Enterprise search method and system

Patent metadata

Field	Value
Publication number	US-9639609-B2
Application number	US-39148409-A
Country	US
Kind code	B2
Filing date	Feb 24, 2009
Priority date	Feb 24, 2009
Publication date	May 2, 2017
Grant date	May 2, 2017

Abstract

A system and method for enterprise search includes one or more computer-readable media storing computer-executable instructions that, when executed on one or more processors that perform acts including extracting one or more of term data, personal data and metadata from one or more predetermined resources; retrieving a set of information derived from the extracted term data, personal data and metadata responsive to a query; and receiving feedback responsive to the set of information, the feedback augmenting at least one of the one or more predetermined resources.

First claim

The invention claimed is:

1. A system comprising:
one or more processors;
memory coupled to the one or more processors; and
a FAQ extraction module stored in the memory and executable by the one or more processors to:
collect candidate files based at least in part on locating one or more keywords associated with frequently asked questions (FAQs) in the candidate files,
extract candidate patterns for the FAQs from the candidate files based at least in part on one or more rules associated with the FAQs,
validate the candidate patterns for the FAQs, validating the candidate patterns comprising:
obtaining a plurality of candidate units based at least in part on the one or more rules,
grouping the plurality of candidate units into a plurality of groups based at least in part on the candidate patterns, one or more candidate units in a group of the plurality of groups corresponding to a same candidate pattern,
selecting first N groups from the plurality of groups, N being an integer greater than zero, and
validating one or more candidate patterns associated with the first N groups based at least in part on a result of a comparison between a number of candidate units corresponding to the one or more candidate patterns and a predetermined number, and

extract one or more pairs of questions and answers from the candidate files after the validation of the candidate patterns; and
an online search module stored in the memory and executable by the one or more processors to provide a search result in response to receiving a query, the search result comprising a plurality of sections showing a plurality of result types, with at least one section showing a result type different from one or more result types of other sections, wherein the result type comprises information derived from one or more of FAQ data, term data, personal data or metadata, and the one or more result types comprises information derived from one or more others of the FAQ data, the term data, the personal data or the metadata.

2. The system of claim 1 , further comprising an offline extraction module stored in the memory and executable by the one or more processors to extract at least one of the metadata, the term data or the personal data.

3. The system of claim 2 , further comprising:
a personal information resource database coupled to the offline extraction module, the personal information resource database including data extracted from an enterprise personal information source;
an intranet source coupled to the offline extraction module, the intranet source for providing documents for data extraction; and
a term resource database coupled to the offline extraction module, the term resource database for receiving user feedback and term information data extracted from one or more sources.

4. The system of claim 2 , further comprising a knowledge database coupled to the online search module, the knowledge database to provide a set of information searchable by the online search module, wherein the knowledge database includes a metadata database, the metadata database configured to store metadata derived from documents including one or more of title, author and date extracted by the offline extraction module.

5. The system of claim 2 , wherein the offline extraction module includes a homepage extraction module configured to extract homepage data using a uniform resource locator (URL).

6. The system of claim 1 , wherein the plurality of result types comprise at least two types associated with: a homepage, an authored document, a person, a definition or a frequently asked question.

7. One or more memory devices storing computer-executable instructions that, when executed by one or more processors, perform acts comprising:
collecting candidate files based at least in part on locating one or more keywords associated with frequently asked questions (FAQs) in the candidate files,
extracting candidate patterns for the FAQs from the candidate files based at least in part on one or more rules associated with the FAQs,
validating the candidate patterns for the FAQs, validating the candidate patterns comprising:
obtaining a plurality of candidate units based at least in part on the one or more rules,
grouping the plurality of candidate units into a plurality of groups based at least in part on the candidate patterns, one or more candidate units in a group of the plurality of groups corresponding to a same candidate pattern,
selecting first N groups from the plurality of groups, N being an integer greater than zero, and
validating one or more candidate patterns associated with the first N groups based at least in part on a result of a comparison between a number of candidate units corresponding to the one or more candidate patterns and a predetermined number, and

extracting one or more pairs of questions and answers from the candidate files after the validation of the candidate patterns; and
providing a search result in response to receiving a query, the search result comprising a plurality of sections showing a plurality of result types, with at least one section showing a result type different from one or more result types of other sections, wherein the result type comprises information derived from one or more of FAQ data, term data, personal data or metadata, and the one or more result types comprises information derived from one or more others of the FAQ data, the term data, the personal data or the metadata.

8. The one or more memory devices of claim 7 , wherein the plurality of result types comprise at least two types associated with: a homepage, an authored document, a person, a definition or a frequently asked question.

9. The one or more memory devices of claim 7 , the acts further comprising extracting at least one of the metadata, the term data or the personal data.

10. The one or more memory devices of claim 7 , the acts further comprising:
receiving an input of additional information for the search result for augmenting one or more predetermined resources related to the search result; and
augmenting the one or more predetermined resources based at least in part on the input of the additional information.

11. The one or more memory devices of claim 10 , wherein the one or more predetermined resources include one or more of a FAQ resource, an active directory resource, an intranet resource or a term resource.

12. The one or more memory devices of claim 10 , wherein receiving the input of the additional information comprises receiving feedback as a correction to the search result, a vote associated with the search result, an addition to the search result, and/or an indication of a quality of the search result.

13. The one or more memory devices of claim 7 , the acts further comprising causing a display of a collaborative interface for receiving an input of additional information, the collaborative interface enabling one or more users to edit, delete, modify and/or replace the search result.

14. The one or more memory devices of claim 7 , the acts further comprising enabling a selection between receiving an organized search result responsive to the query and receiving prior user feedback associated with the query, the prior user feedback being previously provided by one or more users.

15. A method implemented by one or more computing devices, the method comprising:
collecting candidate files based at least in part on locating one or more keywords associated with frequently asked questions (FAQs) in the candidate files,
extracting candidate patterns for the FAQs from the candidate files based at least in part on one or more rules associated

Assignees

• Li Hang
• Hu Yunhua
• Zou Xin
• Cui Xiaoyuan
• Gao Guangping
• Meyerzon Dmitriy
• Poznanski Victor
• Microsoft Technology Licensing Llc

Inventors

• Li Hang
• Hu Yunhua
• Zou Xin
• Cui Xiaoyuan
• Gao Guangping
• Meyerzon Dmitriy
• Poznanski Victor

Classifications

• G06F17/30864 Primary

  Physics · mapped topic

• G06F16/248

  Presentation of query results · CPC title

• G06F16/24578

  using ranking · CPC title

• G06F16/24575

  using context · CPC title

• G06F16/951 Primary

  Indexing; Web crawling techniques · CPC title