System for static analysis of binary executable code and source code using fuzzy logic and method thereof

1. A computer-implemented static analysis system for binary executables and source code to detect vulnerabilities, undocumented features and other input program errors, the system comprising:

at least one computing device having a processor;

a decompiler operating on the processor configured to receive a source code of an input program and represent the input program in a target low-level language code (LLC) intermediate representation (IR);

a frontend operating on the processor configured to receive a binary code representation of the input program and represent the input program in the target LLC IR;

an analyzer configured to receive the target LLC IR and analyze the target LLC IR to detect vulnerabilities, undocumented features and input program errors using predetermined rules stored in a rule module and provided to the analyzer,

the analyzer having a fuzzy logic engine to reduce a number of false positives and false negatives results; and

a security report module comprises outputting a security report that comprises the vulnerabilities, the undocumented features and the input program errors detected by the analyzer,

wherein the fuzzy logic engine uses a fuzzy set of pairs (P, m), where P is a set of positives, all of which have a patternId, and m: P→[0, 1] a membership function which determines confidence.

2. The system according to claim 1 , wherein the target LLC IR is a Low Level Virtual Machine (LLVM) Intermediate Representation (IR).

3. The system according to claim 1 , wherein the fuzzy logic engine considers a plurality of metrics, the plurality of metrics comprising:

variety of complaining clients on each project;

number of complaining client on each project;

variety of projects from which complaints are received;

number of the projects; and

number of complaints.

4. The system according to claim 3 , wherein the fuzzy logic engine applies the following formula (I):

H

=

-

∑

n

=

1

N

p

n

⁢

log

⁢

p

n

,

∑

n

=

1

N

p

n

=

1

,

(

I

)

where p n is a percentage of errors among all mistakes made for all projects relative to a single pattern and N is a number of patterns.

5. The system according to claim 1 , wherein the analyzer configured to use taint analysis and/or symbolic interpretation analysis.

6. The system according to claim 1 , wherein the security report comprises: (a) detected vulnerabilities and undocumented features; (b) results within one project or different groups of projects to track the progress of vulnerability elimination or occurrence; (c) recommendations for cybersecurity and/or development teams; and (d) web application firewall (WAF) configuration recommendations to block and prevent possibility of exploiting vulnerabilities during the input program corrections.

7. A computer-implemented method for analyzing binary executables and source code of an input program to detect vulnerabilities, undocumented features and other input program errors, the method comprising:

identifying a presentation of the input program to determine whether the input program is presented in a source code or a binary code;

if the input program is presented in the source code, processing the input program by a decompiler to represent the input program in a target low-level language code (LLC) intermediate representation (IR);

if the input program is presented in the source code, processing the input program by a frontend to represent the input program and represent the input program in the target LLC IR;

analyzing by an analyzer the target LLC IR to detect vulnerabilities, undocumented features and input program errors using predetermined rules stored in a rule module and provided to the analyzer, wherein the analyzer comprises a fuzzy logic engine to reduce a number of false positives and false negatives results; and

outputting a security report by a security report module that comprises the vulnerabilities, the undocumented features and the input program errors detected by the analyzer,

wherein the fuzzy logic engine uses a fuzzy set of pairs (P, m), where P is a set of positives, all of which have a patternId, and in: P→[0, 1] a membership function which determines confidence.

8. The computer-implemented method according to claim 7 , wherein the target LLC IR is a Low Level Virtual Machine (LLVM) Intermediate Representation (IR).

9. The computer-implement method according to claim 8 , wherein the security report comprises: (a) detected vulnerabilities and undocumented features; (b) results within one project or different groups of projects to track the progress of vulnerability elimination or occurrence; (c) recommendations for cybersecurity and/or development teams; and (d) WAF configuration recommendations to block and prevent possibility of exploiting vulnerabilities during the input program corrections.

10. The computer-implement method according to claim 7 , wherein the fuzzy logic engine considers a plurality of metrics, the plurality of metrics comprising:

variety of complaining clients on each project;

number of complaining client on each project;

variety of projects from which complaints are received;

number of the projects; and

number of complaints.

11. The computer-implement method according to claim 10 , wherein the fuzzy logic engine applies the following formula (I):

H

=

-

∑

n

=

1

N

p

n

⁢

log

⁢

p

n

,

∑

n

=

1

N

p

n

=

1

,

(

I

)

where p n is a percentage of errors among all mistakes made for all projects relative to a single pattern and N is a number of patterns.

12. The computer-implement method according to claim 7 , wherein the analyzer configured to use taint analysis and/or symbolic interpretation analysis.