Systems and methods for contextual analysis and segmentation using dynamically-derived topics
View Patent ↗Systems and methods are disclosed for contextual analysis and segmentation of information objects. According to one implementation, information objects, such as web pages and user profiles, may be analyzed to identify key terms. These key terms may be included in a contextual representation of an information object. By comparing the contextual representations of a plurality of information objects, one or more contextual segments (i.e., categories of information objects) may be created. Each contextual segment may also be associated with its own contextual representation. Once a contextual segment has been created, information objects may be assigned to the contextual segment. These contextual segments may be used to deliver targeted advertising, for example.
1. A method of contextual analysis and segmentation of information objects, comprising the following operations performed by at least one processor:
accessing a plurality of information objects;
generating a contextual representation for each of the plurality of information objects;
identifying, using a processor, similarities between the contextual representations;
based on the identified similarities, creating a plurality of contextual segments, each of the plurality of contextual segments representing a subset of the plurality of information objects;
generating a contextual representation for each contextual segment, wherein the contextual representation for each contextual segment comprises a plurality of key terms aggregated from the plurality of information objects represented by the contextual segment and a plurality of weights associated with each of the plurality of key terms; and
assigning at least one information object to the at least one each contextual segment.
2. The method of claim 1 , wherein the plurality of information objects comprises a plurality of web pages.
3. The method of claim 1 , wherein the plurality of information objects comprises a plurality of user profiles.
4. The method of claim 3 , wherein each of the plurality of user profiles is comprised of a plurality of web pages obtained from a browsing history of a user.
5. The method of claim 1 , wherein generating a contextual representation for each of the plurality of information objects comprises identifying a plurality of key terms associated with each of the plurality of information objects.
6. The method of claim 5 , wherein generating a contextual representation for each of the plurality of information objects further comprises assigning a weight to each of the key terms associated with each of the plurality of information objects.
7. The method of claim 6 , wherein the assigned weight is based on a frequency that the key term appears in the information object.
8. The method of claim 6 , wherein the assigned weight is based on a location within the information object from which the key term was obtained.
9. The method of claim 5 , wherein identifying similarities between the contextual representations comprises comparing the pluralities of key terms.
10. The method of claim 1 , wherein generating a contextual representation for each of the plurality of information objects comprises generating n-grams from each of the plurality of information objects and identifying key n-grams to include in each of the contextual representations.
11. The method of claim 1 , wherein creating the plurality of contextual segments comprises creating each contextual segment upon determining that a number of identified similarities between the contextual representations exceeds a threshold value.
12. The method of claim 1 , wherein assigning the at least one information object to the each contextual segment comprises assigning at least one information object from the plurality of information objects to each contextual segment.
13. The method of claim 1 , wherein assigning the at least one information object to each contextual segment comprises assigning at least one information object not present in the plurality of information objects to each contextual segment.
14. A system of contextual analysis and segmentation of information objects, comprising:
a memory; and a processor coupled to the memory and configured to:
access a plurality of information objects;
generate a contextual representation for each of the plurality of information objects;
identify similarities between the contextual representations;
based on the identified similarities, create a plurality of contextual segments, each of the plurality of contextual segments representing a subset of the plurality of information objects; and
generate a contextual representation for each contextual segment, wherein the contextual representation for each contextual segment comprises a plurality of key terms aggregated from the plurality of information objects represented by the contextual segment and a plurality of weights associated with each of the plurality of key terms; and
assign at least one information object to the at least one each contextual segment.
15. The system of claim 14 , wherein the processor is further configured to access a plurality of information objects comprising a plurality of web pages.
16. The system of claim 14 , wherein the processor is further configured to access a plurality of information objects comprising a plurality of user profiles.
17. The system of claim 14 , wherein the processor is further configured to access a plurality of information objects comprising a plurality of user profiles, wherein each of the plurality of user profiles is comprised of a plurality of web pages obtained from a browsing history of a user.
18. The system of claim 14 , wherein the processor is further configured to generate a contextual representation for each of the plurality of information objects by identifying a plurality of key terms associated with each of the plurality of information objects.
19. The system of claim 18 , wherein the processor is further configured to generate a contextual representation for each of the plurality of information objects by assigning a weight to each of the key terms associated with each of the plurality of information objects.
20. The system of claim 19 , wherein the processor is further configured to assign a weight to each of the key terms based on a frequency that each of the key term appears in the information object.
21. The system of claim 19 , wherein the processor is further configured to assign a weight to each of the key terms based on a location within the information object from which each of the key terms was obtained.
22. The system of claim 18 , wherein the processor is further configured to identify similarities between the contextual representations by comparing the pluralities of key terms.
23. The system of claim 14 , wherein the processor is further configured to generate a contextual representation for each of the plurality of information objects by generating n-grams from each of the plurality of information objects and identifying key n-grams to include in each of the contextual representations.
24. The system of claim 14 , wherein the processor is further configured to create the plurality of contextual segments by creating each contextual segment upon determining that a number of identified similarities between the contextual representations exceeds a threshold value.
25. The system of claim 14 , wherein the processor is further configured to assign at least one information object from the plurality of information objects to each contextual segment.
26. The system of claim 14 , wherein the processor is further configured to assign at least one information object not present in the plurality of information objects to each contextual segment.
27. A computer-readable storage medium including instructions for contextual analysis and segmentation of information objects, which, when executed, perform steps comprising:
accessing a plurality of information objects;
generating a contextual representation for each of the plurality of information objects;
identifying similarities between the contextual representations;
based on the identified similarities, creating a plurality of contextual segments, each of the plurality of contextual segments representing a subset of the plurality of information objects;
generating a contextual representation for each contextual segment, wherein the contextual representation for each contextual segment comprises a plurality of key terms aggregated from the plurality of information objects represented by the contextual segment and a plurality of weights associated with each of the plurality of key terms; and
assigning at least one information object to the at least one each contextual segment.
28. The computer-readable storage medium of claim 27 , wherein the step of generating a contextual representation for each of the plurality of information objects comprises generating n-grams from each of the plurality of information objects and identifying key n-grams to include in each of the contextual representations.
29. The computer-readable storage medium of claim 27 , wherein the step of creating the plurality of contextual segments comprises creating each contextual segment upon determining that a number of identified similarities between the contextual representations exceeds a threshold value.