Method, apparatus, and computer program product for classifying user search data

• US 9,947,024 B1
• Filed: 06/26/2014
• Issued: 04/17/2018
• Est. Priority Date: 10/04/2012
• Status: Active Grant

• Current Assignee: Groupon Incorporated
• Original Assignee: Groupon Incorporated
• Inventors: Shariff, Shafiq, Nordquist, Derek
• Primary Examiner(s): Gills, Kurtis
• Application Number:
  US14/316,228
• Time in Patent Office: 1,391 Days
• Field of Search: 705 731
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0259 based on store location
• CPC Class Codes: H05K 999/99 dummy group

Abstract

Provided herein are systems, methods and computer readable media for classifying user search data. An example method comprises accessing user search data, the user search data generated by a user interacting with a promotion and marketing service to identify a requested promotion, determining, using a processor, a first classification of the user search data by parsing the user search data into one or more terms, applying one or more terms within the user search data to a search index, wherein the first classification is configured to generate a mapping from the one or more terms within the user search data to at least one of a service or merchant, determining, using the processor, a second classification of the user search data, wherein the second classification identifies a requested redemption location for the requested promotion and generating an identification pair for the search data that is representative of a promotion tuple, the identification pair comprising the first classification and the second classification.

Claims

1. A method for utilizing user search data to infer consumer demand for a presently unavailable product or service, the method comprising:

accessing user search data, the user search data generated by a user interacting with a promotion and marketing service to identify a requested promotion, the user search data comprising data captured in the form of characters input by a consumer and data in the form of a browsing pattern,

the user search data generated by;

identifying mobile application and web browser activity of a consumer, the mobile application and web browser activity being identifying clicking of a link, selecting particular text or image data, and inputting or entering of data; and

storing the mobile application and web browser activity of the consumer as search data;

determining, using a first classifier, a first classification of the user search data, the first classification indicative of at least one of (i) a category or subcategory;

or (ii) one or more services offered by a merchant, and determining, using a second classifier, a second classification of the user search data,

wherein each of the first classifier and the second classifier is a trainable classifier adapted using a supervised learning method, the first classifier and the second classifier adapted based on a training data set,

wherein the determination of the first classification performed by;

(1) parsing the input or entered user search data into one or more terms;

(2) utilizing click data and selection data of the user search data and location data to identify a correct merchant, increasing accuracy of a merchant level search classification; and

(3) applying the one or more terms within the user search data to a search index, wherein the first classification is configured to generate a mapping from the one or more terms within the user search data to at least one of (1) the category or subcategory;

or (2) the one or more services offered by a merchant, wherein the second classification identifies a requested redemption location for the requested promotion, the request location indicative of where the requested promotion is to be redeemed, and

wherein the first classifier and the second classifier are trained by providing positive samples extracted from searched attributes annotated as relevant to a particular attribute and providing negative samples extracted from searched attributes annotated as relevant to a parent of the particular attribute or not relevant to the particular attribute; and

generating an identification pair for the search data that is representative of a promotion tuple, the identification pair comprising the first classification indicative of (1) the category or subcategory or (2) the one or more services offered by a merchant, and the second classification indicative of the requested redemption location.

2. The method of claim 1, further comprising:

aggregating a plurality of generated identification pairs for a predetermined time period;

identifying a geographic area;

generating a real time demand indication for a geographic area by ranking the first classification of the plurality that have a second classification within the geographic area; and

displaying the real time demand indication.

3. The method of claim 1, wherein the search index comprises at least a taxonomy search index or a merchant search index.

4. The method of claim 3, further comprising:

generating the taxonomy search index, the taxonomy search index comprising a mapping of one or more terms with at least one category or sub-category defined by a taxonomy; and

generating the merchant search index, the merchant search index comprising a mapping of one or more terms to at least one merchant.

5. The method of claim 3, wherein the taxonomy is generated by:

inputting a service corpus, the service corpus comprising one or more terms and one or more services defined by a taxonomy;

analyzing one or more click events in the promotion and marketing system, the one or more click events received in response to displaying one or more promotions that correspond to the one or more terms, wherein the one or more click events identify a selected promotion;

identifying one or more services for the selected promotion; and

generating a mapping between the one or more terms and the one or more services.

6. The method of claim 3, wherein the taxonomy is generated by:

inputting a merchant corpus, the merchant corpus comprising one or more terms and one or more merchants that are capable of offering at least one of goods, services or experiences;

analyzing one or more click events in the promotion and marketing system, the one or more click events received in response to displaying one or more promotions that correspond to the one or more terms, wherein the one or more click events identify a user selected promotion;

identifying one or more merchants that are offering the selected promotion; and

generating a mapping between the one or more terms and the one or more merchants.

