Difference between revisions of "TOOBAD4ML"
From Security Unileon
SecurityAdm (talk | contribs) |
SecurityAdm (talk | contribs) |
||
| Line 4: | Line 4: | ||
The input consists of one or several ''.c'' files that are processed for lexical analysis. The aim of this analysis is to extract an arbitrary number of features that describe a Buffer Overflow. To do so, the input must be pre-tagged: either manually or by means of static code analysis. Finally, the output is a set of vector descriptors, which can be exported to various formats in order to create a dataset suitable to be analyzed using Machine Learning techniques. | The input consists of one or several ''.c'' files that are processed for lexical analysis. The aim of this analysis is to extract an arbitrary number of features that describe a Buffer Overflow. To do so, the input must be pre-tagged: either manually or by means of static code analysis. Finally, the output is a set of vector descriptors, which can be exported to various formats in order to create a dataset suitable to be analyzed using Machine Learning techniques. | ||
| + | |||
| + | == Built With == | ||
| + | |||
| + | * C++ | ||
| + | * [https://clang.llvm.org/docs/LibTooling.html LibTooling] - Library to support writing standalone tools based on Clang. | ||
== References == | == References == | ||
Latest revision as of 18:12, 9 November 2021
TOOl to Buffer Overflow Analysis and Description For Machine Learning (TOOBAD4ML) is a tool for extracting features of Buffer Overflow vulnerabilities written in C code in order to further analyze them with Machine Learning techniques.
TOOBAD4ML overview
The input consists of one or several .c files that are processed for lexical analysis. The aim of this analysis is to extract an arbitrary number of features that describe a Buffer Overflow. To do so, the input must be pre-tagged: either manually or by means of static code analysis. Finally, the output is a set of vector descriptors, which can be exported to various formats in order to create a dataset suitable to be analyzed using Machine Learning techniques.
Built With
- C++
- LibTooling - Library to support writing standalone tools based on Clang.
References
External Links
- Source code
