Abstract
Since the cost of fixing vulnerabilities can be thirty times greater after an application has been deployed, it is recognized that properly-written code can yield potentially large savings. Accordingly, approaches presented herein apply machine learning and Artificial Intelligence (AI) techniques to improve developer experience by enabling developers to avoid introducing potential bugs and/or vulnerabilities while coding. Billions of lines of source code, which have already been written, are utilized as examples of how to write functional and secure code that is easy to read and to debug. By leveraging this wealth of available data, which is complemented with state-of-art machine learning models, enterprise-level software solutions can be developed that have a high standard of coding and are potentially bug-free.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Sundaresan, Krishna; Tanwar, Anshul; Ganesan, Prasanna; Ravi, Sriram; Chandrasekaran, Sathish Kumar; and A., Parmesh, "SECURITY AND OTHER VULNERABILITY PREDICTION USING NOVEL DEEP REPRESENTATION OF SOURCE CODE WITH ACTIVE FEEDBACK LOOP", Technical Disclosure Commons, (June 17, 2020)
https://www.tdcommons.org/dpubs_series/3341