7. The method of claim 1, wherein determining the second classification further comprises:

receiving a latitude longitude pair generated at the time of the user search data; and

assigning a hyper location region based on the latitude longitude pair.

8. The method of claim 1, wherein the second classification further comprises:

receiving a latitude longitude pair generated at the time of the user search data; and

assigning a hyper location region as the redemption location based on the latitude longitude pair.

9. The method of claim 1, wherein determining the second classification further comprises:

comparing the one or more terms to a location index, wherein the location index is trained based on one or more geographical identifiers of a geographic area;

determining that at least one of the one or more terms contains a geographical identifier; and

assigning a hyper location region as the redemption location based on the geographical identifier.

10. A computer program product for utilizing user search data to infer consumer demand for a presently unavailable product or service, the computer program product comprising at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein, the computer-executable program code instructions comprising program code instructions for:

accessing user search data, the user search data generated by a user interacting with a promotion and marketing service to identify a requested promotion, the user search data comprising data captured in the form of characters input by a consumer and data in the form of a browsing pattern,

the user search data generated by;

identifying mobile application and web browser activity of a consumer, the mobile application and web browser activity being identifying clicking of a link, selecting particular text or image data, and inputting or entering of data; and

storing the mobile application and web browser activity of the consumer as search data;

determining, using a first classifier, a first classification of the user search data, the first classification indicative of at least one of (i) a category or subcategory;

or (ii) one or more services offered by a merchant, and determining, using a second classifier, a second classification of the user search data,

wherein each of the first classifier and the second classifier is a trainable classifier adapted using a supervised learning method, the first classifier and the second classifier adapted based on a training data set,

wherein the determination of the first classification performed by;

(1) parsing the input or entered user search data into one or more terms;

(2) utilizing click data and selection data of the user search data and location data to identify a correct merchant, increasing accuracy of a merchant level search classification; and

(3) applying the one or more terms within the user search data to a search index, wherein the first classification is configured to generate a mapping from the one or more terms within the user search data to at least one of (1) the category or subcategory;

or (2) the one or more services offered by a merchant,

wherein the second classification identifies a requested redemption location for the requested promotion, the request location indicative of where the requested promotion is to be redeemed, and

wherein the first classifier and the second classifier are trained by providing positive samples extracted from searched attributes annotated as relevant to a particular attribute and providing negative samples extracted from searched attributes annotated as relevant to a parent of the particular attribute or not relevant to the particular attribute; and

generating an identification pair for the search data that is representative of a promotion tuple, the identification pair comprising the first classification indicative of (1) the category or subcategory or (2) the one or more services offered by a merchant, and the second classification indicative of the requested redemption location.

11. The computer program product according to claim 10, wherein the computer-executable program code instructions further comprise program code instructions for:

aggregating a plurality of generated identification pairs for a predetermined time period;

identifying a geographic area;

generating a real time demand indication for a geographic area by ranking the first classification of the plurality that have a second classification within the geographic area; and

displaying the real time demand indication.

12. The computer program product of claim 10, wherein the search index comprises at least a taxonomy search index or a merchant search index.

13. The computer program product according to claim 12, wherein the computer-executable program code instructions further comprise program code instructions for:

generating the taxonomy search index, the taxonomy search index comprising a mapping of one or more terms with at least one category or sub-category defined by a taxonomy; and

generating the merchant search index, the merchant search index comprising a mapping of one or more terms to at least one merchant.

14. The computer program product of claim 13, wherein the computer-executable program code instructions for generating the taxonomy comprise program code instructions for:

inputting a service corpus, the service corpus comprising one or more terms and one or more services defined by a taxonomy;

analyzing one or more click events in the promotion and marketing system, the one or more click events received in response to displaying one or more promotions that correspond to the one or more terms, wherein the one or more click events identify a selected promotion;

identifying one or more services for the selected promotion; and

generating a mapping between the one or more terms and the one or more services.

15. The computer program product of claim 13, wherein the computer-executable program code instructions for generating the taxonomy comprise program code instructions for:

inputting a merchant corpus, the merchant corpus comprising one or more terms and one or more merchants that are capable of offering at least one of goods, services or experiences;

analyzing one or more click events in the promotion and marketing system, the one or more click events received in response to displaying one or more promotions that correspond to the one or more terms, wherein the one or more click events identify a user selected promotion;

identifying one or more merchants that are offering the selected promotion; and

generating a mapping between the one or more terms and the one or more merchants.

16. The computer program product of claim 10, wherein the computer-executable program code instructions for determining the second classification comprise program code instructions for:

receiving a latitude longitude pair generated at the time of the user search data; and

assigning a hyper location region based on the latitude longitude pair.

