Computer-implemented system and method for analyzing search queries

• US 8,326,817 B2
• Filed: 05/13/2009
• Issued: 12/04/2012
• Est. Priority Date: 09/07/2004
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Koch, Donald O., Pope, David C., Scott, Kevin L.
• Primary Examiner(s): TRUONG, CAM Y T
• Application Number:
  US12/465,408
• Publication Number:
  US 20090222426A1
• Time in Patent Office: 1,301 Days
• Field of Search: 707/706, 707/770, 707/708
• US Class Current: 707/708
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9532 Query formulation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

Systems and methods for handling input search engine data. A data store is searched based upon one or more data items contained within the input search engine data. The result of searching the data store is used to determine which data items in the input search engine data might be a search term. Search terms are extracted based upon the result of searching the data store.

Claims

1. A computer-implemented method of extracting search terms from a log, comprising:

accessing, using one or more data processors, a log containing a plurality of search strings compiled from previously executed searches submitted to a plurality of different known and new search engines;

parsing, using the one or more data processors, a search string from the log containing the plurality of search strings from previously executed searches to identify one or more name-value pairs within the search string, wherein a name-value pair is associated with a variable name and a variable value;

searching, using the one or more data processors, a non-search term table to identify the variable name or the variable value associated with each name-value pair, wherein the non-search term table includes variable names or variable values known as non-search terms for known search engines;

determining, using the one or more data processors, that a variable value associated with a name-value pair is not a search term when the associated variable name or the variable value is located in the non-search term table, wherein name-value pairs having associated variable names and variable values that are not located in the non-search term table are remaining name-value pairs;

searching, using the one or more data processors, a search term table to identify the variable name or the variable value associated with each remaining name-value pair, wherein the search term table includes variable names or variable values known as search terms for known search engines;

determining, using the one or more data processors, that a variable value associated with a remaining name-value pair is a search term when the associated variable name or the variable value is located in the search term table, wherein remaining name-value pairs having associated variable names and variable values that are not located in the search term table are unknown name-value pairs; and

applying, using the one or more data processors, a name-value pair test to determine whether the variable values associated with each unknown name-value pair are search terms, wherein the name-value pair test identifies search terms associated with new search engines.

2. The method of claim 1, further comprising:

storing variable values in a computer-readable medium when those variable values are determined to be search terms.

3. The method of claim 1, wherein at least two of the different known and new search engines use different formats and identifiers for search terms in a search string.

4. The method of claim 1, wherein the non-search term table and the search term table comprise hash tables.

5. The method of claim 1, wherein one of the search strings is formatted according to a hypertext transfer protocol.

6. The method of claim 1, wherein one of the search strings includes a name-value pair associated with a referrer.

7. The method of claim 1, wherein the name-value pair test is a test to determine whether a variable value is longer than a threshold length.

8. The method of claim 1, wherein the name-value pair test is a test to determine whether a variable value includes only numbers and symbols.

9. The method of claim 1, wherein a particular variable name is added to the search term table when a particular variable value associated with the particular variable name is determined to be a search term.

10. The method of claim 1, wherein the log comprises click-stream data generated from one or more Internet web page accesses.

11. The method of claim 1, wherein the log comprises click-stream referrer data.

12. The method of claim 1, wherein the variable values determined to be search terms are presented in a searching profile report.

13. The method of claim 12, wherein the report is provided to an e-business web site manager.

14. The method of claim 1, wherein the name-value pair test includes searching for a variable value associated with an unknown name-value pair in a variable value table.

15. The method of claim 1, wherein the log is associated with a weblog.

16. The method of claim 1, further comprising:

identifying a referrer name-value pair in one of the search strings;

locating a search term name-value pair based upon the identified referrer; and

adding a variable value associated with the search term name-value pair to the search term table.

17. A computer-implemented system for extracting search terms from a log, comprising:

one or more data processors;

one or more computer-readable storage mediums containing instructions configured to cause the one or more processors to perform operations including;

accessing a log containing a plurality of search strings compiled from previously executed searches submitted to a plurality of different known and new search engines;

parsing a search string to identify one or more name-value pairs within the search string from the log containing the plurality of search strings from previously executed searches, wherein a name-value pair is associated with a variable name and a variable value;

searching a non-search term table to identify the variable name or the variable value associated with each name-value pair, wherein the non-search term table includes variable names or variable values known as non-search terms for known search engines;

determining that a variable value associated with a name-value pair is not a search term when the associated variable name or the variable value is located in the non-search term table, wherein name-value pairs having associated variable names and variable values that are not located in the non-search term table are remaining name-value pairs;

searching a search term table to identify the variable name or the variable value associated with each remaining name-value pair, wherein the search term table includes variable names or variable values known as search terms for known search engines;

determining that a variable value associated with a remaining name-value pair is a search term when the associated variable name or the variable value is located in the search term table, wherein remaining name-value pairs having associated variable names and variable values that are not located in the search term table are unknown name-value pairs; and

applying a name-value pair test to determine whether the variable values associated with each unknown name-value pair are search terms, wherein the name-value pair test identifies search terms associated with new search engines.

18. A computer-program product for extracting search terms from a log, tangibly embodied in a machine-readable non-transitory storage medium, including instructions configured to cause a data processing apparatus to:

access a log containing a plurality of search strings compiled from previously executed searches submitted to a plurality of different known and new search engines;

parse a search string to identify one or more name-value pairs within the search string from the log containing the plurality of search strings from previously executed searches, wherein a name-value pair is associated with a variable name and a variable value;

search a non-search term table to identify the variable name or the variable value associated with each name-value pair, wherein the non-search term table includes variable names or variable values known as non-search terms for known search engines;

determine that a variable value associated with a name-value pair is not a search term when the associated variable name or the variable value is located in the non-search term table, wherein name-value pairs having associated variable names and variable values that are not located in the non-search term table are remaining name-value pairs;

search a search term table to identify the variable name or the variable value associated with each remaining name-value pair, wherein the search term table includes variable names or variable values known as search terms for known search engines;

determine that a variable value associated with a remaining name-value pair is a search term when the associated variable name or the variable value is located in the search term table, wherein remaining name-value pairs having associated variable names and variable values that are not located in the search term table are unknown name-value pairs; and

apply a name-value pair test to determine whether the variable values associated with each unknown name-value pair are search terms, wherein the name-value pair test identifies search terms associated with new search engines.