Method, system, and computer-readable medium for product and vendor selection

• US 10,296,956 B2
• Filed: 01/14/2015
• Issued: 05/21/2019
• Est. Priority Date: 01/14/2015
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP SE
• Inventors: Shi, Zhan, Cheng, Yu, Sun, Gufei, Li, Wen-Syan
• Primary Examiner(s): Smith, Jeffrey A.
• Assistant Examiner(s): Duraisamygurusamy, Lalith M
• Application Number:
  US14/596,219
• Publication Number:
  US 20160203540A1
• Time in Patent Office: 1,588 Days
• Field of Search: 705 2662
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06Q 30/0625 Directed, with specific intent or strategy

Abstract

Disclosed herein are technologies for providing recommendations as to particular products and/or services that are customer specific and general, based on customer preference and inquiry. The recommendations are provided as part of an online shopping system. In accordance with one aspect, an item query is received from a customer, and analyzed by a query analyzer to determine if the query is a general item query or a specific item query. A search may be performed for items based on the item query in an items database listing items offered for purchase. If the query is the general item query, customer preference is determined from results of the search. If the query is the specific item query, the items from the results of the search are grouped based on cost performance. The items of the search result are ranked and provided to the customer.

Claims

1. A computer-implemented method for providing personalized search in an online shopping platform comprising:

receiving a query from a customer at the online shopping platform;

analyzing the query by a query analyzer to determine a type of the query, wherein the type of the query is selected from a group consisting of a general item query and a specific item query;

performing a search for items based on the query in an items database listing items offered for purchase, wherein performing the search includes, based on determining that the query is the specific item query, categorizing each of the items from results of the search into a respective one of a plurality of cost performance categories based on cost performance for each item, wherein the cost performance for a respective item includes a price-quality relationship indicating a proportion of price to quality for the respective item;

retrieving most relevant items as a search result from the items database based on the performed search and the type of the query;

ranking the most relevant items of the search result based at least partly on the respective cost performance categories of the most relevant items, including ranking the most relevant items based at least partly on the price-quality relationship indicating the proportion of price to quality for each of the most relevant items;

providing the ranked items to the customer at a user interface of a device connected to the online shopping platform;

storing data for user off-line and online behavior in a behavior log at the items database, wherein the data includes click-through data that indicates which of the provided ranked items the customer clicked on, and wherein the click-through data includes data for selection interactions of the customer after providing the ranked items as the search result;

optimizing retrieval of search results for searches requested by the customer based on the stored data for user off-line and online behavior to determine an optimized search result; and

initiating an optimized search to determine a different set of items compared to the ranked items, wherein the optimized search is associated with the query.

2. The computer-implemented method of claim 1 wherein:

performing the search using the general item query comprises performing a general item search by a general retrieval module; and

performing the search using the specific item query comprises performing a specific item search by a specific retrieval module.

3. The computer-implemented method of claim 2 wherein performing the general item search comprises:

retrieving general search items from the items database based on the general item query;

determining customer preference by a behavior analyzer using the behavior log of the customer; and

ranking the general search items by a personalized ranking function taking account of the customer preference.

4. The computer-implemented method of claim 3 wherein determining customer preference comprises:

deriving an appropriate price interval for the customer from the behavior log of the customer; and

generating Gaussian weight values for general search items.

5. The computer-implemented method of claim 4 wherein ranking comprises:

calculating ranking scores for the general search items using the personalized ranking function and the Gaussian weight values; and

ranking the items based on the ranking scores of the items.

6. The computer-implemented method of claim 4 further comprising:

adjusting the appropriate price interval using real-time data.

7. The computer-implemented method of claim 6 wherein the real-time data comprises click through data of the customer.

8. The computer-implemented method of claim 2 wherein performing the specific item search comprises:

retrieving specific search items from the items database based on the specific item query;

determining a distribution of price-quality relationship of the specific search items; and

grouping the specific search items into groups based on cost performance using the distribution.

9. The computer-implemented method of claim 8 wherein:

items in the items database comprise prices of the items;

the online shopping platform comprises a quality rating of items in the items database; and

determining the distribution of price-quality relationship takes into account of quality and pricing of the items.

10. The computer-implemented method of claim 9 wherein grouping comprises grouping items into low cost performance, moderate cost performance and high cost performance groups.

11. The computer-implemented method of claim 8 further comprising ranking specific search items within each group based on customer preference.

12. The computer-implemented method of claim 1 wherein the database stores items in an items table comprising:

category fields for categorizing items; and

a price field containing a price of the items.

13. The computer-implemented method of claim 1 wherein:

items in the items database comprise category levels to facilitate search using general and specific item queries; and

the category levels comprise item level, category level, name level and brand level.

