Network optimization

Patent metadata

Field	Value
Publication number	US-11775852-B2
Application number	US-201916555375-A
Country	US
Kind code	B2
Filing date	Aug 29, 2019
Priority date	Aug 29, 2019
Publication date	Oct 3, 2023
Grant date	Oct 3, 2023

Abstract

A system may receive a cluster prediction requirement. The system may determine a first node conglomerate by sorting a first dataset into a first plurality of nodes. The system may determine a plurality of attributes by sorting a second dataset associated with the cluster prediction requirement. The system may determine a second node conglomerate for each of the plurality of attributes. A node confidence score may be assigned to each of the second plurality of nodes. The system may determine a node graph based on a comparison between the first node conglomerate and the second node conglomerate. The node graph may be iteratively modified based on a node optimization threshold value to generate a harmonized node graph. The node optimization threshold value may be based on a map confidence score allotted to the node graph.

First claim

We claim:

1. A system comprising:
a processor;
a data analyzer coupled to the processor, the data analyzer to:
receive a query from a user, the query to indicate a cluster prediction requirement; and
determine a first node conglomerate by sorting a first dataset into a first plurality of nodes, each of the first plurality of nodes being associated with the cluster prediction requirement;
a node analyzer coupled to the processor, the node analyzer to:
determine a plurality of attributes by sorting a second dataset associated with the cluster prediction requirement, the plurality of attributes to be associated with the cluster prediction requirement;
determine a second node conglomerate for each of the plurality of attributes, each of the second node conglomerate to include a second plurality of nodes associated with each of the plurality of attributes; and
assign a node confidence score to each of the second plurality of nodes associated with each of the plurality of attributes; and
a node optimizer coupled to the processor, the node optimizer to:
determine a node graph based on a comparison between the first node conglomerate and the second node conglomerate for each of the plurality of attributes, the node graph to include the first node conglomerate mapped to the second node conglomerate for each of the plurality of attributes through a plurality of mapped connections;
determine a map confidence index from the node graph, the map confidence index including a confidence value assigned to each of the plurality of mapped connections based on the node confidence score of the each of the second plurality of nodes associated with a corresponding mapped connection from each of the plurality of mapped connections; modify the node graph iteratively based on a node optimization threshold value to generate a harmonized node graph, the processing of the cluster prediction requirement being based on the harmonized node graph; and
create a node graph library, by associating the plurality of attributes, the second plurality of nodes associated with each of the plurality of attributes, the harmonized node graph, and the map confidence index with the cluster prediction requirement.

2. The system as claimed in claim 1 , wherein the node analyzer is to modify the node confidence score assigned to each of the second plurality of nodes associated with each of the plurality of attributes based on an input from the user.

3. The system as claimed in claim 1 , wherein the node analyzer is to predict at least one of the attributes from the plurality of attributes of the second dataset to generate a set of predicted plurality of attributes.

4. The system as claimed in claim 3 , wherein the system is to consider the set of predicted plurality of attributes as a unit of the plurality of attributes sorted from the second dataset.

5. The system as claimed in claim 1 , wherein the node optimizer is to generate the harmonized node graph by iteratively modifying the confidence value assigned to each of the plurality of mapped connections until the node optimization threshold value is achieved.

6. The system as claimed in claim 1 , wherein the system is to further deploy the node graph library for validation of the harmonized node graph.

7. A method comprising:
receiving, by a processor, a query from a user, the query to indicate a cluster prediction requirement;
determining, by the processor, a first node conglomerate by sorting a first dataset into a first plurality of nodes, each of the first plurality of nodes being associated with the cluster prediction requirement;
determining, by the processor, a plurality of attributes by sorting a second dataset associated with the cluster prediction requirement, the plurality of attributes to be associated with the cluster prediction requirement;
determining, by the processor, a second node conglomerate for each of the plurality of attributes, each of the second node conglomerate to include a second plurality of nodes associated with each of the plurality of attributes;
assigning, by the processor, a node confidence score to each of the second plurality of nodes associated with each of the plurality of attributes;
determining, by the processor, a node graph based on a comparison between the first node conglomerate and the second node conglomerate for each of the plurality of attributes, the node graph to include the first node conglomerate mapped to the second node conglomerate for each of the plurality of attributes through a plurality of mapped connections;
determining, by the processor, a map confidence index from the node graph, the map confidence index including a confidence value assigned to each of the plurality of mapped connections based on the node confidence score of the each of the second plurality of nodes associated with a corresponding mapped connection from each of the plurality of mapped connections;
modifying, by the processor, the node graph iteratively based on a node optimization threshold value to generate a harmonized node graph, the processing of the cluster prediction requirement being based on the harmonized node graph; and
creating, by the processor, a node graph library, by associating the plurality of attributes, the second plurality of nodes associated with each of the plurality of attributes, the harmonized node graph, and the map confidence index with the cluster prediction requirement.

8. The method as claimed in claim 7 , wherein the method further comprises modifying, by the processor, the node confidence score assigned to each of the second plurality of nodes associated with each of the plurality of attributes based on an input from the user.

9. The method as claimed in claim 7 , wherein the method further comprises predicting, by the processor, at least one of the attributes from the plurality of attributes of the second dataset to generate a set of predicted plurality of attributes.

10. The method as claimed in claim 9 , wherein the method further comprises considering, by the processor, the set of predicted plurality of attributes as a unit of the plurality of attributes sorted from the second dataset.

11. The method as claimed in claim 7 , wherein the method further comprises generating, by the processor, the harmonized node graph by iteratively modifying the confidence value assigned to each of the plurality of mapped connections until the node optimization threshold value is achieved.

12. The method as claimed in claim 7 , wherein the method further comprises deploying, by the processor, the node graph library for validation of the harmonized node graph.

13. A non-transitory computer readable medium including machine readable instructions that are executable by a processor to:
receive a query from a user, the query to indicate a cluster prediction requirement;
determine a first node conglomerate by sorting a first dataset into a first plurality of nodes, each of the first plurality of nodes being associated with the cluster prediction requirement;
determine a plurality of attributes by sorting a second dataset associated with the cluster prediction requirement, the plurality of attributes to be associated with the cluster prediction requirement;
determine a second node conglomerate for each of the plurality of attributes, each of the second node conglomerate to include a second plurality of nodes associated with each of the plurality of attributes;
assign a node confidence score to each of the second plurality of nodes associated with each of the plurality of attributes;
determine a node graph based on a comparison between the first node conglomerate and the second node conglomerate

Assignees

• Accenture Global Solutions Ltd

Inventors

• Parikh Ashesh
• Husain Afzal
• Arad Alon
• Aswathnarayana Suresh
• Myhrer Ragnar-Miguel
• Rao Tejas
• Sachdev Sharad
• Intriligator Joshua
• Alfieri Scott Andrew

Classifications

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N7/01 Primary

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F16/906

  Clustering; Classification · CPC title