Identifying child node correlations

Patent metadata

Field	Value
Publication number	US-10102560-B1
Application number	US-201615079525-A
Country	US
Kind code	B1
Filing date	Mar 24, 2016
Priority date	Mar 24, 2016
Publication date	Oct 16, 2018
Grant date	Oct 16, 2018

Abstract

Disclosed are various embodiments for identifying correlations between child nodes of a taxonomy. A correlation service identifies a correlation between parent nodes of the taxonomy. A data set such as an interaction history is filtered according to the parent node correlation. The filtered data set is then used to identify correlations between the respective child nodes of the correlated parent nodes.

First claim

Therefore, the following is claimed:

1. A non-transitory computer-readable medium embodying a program executable in at least one computing device, the program, when executed, causing the at least one computing device to at least:
identify, from a plurality of items embodied as a plurality of taxonomy nodes, a first correlation between a first taxonomy node among the plurality of taxonomy nodes and a second taxonomy node among the plurality of taxonomy nodes based at least in part on an interaction history with respect to the plurality of items and responsive to a first confidence score meeting a first predefined threshold;
select a subset of the interaction history corresponding to a first child node of the first taxonomy node and at least one second child node of the second taxonomy node;
identify, based at least in part on the subset of the interaction history, at least one second correlation between the first child node and a respective one of the at least one second child node responsive to at least one second confidence score meeting a second predefined threshold less than the first predefined threshold; and
store an indication of the at least one second correlation in a data store.

2. The non-transitory computer-readable medium of claim 1 , wherein the program further causes the at least one computing device to at least:
load, from the data store, the indication in response to a request for a network page associated with the first child node; and
generate the network page comprising a navigation aid directed to the respective one of the at least one second child node.

3. The non-transitory computer-readable medium of claim 1 , wherein the interaction history comprises at least one of a browse history or a purchase history.

4. A system, comprising:
A memory configured to store computer-readable instructions thereon; and
at least one computing device configured to, through the execution of the computer-readable instructions, at least:
identify, from a plurality of items embodied as a plurality of taxonomy nodes, a first correlation between a first taxonomy node of the plurality of taxonomy nodes and a second taxonomy node of the plurality of taxonomy nodes based at least in part on an interaction history with respect to the plurality of items;
select a subset of the interaction history corresponding to a first child node of the first taxonomy node and at least one second child node of the second taxonomy node; and
identify, based at least in part on the subset of the interaction history, at least one second correlation between the first child node and a respective one of the at least one second child node.

5. The system of claim 4 , wherein the interaction history comprises at least one of a browse history or a purchase history.

6. The system of claim 4 , wherein selecting the subset of the interaction history comprises selecting the subset of the interaction history as indicating a first action taken with respect to the first child node and a second action taken with respect to the at least one second child node.

7. The system of claim 5 , wherein the first action and the second action are distinct.

8. The system of claim 4 , wherein the at least one computing device is further configured to identify the first correlation by identifying, based at least in part on the interaction history, the first correlation between the first taxonomy node and the second taxonomy node responsive to a confidence score meeting a predefined threshold.

9. The system of claim 8 , wherein the predefined threshold is a first predefined threshold, the confidence score is a first confidence score, and wherein the at least one computing device is further configured to identify the at least one second correlation by identifying at least one second confidence score corresponding to the first child node and the respective one of the at least one second child node meeting a second predefined threshold.

10. The system of claim 9 , wherein the second predefined threshold is less than the first predefined threshold.

11. The system of claim 9 , wherein the at least one computing device is further configured to identify the at least one second confidence score by identifying, from the at least one second child node, the respective one of the at least one second child node as having a highest confidence score meeting the second predefined threshold.

12. The system of claim 4 , wherein the at least one computing device is further configured to at least store an indication of the at least one second correlation in a data store.

13. The system of claim 4 , wherein the at least one computing device is further configured to at least generate, in response to a request for a network page associated with the first child node, the network page comprising a navigation aid directed to the respective one of the at least one second child node.

14. A method, comprising:
identifying, by at least one computing device, from a plurality of items embodied as a plurality of taxonomy nodes, a first correlation between a first taxonomy node of the plurality of taxonomy nodes and a second taxonomy node of the plurality of taxonomy nodes based at least in part on an interaction history with respect to the plurality of items;
selecting, by the at least one computing device, a subset of the interaction history corresponding to a first child node of the first taxonomy node and at least one second child node of the second taxonomy node; and
identify, by the at least one computing device, at least one second correlation between the first child node and a respective one of the at least one second child node based at least in part on the subset of the interaction history.

15. The method of claim 14 , wherein identifying the first correlation is performed at a first interval, and identifying the at least one second correlation is performed at a second interval distinct from the first interval.

16. The method of claim 14 , wherein the second interval is less than the first interval.

17. The method of claim 14 , wherein the subset of the interaction history is selected as indicating a browse action with respect to the first child node and a purchase action taken with respect to the at least one second child node.

18. The method of claim 14 , wherein identifying the first correlation comprises identifying, based at least in part on the interaction history, the first correlation between the first taxonomy node and the second taxonomy node responsive to a confidence score meeting a predefined threshold.

19. The method of claim 18 , wherein the predefined threshold is a first predefined threshold, the confidence score is a first confidence score, and wherein identifying the at least one second correlation comprises identifying at least one second confidence score corresponding to the first child node and the respective one of the at least one second child node meeting a second predefined threshold less than the first predefined threshold.

20. The method of claim 14 , further comprising replacing, in a data store, a first indication of the first correlation with at least one second indication of the at least one second correlation.

Assignees

• Amazon Tech Inc

Inventors

• Kurup Madhu Madhava
• Emery Grant Michael

Classifications

• G06Q30/0631 Primary

  Recommending goods or services · CPC title