Discovering expertise using document metadata in part to rank authors

Patent metadata

Field	Value
Publication number	US-9589072-B2
Application number	US-201113150710-A
Country	US
Kind code	B2
Filing date	Jun 1, 2011
Priority date	Jun 1, 2011
Publication date	Mar 7, 2017
Grant date	Mar 7, 2017

Abstract

Expertise mining features are provided based in part on the use of an expertise mining algorithm and expertise mining queries. A method of an embodiment operates to provide an expanded feedback query based in part on search results using an expertise mining query and a number of author-ranking heuristics used to rank authors and/or co-authors (e.g., primary authors, secondary authors, etc.) as part of an expertise mining operation. A search system of an embodiment includes an author ranker component to rank authors based in part on an expertise mining query and author-ranking heuristics, and a query expander component to provide expanded queries as part of identifying relevant search results. Other embodiments are also disclosed.

First claim

What is claimed is:

1. A method comprising:
using one or more search terms of an original query to generate an expertise mining query;
executing the expertise mining query to create search results;
filtering the search results based in part on a user identification;
identifying well-structured search items from the filtered search results;
ranking authors associated with the well-structured search items using an expertise mining algorithm comprising:
a first post-processing stage to:
calculate a cumulative rank calculation for each author based in part on a document rank and a number of authored documents, and
generate a first author ranking output based on the cumulative rank calculation; and

a second post-processing stage to:
identify an association between the first author ranking output and the original query, and
generate a second author ranking output based on the identified association,

wherein the identified association is based in part on occurrences of one or more of the search terms in title and uniform source locator (URL) attributes of the first author ranking output;

determining one or more ranked authors by merging the first author ranking output from the first post-processing stage and the second author ranking output from the second post-processing stage; and
automatically providing one or more expanding query terms with the original query including one or more of the ranked authors to provide an expanded query for use in mining expertise information.

2. The method of claim 1 , further comprising receiving the original query from a search interface including one or more original search terms and returning a weighted set of authors as experts.

3. The method of claim 1 , further comprising automatically executing the expanded query to generate search results that satisfy the expanded query including providing one or more user profiles and documentary evidence as part of the search results.

4. The method of claim 3 , further comprising automatically executing the expanded query to provide user profile information including any authored documents used to mine profiles based on the expanded query.

5. The method of claim 1 , further comprising using the expertise mining algorithm and the first and second post-processing stages to rank and update author ranks by merging author ranking result sets from the first and second post-processing stages.

6. The method of claim 1 , further comprising extracting a list of authors from search results associated with the expertise mining query and using the expertise mining algorithm to compute a coefficient corresponding to a relation of each document author to the original query.

7. The method of claim 6 , further comprising updating author ranks by merging result sets from the first and second post-processing stages, and sorting the updated author ranks in an order by rank to obtain a final result set including one or more user profiles.

8. The method of claim 1 , further comprising limiting the execution of the expertise mining query to well-structured search items including word processing application files, spreadsheet application files, portable data files, and presentation application files.

9. The method of claim 1 , further comprising ranking authored files including documents based on results of the expertise mining query, extracting authors from a number of the ranked authored files, ranking the extracted authors, re-ranking the extracted authors using URL, author, and title tuples, and providing the expanded query using the one or more search terms and one or more re-ranked authors.

10. The method of claim 1 , returning authored files or links to the authored files as part of ranking authors having a level of expertise as part of providing substantive evidence pertaining to an expertise type.

11. The method of claim 1 , further comprising generating the expanded query using one or more query expertise expanders including one or more first and last author names.

12. The method of claim 1 , further comprising using metadata to identify primary and secondary authors for returned authored documents based in part on the expertise mining query, a total number of documents authored by each author, and an output associated with weighted sums of query term occurrences in a document title and URL portions of each ranked document.

13. The method of claim 1 , further comprising using an association of document metadata with the original query as part of determining that an associated author of the document is relevant to a particular expertise sought by the original query including ranking authors by:

R
′

⁡

(
A
)

→

∑

m
=

1
-

Top

Q
′

⁢

⁢

W
T

(

∑

n
=

1
-

Top

K
′

⁢

⁢

T
⁡

(

Q
m

⁢

D
nA

)

)

+

W
UL

(

∑

n
=

1
-

Top

K
′

Assignees

• Ray Aninda
• Meyerzon Dmitriy
• Khan Sana Fahim
• Alspaugh Jessica Anne
• Microsoft Technology Licensing Llc

Inventors

• Ray Aninda
• Meyerzon Dmitriy
• Khan Sana Fahim
• Alspaugh Jessica Anne

Classifications

• G06F16/90335 Primary

  Query processing · CPC title

• G06F17/30979 Primary

  Physics · mapped topic