Categorizing web sites based on content-temporal locality

• US 8,190,611 B1
• Filed: 09/28/2007
• Issued: 05/29/2012
• Est. Priority Date: 09/28/2007
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: Symantec Corporation (NortonLifeLock Inc.)
• Inventors: Nachenberg, Carey S., Bromwich, Adam
• Primary Examiner(s): HICKS, MICHAEL J
• Application Number:
  US11/864,307
• Time in Patent Office: 1,705 Days
• Field of Search: None
• US Class Current: 707/737
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing

Abstract

A categorization server is coupled to multiple clients via a network. Each client has a security module that monitors web browsing performed on the client and reports a web site browsing stream to the categorization server. The categorization server identifies a site from the browsing stream of a client that is of a known category. The categorization server uses content-temporal locality to determine whether other sites in the browsing stream belong to the same category as the site having the known category. This determination can be performed by assigning probabilities to other sites in the browsing stream, and by considering probabilities assigned to sites in browsing streams of other clients. The categorization server provides categories of sites to the clients, and the client security modules can implement category-based security policies.

Claims

1. A method of categorizing a web site on a network, comprising:

using a computer to perform steps comprising;

receiving a web site browsing stream describing temporal proximities of web sites browsed by a client, wherein a temporal proximity indicates a time interval between visits to browsed web sites;

identifying a category of a first web site in the web site browsing stream;

determining a temporal proximity of a second web site relative to the first web site in the web site browsing stream;

determining a probability that the second web site is of a same category as the first web site in the web site responsive at least in part to the temporal proximity of the second web site relative to the first web site and a degree of content-temporal locality for the category of the first web site; and

storing the determined category probability for the second web site.

2. The method of claim 1, wherein receiving a web site browsing stream comprises:

receiving a client identifier identifying the client; and

receiving a uniform resource locator (URL) referencing a web site browsed by the client.

3. The method of claim 1, wherein identifying the category of the first web site in the web site browsing stream comprises:

accessing a seed list, the seed list identifying web sites in categories in which other web sites on the network can be categorized.

4. The method of claim 1, further comprising:

receiving a plurality of web site browsing streams describing temporal proximities of web sites browsed by a plurality of clients; and

identifying web sites in the plurality of web site browsing streams that are of the same category as the first web site;

wherein the probability that the second web site is of the same category as the first web site is determined responsive at least in part to temporal proximities of the second web site relative to the web sites in the plurality of web site browsing streams that are of the same category as the first web site and the degree of content-temporal locality of the category of the first web site.

5. The method of claim 1, wherein a category of a web site identifies a type of content provided by the web site.

6. The method of claim 1, further comprising:

determining a category of the second web site based on the probability that the second web site is of the same category as the first web site and a probability that the second web site is of a same category as a third web site in the web site browsing stream, the first and third web sites having different categories and different degrees of content-temporal locality.

7. The method of claim 6, further comprising:

providing the determined category of the second web site to the client.

8. The method of claim 1, wherein different categories of web sites have different degrees of content-temporal locality and wherein determining a probability that the second web site is of the same category as the first web site comprises adjusting the probability that the second web site is of the same category as the first web site responsive to the degree of content-temporal locality for the category of the first web site.

9. A computer-implemented system for categorizing a web site on a network, comprising:

a computer processor; and

a computer-readable storage medium storing computer program modules configured to execute on the computer processor, the computer program modules comprising;

a communication module configured to receive a web site browsing stream describing temporal proximities of web sites browsed by a client, wherein a temporal proximity indicates a time interval between visits to browsed web sites;

a web site categorization module configured to;

identify a category of a first web site in the web site browsing stream;

determine a temporal proximity of a second web site relative to the first web site in the web site browsing stream; and

determine a probability that the second web site is of a same category as the first web site responsive at least in part to the temporal proximity of the second web site relative to the first web site and a degree of content-temporal locality for the category of the first web site; and

a data store module configured to store the determined category probability for the second web site.

10. The computer-implemented system of claim 9, wherein the client communication module is configured to:

receive a client identifier identifying the client;

receive a uniform resource locator (URL) referencing a web site browsed by the client; and

receive a timestamp specifying a time at which the client browsed the web site referenced by the URL.

11. The computer-implemented system of claim 9, wherein the web site categorization module is configured to:

access a seed list, the seed list identifying web sites in categories in which other web sites on the network can be categorized.

12. The computer-implemented system of claim 9, the computer program modules further comprising:

a review module configured to prioritize and/or triage the second web site for manual review responsive at least in part to the determined probability.

13. The computer-implemented system of claim 9, wherein the web site categorization module is further configured to:

receive a plurality of web site browsing streams describing temporal proximities of web sites browsed by a plurality of clients; and

identify web sites in the plurality of web site browsing streams that are of the same category as the first web site;

wherein the probability that the second web site is of the same category as the first web site is determined responsive at least in part to temporal proximities of the second web site relative to the web sites in the plurality of web site browsing streams that are of the same category as the first web site and the degree of content-temporal locality of the category of the first web site.

14. The computer-implemented system of claim 9, wherein the web site categorization module is configured to:

determine a category of the second web site based on the probability that the second web site is of the same category as the first web site and a probability that the second web site is of a same category as a third web site in the web site browsing stream, the first and third web sites having different categories and different degrees of content-temporal locality.

15. A non-transitory computer-readable storage medium storing computer program modules for categorizing a web site on a network, the modules comprising:

a communication module configured to receive a web site browsing stream describing temporal proximities of web sites browsed by a client, wherein a temporal proximity indicates a time interval between visits to browsed web sites;

a web site categorization module configured to;

identify a category of a first web site in the web site browsing stream;

determine a temporal proximity of a second web site relative to the first web site in the web site browsing stream; and

determine a probability that the second web site is of a same category as the first web site responsive at least in part to the temporal proximity of the second web site relative to the first web site and a degree of content-temporal locality for the category of the first web site; and

a data store module configured to store the determined probability for the second web site.

16. The computer-readable storage medium of claim 15, wherein the communication module is configured to:

receive a client identifier identifying the client;

receive a uniform resource locator (URL) referencing a web site browsed by the client; and

receive a timestamp specifying a time at which the client browsed the web site referenced by the URL.

17. The computer-readable storage medium of claim 15, wherein the web site categorization module is configured to:

access a seed list, the seed list identifying web sites in categories in which other web sites on the network can be categorized.

18. The computer-readable storage medium of claim 15, wherein the web site categorization module is further configured to:

receive a plurality of web site browsing streams describing temporal proximities of web sites browsed by a plurality of clients; and

identify web sites in the plurality of web site browsing streams that are of the same category as the first web site;

wherein the probability that the second web site is of the same category as the first web site is determined responsive at least in part to temporal proximities of the second web site relative to the web sites in the plurality of web site browsing streams that are of the same category as the first web site and the degree of content-temporal locality of the category of the first web site.

19. The computer-readable storage medium of claim 15, wherein the web site categorization module is configured to:

determine a category of the second web site based on the probability that the second web site is of the same category as the first web site and a probability that the second web site is of a same category as a third web site in the web site browsing stream, the first and third web sites having different categories and different degrees of content-temporal locality.

20. The computer-readable storage medium of claim 15, wherein different categories of web sites have different degrees of content-temporal locality and wherein determining a probability that the second web site is of the same category as the first web site comprises adjusting the probability that the second web site is of the same category as the first web site responsive to the degree of content-temporal locality for the category of the first web site.