Method and system for detecting credential stealing attacks
An Active Vision detection method and system for detecting credential stealing attacks using an automated machine-learning page examination engine is provided that may be used to detect both brand-based and custom credential stealing attacks. The approach employs similarity analysis in a two-stage process that may be achieved through supervised or self-learning machine learning techniques and is comparable to human analysis. The Active Vision System is capable of self-learning; every new attack detected by the system becomes part of system's long-term memory making it incrementally more accurate in future predictions using its past experience.
1. A method for detecting a credential stealing attack comprising:
(a) loading a candidate web page into a browser memory;
(b) extracting different types of features from the candidate web page rendered in the browser memory;
(c) generating a similarity features set by comparing the different types of features extracted in (b) to corresponding custom features from a known custom credential stealing page fetched from a knowledge base storing a plurality of records of known custom credential stealing page;
(d) taking as input the similarity features set in (c), utilizing a trained binary classifier, to predict a binary result whether the candidate web page is the known custom credential stealing page; and
(e) upon determining the candidate web page is not the known custom credential stealing page, repeating (c)-(d) until the candidate web page is predicted to be a known custom credential stealing page or terminating the process by generating a verdict indicating the candidate web page is benign when all of the plurality of records are compared to the candidate web page.
2. The method of claim 1 , further comprising providing a graphical interface for displaying information regarding the candidate web page, the information comprising: (i) an identity of an infected machine on a network that has accessed the candidate web page if the candidate web page is predicted as a known custom credential stealing page and (ii) a feature of the infected machine, wherein the feature is selected from the group consisting of a machine location, a machine usage, a MAC ID, a type of machine, a machine operating system, and an identity of a machine user.
3. The method of claim 1 , further comprising providing a graphical interface for displaying information regarding a credential attack, the information comprising: a number of connection attempts made, a name of a Malware, a type of attack or attack category, a hacker group that performed the attack, a type of malware that was used, or a time of the attack.
4. The method of claim 1 , wherein the different types of features comprise at least two from the group consisting of visual features, natural language features and source code features.
5. The method of claim 1 , further comprising interacting with the candidate web page within the browser memory to collect forensics intelligence on the candidate web page's behavior.
6. The method of claim 1 , further comprising interacting with the candidate web page by establishing an out-of-band connection using an anonymous server to collect forensics intelligence on the candidate web page's behavior.
7. The method of claim 6 , wherein the anonymous server is a virtual private network or a proxy server.
8. The method of claim 6 , further comprising using a virtual mouse or virtual keyboard to interact with the candidate web page.
9. The method of claim 6 , wherein the candidate web page is predicted to be the known custom credential stealing page further based on the forensics intelligence on the candidate web page's behavior.
10. A system detecting a credential stealing attack comprising:
(i) a memory for storing a set of software instructions,
(ii) one or more processors configured to execute the set of software instructions to:
(a) load a candidate web page into a browser memory;
(b) extract different types of features from the candidate web page rendered in the browser memory;
(c) generate a similarity features set by comparing the different types of features extracted in (b) to corresponding custom features from a known custom credential stealing page fetched from a knowledge base storing a plurality of records of known custom credential stealing page;
(d) take as input the similarity features set in (c), utilizing a trained binary classifier, to predict a binary result whether the candidate web page is the known custom credential stealing page; and
(e) upon determining the candidate web page is not the known custom credential stealing page, repeat (c)-(d) until the candidate web page is predicted to be a known custom credential stealing page or terminate the process by generating a verdict indicating the candidate web page is benign when all of the plurality of records are compared to the candidate web page.
11. The system of claim 10 , wherein the one or more processors are further configured to provide a graphical interface for displaying information regarding the candidate web page, the information comprising: (i) an identity of an infected machine on a network that has accessed the candidate web page if the candidate web page is predicted as a known custom credential stealing page and (ii) a feature of the infected machine, wherein the feature is selected from the group consisting of a machine location, a machine usage, a MAC ID, a type of machine, a machine operating system, and an identity of a machine user.
12. The system of claim 10 , wherein the one or more processors are further configured to provide a graphical interface for displaying information regarding a credential attack, the information comprising: a number of connection attempts made, a name of a Malware, a type of attack or attack category, a hacker group that performed the attack, a type of malware that was used, or a time of the attack.
13. The system of claim 10 , wherein the different types of features comprise at least two from the group consisting of visual features, natural language features and source code features.
14. The system of claim 10 , wherein the one or more processors are further configured to interact with the candidate web page within the browser memory to collect forensics intelligence on the candidate web page's behavior.
15. The system of claim 10 , wherein the one or more processors are further configured to interact with the candidate web page by establishing an out-of-band connection using an anonymous server to collect forensics intelligence on the candidate web page's behavior.
16. The system of claim 15 , wherein the anonymous server is a virtual private network or a proxy server.
17. The system of claim 15 , wherein the one or more processors are further configured to use a virtual mouse or virtual keyboard to interact with the candidate web page.
18. The system of claim 15 , wherein the candidate web page is predicted to be the known custom credential stealing page further based on the forensics intelligence on the candidate web page's behavior.