Candidate visualization techniques for use with genetic algorithms

Patent metadata

Field	Value
Publication number	US-11651233-B2
Application number	US-202016787848-A
Country	US
Kind code	B2
Filing date	Feb 11, 2020
Priority date	Sep 23, 2015
Publication date	May 16, 2023
Grant date	May 16, 2023

Abstract

According to one embodiment, a method for generating a plurality of candidate visualizations. The method may include receiving a scenario description. The method may also include collecting a plurality of expert data using a training system based on the received scenario description. The method may further include generating at least one predictive model based on the collected plurality of expert data in order to execute the at least one generated predictive model during an application of a plurality of genetic algorithms.

First claim

What is claimed is:

1. A processor-implemented method for generating a plurality of candidate visualizations, the method comprising:
collecting a plurality of expert data entered through interactions by a subject matter expert user with a graphical user interface of a training system based on a scenario-description, wherein the plurality of expert data comprises aspects and attributes of visualizations favored by the subject matter expert user when assessing a fitness of the visualizations; and wherein the plurality of expert data further comprises metric data, action data, and opinion data, and wherein the plurality of metric data comprises statistical measurements captured from within a visualization or metadata of fields used by the visualization including skewness of a field and kurtosis of a field, and wherein the action data comprises metadata capable of capturing aspects of actions within a visualization, and wherein the opinion data comprises free text opinions regarding the visualization provided by a subject matter expert and data regarding an expertise level of the subject matter expert;
generating at least one predictive model based on the plurality of collected expert data;
calculating a fitness score for each of a plurality of candidate visualizations by executing the at least one generated predictive model during an application of a plurality of genetic algorithms; and

generating a next generation of candidate visualizations using the plurality of genetic algorithms by mutating or cross-breeding candidate visualizations with a calculated fitness score that satisfies a preconfigured threshold value.

2. The method of claim 1 , wherein the training system collects the plurality of expert data through at least one of an expert selecting a preferred visualization within a plurality of candidate visualizations and an expert submitting an opinion related to the preferred visualization.

3. The method of claim 1 , wherein the training system includes at least one of a central data storage facility, a plurality of web-based deployment capabilities, and scalability to a corresponding number of experts using the training system.

4. The method of claim 1 , further comprising:
updating the training system based on the at least one generated predictive model.

5. The method of claim 1 , wherein generating the at least one predictive model comprises using boosted decision trees, text categorization, and neural networking.

6. The method of claim 1 , wherein each at least one generated predictive model may be used to derive an overall fitness evaluation score to rate a candidate visualization when executing the plurality of genetic algorithms.

7. A computer system for generating a plurality of candidate visualizations, the method comprising:
collecting a plurality of expert data entered through interactions by a subject matter expert user with a graphical user interface of a training system based on a scenario-description, wherein the plurality of expert data comprises aspects and attributes of visualizations favored by the subject matter expert user when assessing a fitness of the visualizations; and wherein the plurality of expert data further comprises metric data, action data, and opinion data, and wherein the plurality of metric data comprises statistical measurements captured from within a visualization or metadata of fields used by the visualization including skewness of a field and kurtosis of a field, and wherein the action data comprises metadata capable of capturing aspects of actions within a visualization, and wherein the opinion data comprises free text opinions regarding the visualization provided by a subject matter expert and data regarding an expertise level of the subject matter expert;
generating at least one predictive model based on the plurality of collected expert data;
calculating a fitness score for each of a plurality of candidate visualizations by executing the at least one generated predictive model during an application of a plurality of genetic algorithms; and

generating a next generation of candidate visualizations using the plurality of genetic algorithms by mutating or cross-breeding candidate visualizations with a calculated fitness score that satisfies a preconfigured threshold value.

8. The computer system of claim 7 , wherein the training system collects the plurality of expert data through at least one of an expert selecting a preferred visualization within a plurality of candidate visualizations and an expert submitting an opinion related to the preferred visualization.

9. The computer system of claim 7 , wherein the training system includes at least one of a central data storage facility, a plurality of web-based deployment capabilities, and scalability to a corresponding number of experts using the training system.

10. The computer system of claim 7 , further comprising:
updating the training system based on the at least one generated predictive model.

11. The computer system of claim 7 , wherein generating the at least one predictive model comprises using boosted decision trees, text categorization, and neural networking.

12. The computer system of claim 7 , wherein each at least one generated predictive model may be used to derive an overall fitness evaluation score to rate a candidate visualization when executing the plurality of genetic algorithms.

13. A non-transitory computer-readable storage medium having stored therein instructions which, when executed by a processor, cause the processor to perform a method for generating a plurality of candidate visualizations, the method comprising:
collecting a plurality of expert data entered through interactions by a subject matter expert user with a graphical user interface of a training system based on a scenario-description, wherein the plurality of expert data comprises aspects and attributes of visualizations favored by the subject matter expert user when assessing a fitness of the visualizations; and wherein the plurality of expert data further comprises metric data, action data, and opinion data, and wherein the plurality of metric data comprises statistical measurements captured from within a visualization or metadata of fields used by the visualization including skewness of a field and kurtosis of a field, and wherein the action data comprises metadata capable of capturing aspects of actions within a visualization, and wherein the opinion data comprises free text opinions regarding the visualization provided by a subject matter expert and data regarding an expertise level of the subject matter expert;
generating at least one predictive model based on the plurality of collected expert data;
calculating a fitness score for each of a plurality of candidate visualizations by executing the at least one generated predictive model during an application of a plurality of genetic algorithms; and
generating a next generation of candidate visualizations using the plurality of genetic algorithms by mutating or cross-breeding candidate visualizations with a calculated fitness score that satisfies a preconfigured threshold value.

14. The computer program product of claim 13 , wherein the training system collects the plurality of expert data through at least one of an expert selecting a preferred visualization within a plurality of candidate visualizations and an expert submitting an opinion related to the preferred visualization.

15. The computer program product of claim 13 , wherein the training system includes at least one of a central data storage facility, a plurality of web-based deployment capabilities, and scalability

Assignees

• IBM

Inventors

• Keahey T Alan
• Rope Daniel J
• Wills Graham J

Classifications

• G06T11/26

  Drawing of charts or graphs · CPC title

• G06N3/126 Primary

  Evolutionary algorithms, e.g. genetic algorithms or genetic programming · CPC title

• G06T11/206

  Physics · mapped topic