Method and system for occurrence frequency-based scaling of navigation path weights among online content sources

• US 7,660,815 B1
• Filed: 06/30/2006
• Issued: 02/09/2010
• Est. Priority Date: 06/30/2006
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Scofield, Christopher L., Pope, Elmore Eugene, Marshall, Brad E.
• Primary Examiner(s): Ali; Mohammad
• Assistant Examiner(s): HASAN, SYED HAROON
• Application Number:
  US11/478,783
• Time in Patent Office: 1,320 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A method and system for scaling navigation path weights among online content sources. A method may include determining a first probability of users traversing a first navigation path from a first to a second online content source, dependent upon a ratio of traversals of the first navigation path and traversals of all navigation paths to the second online content source. The method may also include determining a second probability of users traversing navigation paths from the first online content source to any of the online content sources, dependent upon a ratio of the plurality of online content sources to which navigation paths from the first online content source exist and a total number of the plurality of online content sources. The method may further include generating from the probabilities a scaling factor indicative of a strength of the first navigation path relative to other navigation paths among the online content sources.

Claims

1. A method implemented on one or more computing devices, comprising:

determining a first probability of users traversing a first navigation path from a first to a second one of a plurality of online content sources, wherein said first navigation path is one of a plurality of navigation paths from said first online content source to other ones of said plurality of online content sources, and wherein said first probability is determined dependent upon a ratio of a number of user traversals of said first navigation path to a number of user traversals of all navigation paths to said second online content source;

determining a second probability of users traversing navigation paths from said first online content source to any of said plurality of online content sources, wherein said second probability is determined dependent upon a ratio of a number of said plurality of online content sources to which navigation paths from said first online content source exist to a total number of said plurality of online content sources; and

generating a scaling factor indicative of a strength of said first navigation path relative to other navigation paths among said plurality of online content sources that include said first online content source, wherein said scaling factor is generated dependent upon a ratio of said first probability to said second probability.

2. The method as recited in claim 1, wherein said generating said scaling factor is further dependent upon said total number of said plurality of online content sources.

3. The method as recited in claim 1, further comprising:

determining a path weight corresponding to said first navigation path dependent upon a function of content access request traffic along said first navigation path; and

generating a scaled path weight as a function of said path weight and said scaling factor.

4. The method as recited in claim 3, wherein determining said path weight is dependent upon a time rate of change of content access request traffic along said first navigation path.

5. The method as recited in claim 3, wherein generating said scaled path weight includes multiplying said scaling factor and said path weight.

6. The method as recited in claim 3, further comprising respectively determining said first probability, said second probability, said scaling factor, said path weight and said scaled path weight for each of said plurality of navigation paths among said plurality of online content sources.

7. The method as recited in claim 6, further comprising generating one or more interest spaces dependent upon said navigation paths and said respective scaled path weights, wherein each interest space includes at least a subset of said online content sources.

8. A tangible, computer-accessible storage medium comprising instructions, wherein the instructions are executable to:

determine a first probability of users traversing a first navigation path from a first to a second one of a plurality of online content sources, wherein said first navigation path is one of a plurality of navigation paths from said first online content source to other ones of said plurality of online content sources, and wherein said first probability is determined dependent upon a ratio of a number of user traversals of said first navigation path to a number of user traversals of all navigation paths to said second online content source;

determine a second probability of users traversing navigation paths from said first online content source to any of said plurality of online content sources, wherein said second probability is determined dependent upon a ratio of a number of said plurality of online content sources to which navigation paths from said first online content source exist to a total number of said plurality of online content sources; and

generate a scaling factor indicative of a strength of said first navigation path relative to other navigation paths among said plurality of online content sources that include said first online content source, wherein said scaling factor is generated dependent upon a ratio of said first probability to said second probability.

9. The computer-accessible storage medium as recited in claim 8, wherein said scaling factor is generated dependent upon said total number of said plurality of online content sources.

