DOORchain: Deep Ontology-Based Operation Research to Detect Malicious Smart Contracts

Abstract

Blockchains have become of great vogue in different fields after the introduction of Bitcoin. There are some inherent problems that need to be solved. One of these problems is to ensure that secured transactions in blockchain are checked if they are malicious or not. This paper proposes DOORchain that combines three powerful approaches of detecting intrusions and maliciousness. They are Deep learning, Ontology, and Operation Research. This uses the advantage of constraints from operation research to formalize and detect network maliciousness, and ontology to detect behavioral maliciousness in particular. Then it feeds this formalization to deep learning in order to check if the transactions in blockchain are malicious or not. After applying the proposed DOORChain, the final results affirm that accuracy and recall are enhanced with a slight inescapable trade-off in precision.