Model for detecting phishing URLS
Methods, storage systems and computer program products implement embodiments of the present invention for protecting a computing device. These embodiments include detecting that an email is received by the computing device, the email including a Uniform Resource Locator (URL) for a web page in a first domain. The web page is retrieved from the domain, and a set of keywords are extracted from the retrieved web page. A query included the set of keywords is submitted to a search engine, and a response to the query is received from the search engine, the response indicating a set of second domain. Finally, in response to detecting that the first domain does not match any of the second domains, an alert for a phishing attack is generated.
1 . A method for protecting a computing device, comprising:
detecting an email received by the computing device and comprising a Uniform Resource Locator (URL) for a web page in a first domain;
retrieving the web page from the first domain;
determining a plurality of words that would be visible when the web page is rendered;
selecting, from the plurality of words, a set of words to be used as keywords;
submitting, to a search engine, a query comprising the set of keywords;
receiving, from the search engine, a response to the query, the response indicating a set of second domains and respective rankings for the second domains, where the ranking for a given second domain is provided by relative position of the given second domain in the set of second domains and is indicative of a quality of the second domain in relation to the keywords; and
generating an alert for a phishing attack responsively to detecting that either the first domain does not match any of the second domains or the first domain matches a second domain that has a ranking exceeding a specified ranking threshold.
2 . The method according to claim 1 , and further comprising retrieving Hypertext Markup Language (HTML) code associated with the web page, rendering the HTML code, and determining, based on the rendering, the words that would be visible.
3 . The method according to claim 1 , wherein selecting the set of keywords comprises applying a statistical model to the plurality of words so as to rank the words in order of importance, wherein the set of keywords comprises a specific number of the highest ranked words.
4 . The method according to claim 1 , wherein the web page comprises a first web page, wherein the first web page comprises a redirection to a second web page, and wherein selecting the set of keywords comprises determining the plurality of words that would be visible when the second web page is rendered and selecting, from the plurality of words, the set of words to be used as keywords.
5 . The method according to claim 4 , wherein the redirection comprises the first web page redirecting to the second web page within a specified amount of time.
6 . The method according to claim 1 , and further comprising identifying a first owner of the first domain, and identifying respective second owners for the second domains, and wherein detecting that the first domain does not match any of the second domains comprises detecting that the first owner does not match any of the second owners.
7 . The method according to claim 1 , and further comprising generating a screenshot of the retrieved web page, comparing the screenshot to logo images in a set of logo images, each logo image associated with a respective third domain, and generating the alert upon detecting a match between the screenshot and a given logo image in the set of logo images where the third domain associated with the given logo image does not match the first domain.
8 . The method according to claim 7 , wherein comparing the screenshot to the logo images comprises generating a first set of first keypoints for the screenshot, generating respective second sets of second keypoints for the logo images, and comparing the first set to the second sets.
9 . The method according to claim 8 , wherein detecting the match between the generated screenshot and the given logo image comprises detecting at least a specified number of matches between the first set of first keypoints and the second set of second keypoints for the given logo image.
10 . The method according to claim 8 , wherein comparing the first set to the second set comprises measuring respective scale-invariant feature transform (SIFT) distances between the first and the second sets.
11 . The method according to claim 1 , and further comprising generating the alert upon detecting a login form in the retrieved web page.
12 . The method according to claim 11 , wherein detecting the login form comprises rendering HTML code for the retrieved web page, extracting a set of words from the HTML code, comparing the extracted words to a set of login keywords, and detecting a match between a given extracted word and a given login keyword.
13 . The method according to claim 12 , wherein detecting the login form comprises rendering HTML code for the retrieved web page, extracting a set of HTML tags from the HTML code, comparing the extracted words to a set of login tags, and detecting a match between a given extracted word and a given login tag.
14 . The method according to claim 1 , and further comprising ascertaining an age of the first domain, and generating the alert upon detecting that the age exceeds a specified age threshold.
15 . The method according to claim 1 , and further comprising extracting a set of features from the URL, modeling the extracted features so as to classify the URL as either suspicious or unknown, and generating the alert upon classifying the URL as suspicious.
16 . The method according to claim 15 , wherein a given feature comprises a number of times any of one or more specified characters are in the URL.
17 . The method according to claim 15 , wherein a given feature comprises a number of times any of one or more specified words are in the URL.
18 . The method according to claim 15 , wherein a given feature comprises whether or not the web page is hosted by a free hosting service.
19 . The method according to claim 15 , wherein a given feature comprises whether or not the URL comprises an Internet Protocol (IP) address.
20 . The method according to claim 15 , wherein a given feature comprises a number of subdomains in the URL.
21 . The method according to claim 16 , wherein a given feature is selected from a group including a length of a path in the URL, a length of the URL and a length of the domain.
22 . A computing device, comprising:
a memory; and
a processor configured:
to detect an email received by the computing device and comprising a Uniform Resource Locator (URL) for a web page in a first domain,
to retrieve the web page from the first domain,
to determine a plurality of words that would be visible when the web page is rendered,
to select, from the plurality of words, a set of words to be used as keywords,
to submit, to a search engine, a query comprising the set of keywords,
to receive, from the search engine, a response to the query, the response indicating a set of second domains and respective rankings for the second domains, where the ranking for a given second domain is provided by relative position of the given second domain in the set of second domains and is indicative of a quality of the second domain in relation to the keywords, and
to generate an alert for a phishing attack responsive to detecting that either the first domain does not match any of the second domains or the first domain matches a second domain that has a ranking exceeding a specified ranking threshold.
23 . A computer software product for protecting a computing device, the computer software product comprising a non-transitory computer-readable medium, in which program instructions are stored, which instructions, when read by a computer, cause the computer:
to detect an email received by the computing device and comprising a Uniform Resource Locator (URL) for a web page in a first domain;
to retrieve the web page from the first domain;
to determine a plurality of words that would be visible when the web page is rendered;
to select, from the plurality of words, a set of words to be used as keywords;
to submit, to a search engine, a query comprising the set of keywords;
to receive, from the search engine, a response to the query, the response indicating a set of second domains and respective rankings for the second domains, where the ranking for a given second domain is provided by relative position of the given second domain in the set of second domains and is indicative of a quality of the second domain in relation to the keywords; and
to generate an alert for a phishing attack responsive to detecting that either the first domain does not match any of the second domains or the first domain matches a second domain that has a ranking exceeding a specified ranking threshold.