User profile classification by web usage analysis
View Patent ↗Demographic information of an Internet user is predicted based on an analysis of accessed web pages. Web pages accessed by the Internet user are detected and mapped to a user path vector which is converted to a normalized weighted user path vector. A centroid vector identifies web page access patterns of users with a shared user profile attribute. The user profile attribute is assigned to the Internet user based on a comparison of the vectors. Bias values are also assigned to a set of web pages and a user profile attribute can be predicted for an Internet user based on the bias values of web pages accessed by the user. User attributes can also be predicted based on the results of an expectation maximization process. Demographic information can be predicted based on the combined results of a vector comparison, bias determination, or expectation maximization process.
1. A machine-implemented method for extrapolating user profile information from user web page access patterns, comprising:
detecting a set of web pages accessed by a test user having an unknown user profile attribute;
mapping at least a subset of said detected web pages to a first data structure, said first data structure representing a web page access pattern of said test user;
comparing said first data structure to a plurality of a second data structure to obtain a comparison result, the plurality of said second data structure representing clusters of web page access patterns of a sample data set of users having a known user profile attribute in common;
evaluating based on said comparison result the plurality of said second data structure and said first data structure to identify a second data structure matching the web page access pattern of the first data structure; and
assigning said unknown user profile attribute of said test user from the matching second data structure to said test user;
wherein the known user profile attribute in common of the sample data set of users corresponds to the unknown user profile attribute of said test user;
wherein said first and second data structures are multi-dimensional vectors; and
wherein each dimension of said first and said second multi-dimensional vectors corresponds to a separate web page.
2. The method of claim 1 , wherein each dimension of said second data structure corresponds to an average of user multi-dimensional vectors in its corresponding cluster.
3. The method of claim 2 , wherein said comparing further comprises determining a distance between said multi-dimensional vectors.
4. The method of claim 3 , wherein said determining further comprises computing a cosine of an angle between said multi-dimensional vectors.
5. The method of claim 2 , wherein said unknown user profile attribute is demographic information.
6. The method of claim 5 , wherein said demographic information is gender information of the test user.
7. The method of claim 5 , wherein said demographic information is age information of the test user.
8. The method of claim 2 , wherein the multi-dimensional vectors are weighted to dampen disparity between how many time web pages are access by different users.
9. An apparatus, comprising:
a memory, said memory adapted to store program code;
a processor in communication with said memory, said program code capable of programming said processor to perform a method for extrapolating user profile information from user web page access patterns, the method comprising:
detecting a set of web pages accessed by a test user having an unknown user profile attribute;
mapping at least a subset of said detected web pages to a first data structure, said first data structure representing a web page access pattern of said test user;
comparing said first data structure to a plurality of a second data structure to obtain a comparison result, the plurality of said second data structure representing clusters of web page access patterns of a set of users having a known user profile attribute in common;
evaluating based on said comparison result the plurality of said second data structure and said first data structure to identify a second data structure matching the web page access pattern of the first data structure; and
assigning said unknown user profile attribute of said test user from the matching second data structure to said test user;
wherein the known user profile attribute in common of the sample data set of users corresponds to the unknown user profile attribute of said test user;
wherein said first and second data structures are multi-dimensional vectors; and
wherein each dimension of said first and said second multi-dimensional vectors corresponds to a separate web page.
10. The apparatus of claim 9 , wherein each dimension of said second structure corresponds to an average of user multi-dimensional vectors in its corresponding cluster.
11. The apparatus of claim 10 , wherein said comparing further comprises determining a distance between said multi-dimensional vectors.
12. The apparatus of claim 11 , wherein said determining further comprises computing a cosine of an angle between said multi-dimensional vectors.
13. The apparatus of claim 10 , wherein said unknown user profile attribute is demographic information.
14. The apparatus of claim 13 , wherein said demographic information is gender information of the test user.
15. The apparatus of claim 13 , wherein said demographic information is age information of the test user.
16. The apparatus of claim 10 , wherein the multi-dimensional vectors are weighted to dampen disparity between how many time web pages are access by different users.
17. A processor readable storage medium, comprising:
processor readable program code embodied on said processor readable storage medium, said processor readable program code for programming a processor to perform a method for extrapolating user profile information from user web page access patterns, the method comprising:
detecting a set of web pages accessed by a test user having an unknown user profile attribute;
mapping at least a subset of said detected web pages to a first data structure, said first data structure representing a web page access pattern of said test user;
comparing said first data structure to a plurality of a second data structure to obtain a comparison result, the plurality of said second data structure representing clusters of web page access patterns of a sample data set of users having a known user profile attribute in common;
evaluating based on said comparison result the plurality of said second data structure and said first data structure to identify a second data structure matching the web page access pattern of the first data structure; and
assigning said unknown user profile attribute of said test user from the matching second data structure to said test user;
wherein the known user profile attribute in common of the sample data set of users corresponds to the unknown user profile attribute of said test user;
wherein said first and second data structures are multi-dimensional vectors;
wherein each dimension of said first and said second multi-dimensional vectors corresponds to a separate web page.
18. The processor readable storage medium of claim 17 , wherein each dimension of said second data structure corresponds to an average of user multi-dimensional vectors in its corresponding cluster.
19. The processor readable storage medium of claim 18 , wherein said comparing further comprises determining a distance between said multi-dimensional vectors.
20. The processor readable storage medium of claim 19 , wherein said determining further comprises computing a cosine of an angle between said multi-dimensional vectors.
21. The processor readable storage medium of claim 18 , wherein said unknown user profile attribute is demographic information.
22. The processor readable storage medium of claim 21 , wherein said demographic information is gender information of the test user.
23. The processor readable storage medium of claim 21 , wherein said demographic information is age information of the test user.
24. The processor readable storage medium of claim 18 , wherein the multi-dimensional vectors are weighted to dampen disparity between how many time web pages are access by different users.