17. The computer program product of claim 10, wherein the computer-executable program code instructions for determining the second classification comprise program code instructions for:

receiving a latitude longitude pair generated at the time of the user search data; and

assigning a hyper location region as the redemption location based on the latitude longitude pair.

18. The computer program product of claim 10, wherein the computer-executable program code instructions for determining the second classification comprise program code instructions for:

comparing the one or more terms to a location index, wherein the location index is trained based on one or more geographical identifiers of a geographic area;

determining that at least one of the one or more terms contains a geographical identifier; and

assigning a hyper location region as the redemption location based on the geographical identifier.

19. An apparatus comprising at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the processor, cause the apparatus to at least:

access user search data, the user search data generated by a user interacting with a promotion and marketing service to identify a requested promotion, the user search data comprising data captured in the form of characters input by a consumer and data in the form of a browsing pattern,

the user search data generated by;

identifying mobile application and web browser activity of a consumer, the mobile application and web browser activity being identifying clicking of a link, selecting particular text or image data, and inputting or entering of data; and

storing the mobile application and web browser activity of the consumer as search data;

determine, using a first classifier, a first classification of the user search data, the first classification indicative of at least one of (i) a category or subcategory;

or (ii) one or more services offered by a merchant, and determining, using a second classifier, a second classification of the user search data,

wherein each of the first classifier and the second classifier is a trainable classifier adapted using a supervised learning method, the first classifier and the second classifier adapted based on a training data set,

wherein the determination of the first classification performed by;

(1) parsing the input or entered user search data into one or more terms; and

(2) utilizing click data and selection data of the user search data and location data to identify a correct merchant, increasing accuracy of a merchant level search classification; and

(3) applying the one or more terms within the user search data to a search index, wherein the first classification is configured to generate a mapping from the one or more terms within the user search data to at least one of (1) the category or subcategory;

or (2) the one or more services offered by a merchant,

wherein the second classification identifies a requested redemption location for the requested promotion, the request location indicative of where the requested promotion is to be redeemed, and

wherein the first classifier and the second classifier are trained by providing positive samples extracted from searched attributes annotated as relevant to a particular attribute and providing negative samples extracted from searched attributes annotated as relevant to a parent of the particular attribute or not relevant to the particular attribute; and

generate an identification pair for the search data that is representative of a promotion tuple, the identification pair comprising the first classification indicative of (1) the category or subcategory or (2) the one or more services offered by a merchant, and the second classification indicative of the requested redemption location.

20. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to:

aggregate a plurality of generated identification pairs for a predetermined time period;

identify a geographic area;

generate a real time demand indication for a geographic area by ranking the first classification of the plurality that have a second classification within the geographic area; and

display the real time demand indication.

21. The apparatus of claim 19, wherein the search index comprises at least a taxonomy search index or a merchant search index.

22. The apparatus of claim 21, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to:

generate the taxonomy search index, the taxonomy search index comprising a mapping of one or more terms with at least one category or sub-category defined by a taxonomy; and

generate the merchant search index, the merchant search index comprising a mapping of one or more terms to at least one merchant.

23. The apparatus of claim 22, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to generate the taxonomy by:

input a service corpus, the service corpus comprising one or more terms and one or more services defined by a taxonomy;

analyze one or more click events in the promotion and marketing system, the one or more click events received in response to displaying one or more promotions that correspond to the one or more terms, wherein the one or more click events identify a selected promotion;

identify one or more services for the selected promotion; and

generate a mapping between the one or more terms and the one or more services.

24. The apparatus of claim 22, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to generate the taxonomy by:

input a merchant corpus, the merchant corpus comprising one or more terms and one or more merchants that are capable of offering at least one of goods, services or experiences;

analyze one or more click events in the promotion and marketing system, the one or more click events received in response to displaying one or more promotions that correspond to the one or more terms, wherein the one or more click events identify a user selected promotion;

identify one or more merchants that are offering the selected promotion; and

generate a mapping between the one or more terms and the one or more merchants.

25. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to determine the second classification by:

receive a latitude longitude pair generated at the time of the user search data; and

assign a hyper location region based on the latitude longitude pair.

26. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to determine the second classification by:

receive a latitude longitude pair generated at the time of the user search data; and

assign a hyper location region as the redemption location based on the latitude longitude pair.

27. The apparatus of claim 19, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to determine the second classification by:

compare the one or more terms to a location index, wherein the location index is trained based on one or more geographical identifiers of a geographic area;

determine that at least one of the one or more terms contains a geographical identifier; and

assign a hyper location region as the redemption location based on the geographical identifier.