@DisCo: Binary-Level Vulnerability Detection

@DisCo is a procedure for generating function and library embeddings from disassembled binary data using a graph database and processing those embeddings through supervised machine learning techniques. This technique aims to identify relationships and similarities between functions and libraries to identify potential vulnerabilities in project code. The solution utilizes the Orange data mining tool and requires knowledge of Python and Orange to set up and use. The advantage of this technique is that it can identify vulnerabilities at the binary level rather than requiring access to the source code. The benefits of using this technique include an improved understanding of project code and more secure and bug-free applications.

This software is under copyright.To request a copy of @DisCo, please fill out the @DisCo Request for Download form and email it to agradmin@inl.gov.