10. The computer-accessible storage medium as recited in claim 8, wherein the instructions are further executable to:

determine a path weight corresponding to said first navigation path dependent upon a function of content access request traffic along said first navigation path; and

generate a scaled path weight as a function of said path weight and said scaling factor.

11. The computer-accessible storage medium as recited in claim 10, wherein said path weight is determined dependent upon a time rate of change of content access request traffic along said first navigation path.

12. The computer-accessible storage medium as recited in claim 10, wherein to generate said scaled path weight, said instructions are further executable to multiply said scaling factor and said path weight.

13. The computer-accessible storage medium as recited in claim 10, wherein said instructions are further executable to respectively determine said first probability, said second probability, said scaling factor, said path weight and said scaled path weight for each of said plurality of navigation paths among said plurality of online content sources.

14. The computer-accessible storage medium as recited in claim 13, wherein said instructions are further executable to generate one or more interest spaces dependent upon said navigation paths and said respective scaled path weights, wherein each interest space includes at least a subset of said online content sources.

15. A system, comprising:

one or more devices configurable to implement;

a web services interface configured to receive, according to a web services protocol, indications of access requests directed to respective ones of a plurality of online content sources, wherein said indications are indicative of user traversals of navigation paths among said plurality of online content sources; and

an interest analysis system configured to receive said indications from said web services interface and to;

determine a first probability of users traversing a first navigation path from a first to a second one of said plurality of online content sources, wherein said first navigation path is one of a plurality of navigation paths from said first online content source to other ones of said plurality of online content sources, and wherein said first probability is determined dependent upon a ratio of a number of user traversals of said first navigation path to a number of user traversals of all navigation paths to said second online content source;

determine a second probability of users traversing navigation paths from said first online content source to any of said plurality of online content sources, wherein said second probability is determined dependent upon a ratio of a number of said plurality of online content sources to which navigation paths from said first online content source exist to a total number of said plurality of online content sources; and

generate a scaling factor indicative of a strength of said first navigation path relative to other navigation paths among said plurality of online content sources that include said first online content source, wherein said scaling factor is generated dependent upon a ratio of said first probability to said second probability.

16. The system as recited in claim 15, wherein said interest analysis system is further configured to generate said scaling factor dependent upon said total number of said plurality of online content sources.

17. The system as recited in claim 15, wherein said interest analysis system is further configured to:

determine a path weight corresponding to said first navigation path dependent upon a function of content access request traffic along said first navigation path; and

generate a scaled path weight as a function of said path weight and said scaling factor.

18. The system as recited in claim 17, wherein said path weight is determined dependent upon a time rate of change of content access request traffic along said first navigation path.

19. The system as recited in claim 17, wherein to generate said scaled path weight, said interest analysis system is further configured to multiply said scaling factor and said path weight.

20. The system as recited in claim 17, wherein said interest analysis system is further configured to respectively determine said first probability, said second probability, said scaling factor, said path weight and said scaled path weight for each of said plurality of navigation paths among said plurality of online content sources.

21. The system as recited in claim 20, wherein said interest analysis system is further configured to generate one or more interest spaces dependent upon said navigation paths and said respective scaled path weights, wherein each interest space includes at least a subset of said online content sources.

22. A method implemented on one or more computing devices, comprising:

determining a path weight corresponding to a first navigation path from a first to a second one of a plurality of online content sources dependent upon a function of content access request traffic along said first navigation path;

determining a scaling factor corresponding to the navigation path such that the scaling factor is directly proportional to a first factor comprising a ratio of (a) a number of user traversals from said first online content source to said second content source to (b) a number of user traversals from any of said online content sources into said second online content source, and such that the scaling factor is inversely proportional to a second factor comprising a ratio of (c) a number of online content sources linked to by said first online content source to (d) a total number of said online content sources; and

generating a scaled path weight as a function of said path weight and said scaling factor.