Method of detecting potential phishing by analyzing universal resource locators
A method for detecting potential phishing URLs includes extracting a URL from a document, analyzing the URL context, and comparing the URL to stored trusted URLs and stored known phishing URLs. The URL context includes anchor text and surrounding content associated with the URL. The method further includes generating a phishing alert based on the comparing and the analyzing.
1. A method for detecting potential phishing Universal Resource Locators, comprising:
extracting a Universal Resource Locator from a document;
analyzing a context of the Universal Resource Locator including anchor text displayed as a hypertext link associated with the Universal Resource Locator, and surrounding content associated with the Universal Resource Locator;
comparing characters of the Universal Resource Locator to characters of stored trusted Universal Resource Locators and characters of stored known phishing Universal Resource Locators, wherein comparing the characters of the Universal Resource Locator includes calculating an image distance between the characters of the Universal Resource Locator and the characters of the stored trusted Universal Resource Locators, the image distance being based upon glyphs of the characters of the Universal Resource Locator; and
generating a phishing alert based on the comparing and the analyzing.
2. The method of claim 1 , further comprising building a database of trusted Universal Resource Locators.
3. The method of claim 2 , wherein building the database of trusted Universal Resource Locators includes a computer search to identify Universal Resource Locators that have been widely linked to by other websites.
4. The method of claim 1 , wherein the stored known phishing Universal Resource Locators are stored in a table.
5. The method of claim 1 , wherein the document can include one of a web page, an email message, a twitter text, a mobile text message, a chat message, and an instant message.
6. The method of claim 1 , further comprising segmenting the Universal Resource Locator into a domain name, a file path, and parameters.
7. The method of claim 1 , further comprising extracting the anchor text and the surrounding content associated with the Universal Resource Locator.
8. The method of claim 1 , further comprising calculating a consistency of the anchor text and the surrounding context to the Universal Resource Locator.
9. The method of claim 8 , wherein calculating an image distance includes using pre-computed image similarity table.
10. The method of claim 1 , wherein comparing the Universal Resource Locator includes calculating a text string similarity.
11. The method of claim 8 , wherein calculating a text string similarity is based on a model of text similarity learned from the stored known phishing Universal Resource Locator, mining the web, and web logs.
12. The method of claim 1 , wherein the method is performed by one of an email scanner, a web browser, a log scanner, a proxy server router, an intrusion detection/protection sensor, a security information event management system, a mobile device, and a handheld device.
13. A system comprising:
a memory; and
a processor executing instructions stored in the memory to perform a method, the method comprising:
extracting a Universal Resource Locator from a document;
analyzing a context of the Universal Resource Locator including anchor text displayed as a hypertext link associated with the Universal Resource Locator, and surrounding content associated with the Universal Resource Locator;
comparing characters of the Universal Resource Locator to characters of stored trusted Universal Resource Locators and characters of stored known phishing Universal Resource Locators, wherein comparing the characters of the Universal Resource Locator includes calculating an image distance between the characters of the Universal Resource Locator and the characters of the stored trusted Universal Resource Locators, the image distance being based upon glyphs of the characters of the Universal Resource Locator; and
generating a phishing alert based on the comparing and the analyzing.
14. The system of claim 13 , wherein the system includes one of a log scanner, a proxy server router, an intrusion detection/protection sensor, a security information event management system, a mobile device, and a handheld device.
15. A system comprising:
a memory; and
a processor executing instructions stored in the memory to perform a method, the method comprising:
calculating an edit distance and an image distance between characters of a Universal Resource Locator and characters of a reference Universal Resource Locator within a Universal Resource Locator whitelist, wherein the image distance is based upon glyphs of the characters of the Universal Resource Locator; and
generating a phishing alert when the edit distance or the image distance is below a threshold.
16. The system of claim 15 wherein the system is one of a domain name system, a domain name registry system, an intrusion detection scanner, a web proxy, a router filter, and a security event information management system.
17. The system of claim 15 wherein the processor is configured to redirect a user to a website with information about possible phishing.
18. The system of claim 15 wherein the processor is configured to log a potential phishing event.