Systems and methods to generate a concept graph

Patent metadata

Field	Value
Publication number	US-10157240-B2
Application number	US-201514873099-A
Country	US
Kind code	B2
Filing date	Oct 1, 2015
Priority date	Oct 1, 2015
Publication date	Dec 18, 2018
Grant date	Dec 18, 2018

Abstract

In various example embodiments, a system and method for generating a concept graph are presented. Trending queries are determined from a query log stored in a database. Item pages that correspond to the trending queries are identified. Aspects from the item pages are also identified. The aspects from the item page may be stored as pre-defined data within the database. Query nodes for each of the trending queries are generated, and the query nodes are included in a graph. Leaf nodes for each of the aspects are generated. Connections between the leaf nodes and the query nodes are generated. A destination page that includes various aspects is generated.

First claim

What is claimed is:

1. A system comprising:
hardware processing circuitry;
a hardware memory storing instructions that configure the hardware processing circuitry to perform operations, the operations comprising:
determining, from a query log stored in a database, trending queries that are each indicated as being received at least a threshold number of instances, the trending queries being performed by users of a network publication system;
identifying item pages that correspond to the trending queries, each of the item pages having a corresponding item; and
identifying aspects of the items corresponding to the identified item pages, the aspects being stored as data within the database, each aspect identifying a feature of its corresponding item; and
generating a graph to include query nodes for each of the trending queries
generating a leaf node in the graph for each of the aspects from the item pages;
generating connections between the leaf nodes and the query nodes for each of the trending queries;
generating a destination page indicating at least one of the identified aspects based on the generated graph; and
cause display of the destination page on a client device.

2. The system of claim 1 , the operations further comprising identifying item pages that are selected by the users of the publication system from results generated by the trending queries.

3. The system of claim 1 , the operations further comprising generating a connection between a specific leaf node and a specific query node, the specific leaf node representing an aspect of an item page, the item page corresponding to a specific query being represented by the specific query node.

4. The system of claim 1 ,
the operations further comprising: identifying categories that correspond to items from the item pages; and
generating leaf nodes for each of the identified categories that correspond to the items from the item pages.

5. The system of claim 1 ,
the operations further comprising: identifying features of the items from the item pages; and
generating leaf nodes for each of the identified features of the items from the item pages.

6. The system of claim 1 , the operations further comprising: identifying,
from the trending queries, a first query and a second query, wherein the first query and the second query are indicated as being received in succession at least a threshold number of instances; and
generating a connection between the query nodes for each of the trending queries based on the identification of the first query and the second query.

7. The system of claim 1 , the operations further comprising:
generating nodes for each of the item pages that correspond to the trending queries; and
generating connections between the nodes for each of the item pages and the query nodes for each of the trending queries.

8. The system of claim 1 , the operations further comprising: from a client device, a query that matches one of the query nodes from the generated graph; and wherein:
using the generated graph, identifying a specific query node that matches the query received from the client device.

9. The system of claim 8 , wherein the destination page corresponds to the specific query node from the generated graph, and the various aspects corresponding to leaf nodes that are connected to the specific query node from the generated graph, wherein the destination page includes links that are selectable to display item pages that correspond to a trending query represented by the specific query node.

10. The system of claim 8 , the operations further comprising: receiving the trending queries from devices operated by the users of the publication system.

11. A method comprising:
determining, from a query log stored in a database, trending queries that are each indicated as being received at least a threshold number of instances, the trending queries being performed by users of a network publication system;
identifying item pages that correspond to the trending queries, each of the item pages having a corresponding item;
identifying aspects of the items corresponding to the identified item pages, the aspects being stored as data within the database, each aspect identifying a feature of its corresponding item;
generating, using one or more processors, a graph to include query nodes for each of the trending queries;
generating a leaf node in the graph for each of the aspects from the item pages;
generating connections between the leaf nodes and the query nodes for each of the trending queries;
generating a destination page indicating at least one of the identified aspects based on the generated graph; and
cause display of the destination page on a client device.

12. The method of claim 11 , wherein the identifying the item pages that correspond to the trending queries includes:
identifying item pages that are selected by the users of the publication system from results generated by the trending queries.

13. The method of claim 11 , wherein the generating the connections between the leaf nodes and the query nodes for each of the trending queries includes:
generating a connection between a specific leaf node and a specific query node, the specific leaf node representing an aspect of an item page, the item page corresponding to a specific query being represented by the specific query node.

14. The method of claim 11 , wherein:
identifying the aspects includes identifying categories that correspond to items from the item pages; and
generating leaf nodes for each of the identified categories that correspond to the items from the item pages.

15. The method of claim 11 , wherein:
identifying the aspects includes identifying features of the items from the item pages; and
generating the leaf nodes includes generating leaf nodes for each of the identified features of the items from the item pages.

16. The method of claim 11 , further comprising:
identifying, from the trending queries, a first query and a second query, wherein the first query and the second query are indicated as being received in succession at least a threshold number of instances; and
generating a connection between the query nodes for each of the trending queries based on the identification of the first query and the second query.

17. The method of claim 11 , further comprising:
generating nodes in the graph for each of the item pages that correspond to the trending queries; and
generating connections between the nodes for each of the item pages and the query nodes for each of the trending queries.

18. The method of claim 11 , further comprising:
receiving, from the client device, a query that matches one of the query nodes from the generated graph;
using the generated graph, identifying a specific query node that matches the query received from the client device, and
wherein the destination page corresponds to the specific query node from the generated graph, and the various aspects corresponding to leaf nodes that are connected to the specific query node from the generated graph.

19. The method of claim 18 , wherein the destination page includes links that are selectable to display item pages that correspond to a trending query represented by the specific query node.

20. A non-transitory machine-readable medium storing instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
determining, from a query log stored in a database, trending queries that are each indicated as bei

Assignees

• Ebay Inc

Inventors

• Islam Atiq
• Fazal Mahamood Shifa
• Gopalakrishnan Ganesh
• Talele Ganesh

Classifications

• G06F16/9027

  Trees · CPC title

• G06F16/9024 Primary

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title

• H04L67/02

  based on web technology, e.g. hypertext transfer protocol [HTTP] · CPC title

• G06F16/9038

  Presentation of query results · CPC title

• G06F17/30991

  Physics · mapped topic