Using popularity data for ranking

• US 7,783,632 B2
• Filed: 11/03/2005
• Issued: 08/24/2010
• Est. Priority Date: 11/03/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Rounthwaite, Robert L., Brill, Eric D., Ragno, Robert J., Richardson, Matthew R.
• Primary Examiner(s): Trujillo; James
• Assistant Examiner(s): Moser; Bruce M
• Application Number:
  US11/266,026
• Publication Number:
  US 20070100824A1
• Time in Patent Office: 1,755 Days
• Field of Search: 707/7, 707/5, 707/102, 707/999.007, 707/999.005, 707/999.102, 707/725, 707/727, 707/748
• US Class Current: 707/727
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A unique ranking system and method that facilitates improving the ranking and ordering of objects to further enhance the quality, accuracy, and delivery of search results in response to a search query. The system and method involve monitoring and tracking an object in terms of the number of times it'"'"'s been accessed and optionally by whom, when, for how long, and an access rate. The user'"'"'s interaction with the object can be tracked as well. By tracking the objects, a popularity measure can be determined. Popularity based rankings can be computed based on the popularity measure or some function thereof. The popularity measure can be affected by the access time, who accessed it, access duration or the user'"'"'s interaction with the object upon access. The popularity based rankings can be utilized by a search component to improve the quality and retrieval of search results.

Claims

1. An object ranking system embodied on a computer readable storage medium that facilitates improving search results and efficiencies comprising:

an object tracking component that tracks one or more objects and records data in connection with at least one of accessed objects and non-accessed objects to determine a popularity measure for each object, wherein the popularity measure for an object is reduced upon a user selecting the object and navigating away from the object and wherein the popularity measure for an object is signified as being valid upon a user performing at least one of scrolling through the object, book-marking the object, or saving the object;

a ranking component that determines a score for each of the one or more objects based at least in part on the popularity measure for each object, the ranking component employing the score for each object to generate popularity based rankings for each of the one or more objects;

an analysis component that counts the number of times objects have been accessed within a period of time and determines the popularity measure based upon at least one of;

a count per object, a function of the count per object, and a function of the count per object and the duration of the period of time; and

a ranking aggregation component that combines object rankings as determined by one or more sources with the popularity based ranking to produce a new popularity based ranking for the one or more objects.

2. The system of claim 1, wherein the popularity measure is a function of one or more of the following:

user interaction with the object, length of time the object is accessed, identity of user who accessed the object, or identity of the object accessed.

3. The system of claim 2, wherein the ranking component ranks the one or more objects based in part on one of the following:

object access time, object access date, or object access rate.

4. The system of claim 3, wherein the ranking component comprises a machine learning component that facilitates determining the score and is customizable according to user preferences.

5. The system of claim 4, wherein the one or more objects comprise at least one of the following:

a web page, a document, an image, a video, a message, or an audio file.

6. The system of claim 5, wherein the object tracking component records at least one of an access count, access rate, access time, access date, or by whom the one or more objects are accessed.

7. The system of claim 6, wherein the ranking component determines the score for the one or more objects using at least one ranking function and the popularity measure wherein such usage by the ranking function depends on at least one of the following:

user identity, demographics of the user, user profile data, submission time of query, length of the query, or content of the query.

8. An object ranking method that facilitates improving search engine results and efficiencies comprising employing a processor executing computer executable instructions stored on a computer readable medium to perform the following acts:

tracking one or more objects, and recording data of a plurality of users in connection with at least one of accessed objects or non-accessed objects to determine a popularity measure for each object with respect to a user, wherein the popularity measure for each object is based upon the recorded data, wherein only the recorded data of users that have one or more similar interests to the user is employed;

determining a score for each of the one or more objects based at least in part on the popularity measure for each object;

ranking each of the one or more objects based at least in part on the popularity measure for each object; and

combining at least one object ranking for one or more objects determined by at least a first source with a popularity based ranking for the same one or more objects to obtain a new popularity based ranking.

9. The method of claim 8, further comprising communicating popularity based rankings to improve search results given in response to search queries.

10. The method of claim 9, further comprising collecting data relating to a user'"'"'s interaction with the one or more objects.

11. The method of claim 10, further comprising determining the score for the one or more objects based on the popularity measure and at least a portion of data collected relating to the user'"'"'s interaction with the one or more objects.

12. The method of claim 11, further comprising mitigating the popularity measure based in part on the portion of data collected.

13. The method of claim 12, wherein the data comprises time of access.

14. The method of claim 13, further comprising ranking the one or more objects based on a function of the popularity measure.

15. The method of claim 14, further comprising employing machine learning to determine the scores for the one or more objects.

16. An object ranking system embodied on a computer readable storage medium that facilitates improving search results and efficiencies comprising:

an object tracking component that tracks one or more objects and records data in connection with at least one of accessed objects and non-accessed objects to determine a popularity measure for each object, wherein the popularity measure for an object is reduced upon a user selecting the object and navigating away from the object and wherein the popularity measure for an object is a function of a user interaction with the object, and is signified as being valid upon a user scrolling through the object or book-marking the object;

a ranking component comprising a machine learning component, the ranking component being configured to determine a score for each of the one or more objects based at least in part on the popularity measure for each object, the ranking component employing the score for each object to generate popularity based rankings for each of the one or more objects;

an analysis component that counts the number of times objects have been accessed within a period of time and determines the popularity measure based upon a function of a count per object and a duration of the period of time; and

a ranking aggregation component that combines object rankings as determined by one or more sources with the popularity based ranking to produce a new popularity based ranking for the one or more objects, wherein

the machine learning component facilitates determining the score and is customizable according to user preferences, and

the ranking component ranks the one or more objects based at least partially upon an object access time, and determines the score for the one or more objects using at least one ranking function and the popularity measure, the usage of the at least one ranking function depending upon user profile data and content of a query.