates better performance and detection accuracy compared to the single learner algorithms. This stacked ML model is used to detect and classify the Denial of Service (DoS) attacks present in the network. Based on the predicted attacks, malicious nodes are identified and their registration is revoked from the network. Finally, simulations are performed to evaluate the efficiency of the proposed BEML approach.
Supervisor: Prof. Dr. Nadeem Javaid,