File vulnerability detection system and detection method thereof

A file vulnerability detection method comprising:

• translating a binary file into an intermediate file;
• analyzing the intermediate file to obtain multiple functions to be tested;
• establishing function data of each of the functions to be tested;
• and
• comparing correlations between the function data of each of the functions to be tested and function data of at least one function having a version with vulnerability and a version without vulnerability in a vulnerability database based on a characteristic model, to determine whether each of the functions to be tested corresponding to each function data has a vulnerability, wherein the characteristic model has information representing multiple back-end binary files generated by multiple back-end platforms;
• wherein a processor respectively compares each function data with the function data in each function having the version with vulnerability and the version without vulnerability to respectively generate each pair of a first similarity and a second similarity;
• the processor sums up the first similarity and the second similarity in each function;
• and the processor selects a function with the highest value in all functions to determine whether each of the functions to be tested corresponding to each function data has the vulnerability.

A file vulnerability detection system comprising:

• a memory, configured to store multiple instructions and a vulnerability database;
• and
• a processor, connected to the memory and configured to execute the instructions to:
• translate a binary file into an intermediate file;
• analyze the intermediate file to obtain multiple functions to be tested;
• establish function data of each of the functions to be tested;
• and
• compare correlations between the function data of each of the functions to be tested and function data of at least one function having a version with vulnerability and a version without vulnerability in a vulnerability database based on a characteristic model to determine whether each of the functions to be tested corresponding to each function data has a vulnerability, wherein the characteristic model has information representing multiple back-end binary files generated by multiple back-end platforms;
• wherein the processor is further configured to:
• when each function data has a function name, compare each function data with the function data having the function name in the at least one function having the version with vulnerability by a processor to generate a first similarity;
• when each function data has the function name, compare each function data with the function data having the function name in the at least one function having the version without vulnerability by the processor to generate a second similarity;
• and
• determine whether each of the functions to be tested corresponding to each function data has the vulnerability according to the first similarity and the second similarity.

A file vulnerability detection system comprising:

• a memory, configured to store multiple instructions and a vulnerability database;
• and
• a processor, connected to the memory and configured to execute the instructions to:
• translate a binary file into an intermediate file;
• analyze the intermediate file to obtain multiple functions to be tested;
• establish function data of each of the functions to be tested;
• and
• compare correlations between the function data of each of the functions to be tested and function data of at least one function having a version with vulnerability and a version without vulnerability in a vulnerability database based on a characteristic model to determine whether each of the functions to be tested corresponding to each function data has a vulnerability, wherein the characteristic model has information representing multiple back-end binary files generated by multiple back-end platforms;
• wherein the processor is further configured to:
• respectively compare each function data and the function data in each function having the version with vulnerability and the version without vulnerability by a processor to respectively generate each pair of a first similarity and a second similarity;
• sum up the first similarity and the second similarity in each function by the processor;
• and
• select a function with the highest value in all functions by the processor to determine whether each of the functions to be tested corresponding to each function data has the vulnerability.