14. An online shopping platform comprising:

a query analyzer for receiving a search query from a customer, the query analyzer analyzes the query to determine a type of query, wherein the type of the query is selected from a group consisting of a general item query and a specific item query;

an items database storing items offered for sale;

a general retrieval module comprising;

a user behavior component for determining customer preference of the customer,

a personalized ranking model component for ranking results retrieved from the items database, and

a user behavior log to store data for user off-line and online behavior, wherein the data includes click-through data that indicates which of the provided ranked items the customer clicked on, and wherein the click-through data includes data for selection interactions of the customer after providing the ranked items as the search result;

wherein the general retrieval module retrieves relevant items as a search result from the items database based on the performed search and the type of executed query;

wherein the general retrieval module searches the items database and returns a general search result containing general search items based on the general item query, the general search result items are ranked based on the customer preference to generate a personalized ranked general search result;

a specific retrieval module comprising an item price-quality relationship component, wherein the specific retrieval module searches the items database and returns a specific search result containing specific search items based on the specific item query, wherein each of the specific search items are categorized, based on cost performance for each item, into a respective one of a plurality of cost performance categories, wherein the cost performance for a respective item includes a price-quality relationship indicating a proportion of price to quality for the respective item; and

a user interface module to provide ranked items to the customer, wherein the user interface is provided on a device connected to the online shopping platform, wherein the items are ranked based at least partly on the respective cost performance categories of the items, including ranking the items based at least partly on the price-quality relationship indicating the proportion of price to quality for each of the items;

wherein the general retrieval module is operable to optimize retrieval of search results for performed searches based on the stored data for user off-line and online behavior at the user behavior log and to determine an optimized search result, wherein the general retrieval module is operable to initiate an optimized search to determine a different set of items compared to the ranked items, wherein the optimized search is associated with the query.

15. The online shopping platform of claim 14 wherein the user behavior component analyzes the user behavior log of the customer to determine the customer preference.

16. The online shopping platform of claim 15 wherein the personalized ranking model component:

calculates an appropriate price interval for the customer from the user behavior log of the customer;

generates Gaussian weight values for the general search items;

calculates ranking scores for the general search items using a personalized ranking function with the Gaussian weight values; and

ranks the general search items based on the ranking scores of the general search items.

17. The online shopping platform of claim 14:

comprising a quality table rating quality of items in the items database; and

wherein

items in the items database comprise pricing of the items, and

the item price-quality relationship component

generates a distribution of price-quality relationship based on quality rating and pricing of the specific search items, and

groups the specific search items into cost performance groups based on the distribution.

18. The online shopping platform of claim 14 wherein the specific search items in the cost performance groups are ranked based on customer preference.

19. A non-transitory computer-readable medium having stored thereon program code which, when executed by a computer, causes the computing to perform operations for providing personalized search in an online shopping platform, the operations comprising;

receiving a query from a customer by the online shopping platform;

analyzing the query by a query analyzer to determine whether the query is a general item query or a specific item query;

based on determining that the query is the general item query, performing a general item search in an items database including items offered for sale by a general retrieval module using the general item query to retrieve general search items, wherein performing the general item search includes retrieving most relevant items as the general search items from the items database, wherein performing the general item search includes ranking the general search items based on customer preference;

based on determining that the query is the specific item query,

performing a specific item search in the items database by a specific retrieval module using the specific item query to retrieve specific search items,

categorizing each of the specific search items into a respective one of a plurality of cost performance categories based on cost performance, and ranking the specific search items based at least partly on the respective cost performance categories of the specific search items, wherein the cost performance for a respective item includes a price-quality relationship indicating a proportion of price to quality for the respective item, and wherein the ranking of the specific search items includes ranking the specific search items based at least partly on the price-quality relationship indicating the proportion of price to quality for each of the specific search items;

based on an executed search, providing ranked items to the customer at a user interface of a device connected to the online shopping platform;

storing data for user off-line and online behavior in a behavior log at the items database, wherein the data includes click-through data that indicates which of the provided ranked items the customer clicked on, and wherein the click-through data includes data for selection interactions of the customer after providing the ranked items as the search result;

optimizing retrieval of search results for searches requested by the customer based on the stored data for user off-line and online behavior to determine an optimized search result; and

initiating an optimized search to determine a different set of items compared to the ranked items, wherein the optimized search is associated with the query.

20. The non-transitory computer-readable medium of claim 19 the operations further comprising determining customer preference including:

deriving an appropriate price interval for the customer from a behavior log of the customer; and

generating Gaussian weight values for general and specific search items.