Systems and methods for trend aware self-correcting entity relationship extraction

• US 10,157,344 B2
• Filed: 05/10/2016
• Issued: 12/18/2018
• Est. Priority Date: 03/23/2010
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Sinha, Ayon
• Primary Examiner(s): CHEN, ALAN S
• Application Number:
  US15/150,954
• Publication Number:
  US 20160253592A1
• Time in Patent Office: 952 Days
• Field of Search: None
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Methods and systems for trend aware self-correcting entity relationship extraction are disclosed. For example, a method can include receiving a selected entity, determining a plurality of entities related to the selected entity, determining a plurality of most probable entities, calculating relevance scores, and displaying a subset of the plurality of most probable entities. The selected entity can be received on a network-based transaction system. The plurality of entities related to the selected entity can be determined based on a relationship score. The relationship score can represent navigation transitions, aggregated over time, between the selected entity and each of the plurality of entities. The plurality of most probable entities can be determined based on probabilities. Relevance scores can be calculated for each of the plurality of most probable entities. Finally, the subset of the plurality of most probable entities to be displayed can be determined according to the relevance scores.

Claims

1. A method comprising:

receiving a selected entity on a network-based transaction system;

retrieving a first navigation trail of the selected entity, the first navigation trail including a first log of transitions between content located within the network-based transaction system, the content including any individually addressable portion of the network-based transaction system;

extracting the first log of transitions from the first navigation trail to a first data table at a database;

retrieving a second navigation trail of a second entity, the second navigation trail including a second log of transitions;

extracting the second log of transitions from the second navigation trail to a second data table at the database;

calculating an edge strength between the selected entity and the second entity, the edge strength calculated based on the first log of transitions of the first data table and the second log of transitions of the second data table; and

causing display of a recommendation of the second entity based on the calculated edge strength.

2. The method of claim 1, wherein the calculating the edge strength between the selected entity and the second entity includes:

determining an amount of navigation between the selected entity and the second entity.

3. The method of claim 2, wherein the determining the amount of navigation between the selected entity and the second entity includes:

determining user session boundaries for determining related navigation history.

4. The method of claim 1, wherein the method further comprises calculating a relevance score for the second entity, and wherein the calculating the relevance score includes:

determining a visitation rate of the second entity;

calculating a change in the visitation rate of a period of time, the change in the visitation rate based on the visitation rate and a previous visitation rate;

calculating an entity velocity of the second entity based on the change in the visitation rate; and

calculating the relevance score based on the entity velocity.

5. The method of claim 1, wherein the retrieving the second navigation trail of the second entity includes:

determining a plurality of related entities based at least in part on aggregated navigation transitions between the selected entity and each related entity of the plurality of related entities, the plurality of related entities including the second entity.

6. The method of claim 5, further comprising:

selecting, from among the plurality of related entities, the second entity based at least in part on a probability calculated based on the aggregated navigation transitions.

7. The method of claim 1, further comprising:

aggregating navigation transition between entities, wherein the navigation transitions include the first navigation trail and the second navigation trail.

8. A system comprising:

processors; and

a memory storing instructions that, when executed by at least one processor among the processors, causes the system to perform operations comprising;

receiving a selected entity on a network-based transaction system;

retrieving a first navigation trail of the selected entity, the first navigation trail including a first log of transitions between content located within the network-based transaction system, the content including any individually addressable portion of the network-based transaction system;

extracting the first log of transitions from the first navigation trail to a first data table at a database;

retrieving a second navigation trail of a second entity, the second navigation trail including a second log of transitions;

extracting the second log of transitions from the second navigation trail to a second data table at the database;

calculating an edge strength between the selected entity and the second entity, the edge strength calculated based on the first log of transitions the first data table and the second log of transitions of the second data table; and

causing display of a recommendation of the second entity based on the calculated edge strength.

9. The system of claim 8, wherein the operations for calculating the edge strength between the selected entity and the second entity further comprise:

determining an amount of navigation between the selected entity and the second entity.

10. The system of claim 9, wherein the operations for determining the amount of navigation between the selected entity and the second entity include:

determining user session boundaries for determining related navigation history.

11. The system of claim 8, wherein the operations further comprise:

calculating a relevance score for the second entity, and wherein the calculating the relevance score includes;

determining a visitation rate of the second entity;

calculating a change in the visitation rate of a period of time, the change in the visitation rate based on the visitation rate and a previous visitation rate;

calculating an entity velocity of the second entity based on the change in the visitation rate; and

calculating the relevance score based on the entity velocity.

12. The system of claim 8, wherein the operations for retrieving the second navigation trail of the second entity comprise:

determining a plurality of related entities based at least in part on aggregated navigation transitions between the selected entity and each related entity of the plurality of related entities, the plurality of related entities including the second entity.

13. The system of claim 12, wherein the operations further comprise:

selecting, from among the plurality of related entities, the second entity based at least in part on a probability calculated based on the aggregated navigation transitions.

14. The system of claim 8, wherein the operations further comprise:

aggregating navigation transition between entities, wherein the navigation transitions include the first navigation trail and the second navigation trail.

15. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:

receiving a selected entity on a network-based transaction system;

retrieving a first navigation trail of the selected entity, the first navigation trail including a first log of transitions between content located within the network-based transaction system, the content including any individually addressable portion of the network-based transaction system;

extracting the first log of transitions from the first navigation trail to a first data table at a database;

retrieving a second navigation trail of a second entity, the second navigation trail including a second log of transitions;

extracting the second log of transitions from the second navigation trail to a second data table at the database;

calculating an edge strength between the selected entity and the second entity, the edge strength calculated based on the first log of transitions of the first data table and the second log of transitions of the second data table; and

causing display of a recommendation of the second entity based on the calculated edge strength.

16. The non-transitory machine-readable storage medium of claim 15, wherein the instructions for the calculating the edge strength between the selected entity and the second entity include instructions that cause the machine to perform operations further comprising:

determining an amount of navigation between the selected entity and the second entity.

17. The non-transitory machine-readable storage medium of claim 16, wherein instructions for the determining the amount of navigation between the selected entity and the second entity include instructions that cause the machine to perform operations further comprising:

determining user session boundaries for determining related navigation history.

18. The non-transitory machine-readable storage medium of claim 15, further comprising instructions that cause the system to perform operations comprising:

calculating a relevance score for the second entity, and wherein the calculating the relevance score includes;

determining a visitation rate of the second entity;

calculating a change in the visitation rate of a period of time, the change in the visitation rate based on the visitation rate and a previous visitation rate;

calculating an entity velocity of the second entity based on the change in the visitation rate; and

calculating the relevance score based on the entity velocity.

19. The non-transitory machine-readable storage medium of claim 15, wherein the instructions for the retrieving the second navigation trail of the second entity cause the system to perform operations further comprising:

determining a plurality of related entities based at least in part on aggregated navigation transitions between the selected entity and each related entity of the plurality of related entities, the plurality of related entities including the second entity.

20. The non-transitory machine-readable storage medium of claim 19, further comprising instructions that cause the system to perform operations further comprising:

selecting, from among the plurality of related entities, the second entity based at least in part on a probability calculated based on the aggregated navigation transitions.