System and method to provide query linguistic service

• US 8,949,265 B2
• Filed: 03/05/2010
• Issued: 02/03/2015
• Est. Priority Date: 03/05/2009
• Status: Active Grant

• Current Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Original Assignee: eBay Inc.
• Inventors: Mauge, Karin, Petranov, Radoslav Valentinov, Ruvini, Jean-David, Statelova, Antoniya T., Sundaresan, Neelakantan
• Primary Examiner(s): HASAN, SYED HAROON
• Application Number:
  US12/718,929
• Publication Number:
  US 20100228762A1
• Time in Patent Office: 1,796 Days
• Field of Search: 707/771, 707/E17.015, 707/E17.072, 707/766
• US Class Current: 707/766
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/3338 Query expansion
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms

Abstract

In various example embodiments, a system and method to provide query linguistic service is provided. An initial query term set is received. Phrase recognition is performed on the initial query term set to determine recognized phrases. Using the determined recognized phrases, one or more synonyms for each of the recognized phrases are determined. Results matching the initial query term set and any selected synonyms from the determined one or more synonyms are determined.

Claims

1. A method comprising:

receiving an initial query term set;

based on the initial query term set determining a list of item categories that each represent a category of items available on a network;

causing the list of item categories to be presented in a user interface to a user;

receiving from the user, a selection of an item category from the list of item categories determined based on the initial query term set;

performing phrase recognition on the initial query term set to determine recognized phrases;

determining a synonym for each of the recognized phrases based on the user selected item category, the synonym comprising a plurality of words;

causing the determined synonyms to be presented in the user interface to the user along with corresponding first selectors and corresponding second selectors, each of the first selectors being operable to select the corresponding synonym, each of the second selectors being operable to select a search option for the corresponding synonym, each second selector including;

a first search option operable to cause the corresponding synonym to be matched whenever the plurality of words of the synonym are matched without regard to order,

a second search option operable to cause the corresponding synonym to be matched whenever at least one of the plurality of words of the synonym is matched, and

a third search option operable to cause the corresponding synonym to be matched whenever the plurality of words of the synonym are matched in order;

receiving, from the user, a selection of a subset of the presented synonyms selected by the user from the determined synonyms, the selection indicating corresponding search options selected by the user; and

determining, using one or more processors, results within the user selected item category that match the initial query term set and the selected subset of synonyms, the results being based on the selected search options for the selected subset of synonyms.

2. The method of claim 1, further comprising returning an entire tree for each item category of the list of item categories based on the initial query term set from which the selection of the item category is made.

3. The method of claim 1, wherein the performing of the phrase recognition comprises adding one word at a time to a blank phrase until a matching recognized phrase is found.

4. The method of claim 1, wherein the performing of the phrase recognition comprises removing One word at a time from the initial query term set until a matching recognized phrase is found.

5. The method of claim 1, wherein the performing of the phrase recognition comprises checking if a window of a number of words constitutes a matching recognized phrase.

6. The method of claim 1, further comprising:

analyzing a number of the results to determine terms most frequently encountered in the number of the results;

receiving a refinement input based on the terms most frequently encountered; and

refining the results based on the refinement input.

7. The method of claim 6, wherein the refinement input comprises a term of the terms most frequently encountered to exclude from the results when refining the results.

8. The method of claim 1, wherein the item category is associated with a good or service.

9. The method of claim 1, wherein the list of item categories comprises a top number of most commonly viewed categories corresponding to the initial query term set.

10. A system comprising:

a processor of a machine;

a query tool module to receive an initial query term set;

a category identification module configured to;

determine a list of item categories that each represents a category of items available on a network based on the initial query term set, and receive, from a user, a selection of an item category from the list of item categories determined based on the initial query term set, the selection being received via a user interface presenting the list of items to the user;

a synonym lookup module configured to perform phrase recognition on the initial query term set to determine a recognized phrase and to determine a synonym for the recognized phrase based on the user selected item category, the synonym comprising a plurality of words;

the query tool module being further configured to receive from the user a selection of the determined synonym along with a search option for the synonym, the search option comprising one of;

a first search option requesting that the synonym to be matched whenever the plurality of words of the synonym are matched without regard to order,

a second search option requesting that the synonym to be matched whenever at least one of the plurality of words of the synonym is matched, and

a third search option requesting that the synonym to be matched whenever the plurality of words of the synonym are matched in order; and

a listing finder module configured to determine, using the processor, results within the user selected item category that match the initial query term set and the synonym, the results being based on the search option for the synonym.

11. The system of claim 10, wherein the listing finder module is further to analyze a number of the results to determine terms most frequently encountered in the number of the results, to receive a refinement input based on the terms most frequently encountered, and to refine the results based on the refinement input.

12. A non-transitory machine-readable storage medium in communication with at least one processor, the machine-readable storage medium storing instructions which, when executed by the at least one processor, perform operations comprising:

receiving an initial query term set;

based on the initial query term set, determining a list of item categories that each represent a category of items available on a network;

causing the list of item categories to be presented in a user interface to a user;

receiving, from the user, a selection of an item category from the list of item categories determined based on the initial query term set;

performing phrase recognition on the initial query term set to determine recognized phrases;

determining one or more synonyms for each of the recognized phrases based on the user selected item category, each of the one or more synonyms comprising a plurality of words;

causing the determined synonyms to be presented in the user interface to the user along with corresponding first selectors and corresponding second selectors, each of the first selectors being operable to select the corresponding synonym, each of the second selectors being operable to select a search option for the corresponding synonym, each second selector including;

a first search option operable to cause the corresponding synonym to be matched whenever the plurality of words of the synonym are matched without regard to order,

a second search option operable to cause the corresponding synonym to be matched whenever at least one of the plurality of words of the synonym is matched, and

a third search option operable to cause the corresponding synonym to be matched whenever the plurality of words of the synonym are matched in order;

receiving, from the user, a selection of a subset of the presented synonyms selected by the user from the determined synonyms, the selection indicating corresponding search options selected by the user; and

determining, using one or more processors, results within the user selected item category that match the initial query term set and the selected subset of synonyms, the results being based on the selected search options for the selected subset of synonyms.

13. The machine-readable storage medium of claim 12, wherein the performing of the phrase recognition comprises adding one word at a time to a blank phrase until a matching recognized phrase is found.

14. The machine-readable storage medium of claim 12, wherein the performing of the phrase recognition comprises removing one word at a time from the initial query term set until a matching recognized phrase is found.

15. The machine-readable storage medium of claim 12, wherein the performing of the phrase recognition comprises checking if a window of n words constitutes a matching recognized phrase.

16. The machine-readable storage medium of claim 12, wherein the method further comprises

analyzing a number of the results to determine terms most frequently encountered in the number of the results;

receiving a refinement input based on the terms most frequently encountered; and

refining the results based on the refinement input.

17. The machine-readable storage medium of claim 16, wherein the refinement input comprises a term of the terms most frequently encountered to exclude from the results when refining the results.

18. The machine-readable storage medium of claim 12, wherein the determining of the results further comprises applying an advanced query syntax input.