USING CATEGORICAL METADATA TO RANK SEARCH RESULTS

• US 20110040752A1
• Filed: 08/14/2009
• Published: 02/17/2011
• Est. Priority Date: 08/14/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Svore, Krysta Marie, Bennett, Paul Nathan, Dumais, Susan T.
• Granted Patent: US 9,020,936 B2
• Time in Patent Office: Days
• US Class Current: 707/728
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/907 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A system that facilitates ranking search results returned by a search engine in response to receipt of a query is described herein. The system includes a receiver component that receives categorical metadata pertaining to an item and categorical metadata pertaining to the query and a computation component that computes at least one of a document feature pertaining to the item, a query feature pertaining to the query, or a document-query feature pertaining to the item and the query based at least in part upon one or more of the categorical metadata pertaining to the item or the categorical metadata pertaining to the query. The system also includes a ranker component that selectively places the item in a particular location in a sequence of items based at least in part upon the at least one of the document feature, the query feature, or the document-query feature.

Claims

1. A method configured for execution on a processor of a computing device comprising:

determining at least one of a document feature, a query feature, or a document-query feature based at least in part upon categorical metadata corresponding to at least one of an item accessible by way of a search engine or a query executable by the search engine to retrieve items; and

ranking search results based at least in part upon the at least one of the document feature, the query feature, or the document-query feature.

2. The method of claim 1, wherein the categorical metadata corresponding to the item accessible by way of the search engine comprises at least one category that is assigned to the item.

3. The method of claim 1, wherein the categorical metadata corresponding to the query usable by the search engine to retrieve items comprises at least one category that is assigned to the query.

4. The method of claim 3, further comprising assigning the categorical metadata to the query, wherein assigning the categorical metadata to the query comprises:

accessing a plurality of items that pertain to the query;

determining categories that are assigned, respectively, to the plurality of items that pertain to the query; and

assigning the categorical metadata to the query based at least in part upon the categories that are assigned, respectively, to the plurality of items that pertain to the query.

5. The method of claim 3, further comprising assigning the categorical metadata to the query, wherein assigning the categorical metadata to the query comprises:

executing a search over a plurality of items through use of the query;

generating an initial ranking of items returned in the search;

determining document categories assigned, respectively, to a top threshold number of items returned in the search; and

assigning the categorical metadata to the query based at least in part upon the categories that are assigned, respectively, to the top threshold number of items returned in the search.

6. The method of claim 3, further comprising assigning the categorical metadata to the query, wherein assigning the categorical metadata to the query comprises:

accessing a click graph, wherein the click graph comprises a first set of nodes and a second set of nodes, wherein edges couple nodes in the first set of nodes with nodes in the second set of nodes, wherein nodes in the first set of nodes are representative of queries, nodes in the second set of nodes represent items, and an edge between a node in the first set of nodes and a node in the second set of nodes indicates that a user has clicked on an item represented by the node in the first set of nodes upon issuing a query represented by the node in the second set of nodes to a search engine;

analyzing the click graph to determine a plurality of items that were clicked on by users when the query was issued to the search engine;

determining document categories assigned, respectively, to the plurality of items that were clicked on by the users when the query was issued; and

assigning the categorical metadata to the query based at least in part upon the document categories assigned, respectively, to the plurality of items that were clicked on by the users when the query was issued.

7. The method of claim 6, wherein assigning the categorical metadata to the query additionally comprises:

locating additional items by executing a random walk over the click graph;

determining document categories assigned, respectively, to the additional items located via the random walk; and

assigning the categorical metadata to the query based at least in part upon the document categories assigned, respectively, to the additional items located via the random walk.

8. The method of claim 1, wherein the search results comprise an item that is assigned a plurality of categories, and wherein the document feature is a score assigned to at least one of the plurality of categories.

9. The method of claim 1, wherein the search results comprise a plurality of items, wherein each of the plurality of items has at least one category assigned thereto, and wherein the document feature is a probability of at least one of the categories being assigned to an item.

10. The method of claim 1, wherein the search results comprise a plurality of items, wherein a subset of the plurality of items has at least one category assigned thereto, and wherein the document feature is a posterior entropy of at least one item in the plurality of items.

11. The method of claim 1, further comprising causing the search results to be displayed on a display screen, wherein the search results are ordered on the display screen according to the ranking of the search results.

12. The method of claim 1, wherein the search results comprise a plurality of items, wherein each of the plurality of items has at least one category assigned thereto, and wherein the query feature is an indication of whether the query utilized to execute the search has a particular category assigned thereto.

13. The method of claim 1, wherein the search results comprise a plurality of items, wherein each of the plurality of items has at least one category assigned thereto, and wherein the query feature is a binary value indicating whether a probability corresponding to a category assigned to the query is within a threshold.

14. The method of claim 1, wherein the search results comprise a plurality of items, wherein each of the plurality of items has at least one category assigned thereto, and wherein the query feature is an entropy value pertaining to a distribution of categories assigned to the query.

15. The method of claim 1, wherein the search results comprise a plurality of items, wherein each of the plurality of items has at least one category assigned thereto, and wherein the document-query feature indicates whether a category assigned to an item is also assigned to the query.

16. The method of claim 1, wherein the method is configured to execute responsive to receipt of the query.

17. A system that facilitates ranking search results returned by a search engine in response to receipt of a query, the system comprising the following computer-executable components:

a receiver component that receives categorical metadata pertaining to an item and categorical metadata pertaining to the query;

a computation component that computes at least one of a document feature pertaining to the item, a query feature pertaining to the query, or a document-query feature pertaining to the item and the query based at least in part upon one or more of the categorical metadata pertaining to the item or the categorical metadata pertaining to the query; and

a ranker component that selectively places the item in a particular location in a sequence of items based at least in part upon the at least one of the document feature, the query feature, or the document-query feature.

18. The system of claim 17, further comprising a query metadata determiner component that generates the categorical metadata pertaining to the query based at least in part upon categorical metadata assigned to one or more items returned by a search engine when executing the query.

19. The system of claim 17, wherein the query metadata determiner component generates the categorical metadata as a distribution of categories, wherein the categories in the distribution of categories are assigned to items returned by a search engine when executing a search using the query.

20. A computer-readable medium comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:

receiving a query;

accessing a plurality of items responsive to receipt of the query, wherein a subset of the plurality of items have categorical metadata respectively assigned thereto;

determining categorical metadata that has been assigned to the subset of items;

based at least in part upon the categorical metadata assigned to the subset of items, assigning categorical metadata to the query;

generating a value that indicates that a category assigned to an item in the plurality of items corresponds to a category assigned to the query; and

selectively positioning the item in a ranked list of search results based at least in part upon the generated value.