Adaptive filtering and modeling via adaptive experimental designs to identify emerging data patterns from large volume, high dimensional, high velocity streaming data

Patent metadata

Field	Value
Publication number	US-11443206-B2
Application number	US-202016751051-A
Country	US
Kind code	B2
Filing date	Jan 23, 2020
Priority date	Mar 23, 2015
Publication date	Sep 13, 2022
Grant date	Sep 13, 2022

Abstract

A system for identifying information in high dimensional, low latency streaming data having dynamically evolving data patterns. The system processes, continuously and in real-time, the streaming data. Processing includes filtering the data based on event data to identify diagnostic data points by comparing the event data with an experimental design matrix and performing a modeling operation using the identified diagnostic data points in order to identify efficiently any current and emerging patterns of relationships between at least one outcome variable and predictor variables. The at least one a-priori, pre-designed experimental design matrix is generated based on combinations of the predictor variables and at least one outcome variable. The experimental design matrix is also generated based on at least one of main effects, limitations, constraints, and interaction effects of the predictor variables and combinations.

First claim

What is claimed is:

1. A computer-implementable method for identifying information in high dimensional data streams having dynamically evolving data patterns, the method comprising:
loading at least one a-priori, pre-designed experimental design matrix and at least one modeling operation into memory;
processing streaming data continuously, wherein processing comprises:
filtering the streaming data based on event data to identify diagnostic data points by comparing the event data with the at least one a-priori, pre-designed experimental design matrix; and

performing the modeling operation using the identified diagnostic data points to identify current and emerging patterns of relationships between at least one outcome variable and predictor variables,
wherein the at least one a-priori, pre-designed experimental design matrix is generated based on combinations of the predictor variables, wherein the combinations are based on an outcome variable, wherein the at least one a-priori, pre-designed experimental design matrix is generated further based on at least one of:
main effects of predictor variable values;
limitations of the combinations of predictor variable values;
constraints of the combination of predictor variable values; and
interaction effects between selected predictor variables.

2. The computer-implementable method of claim 1 wherein the modeling operation is one of a prediction modeling operation and a clustering modeling operation.

3. The computer-implemented method of claim 1 wherein limitations of the predictor variable values are determined based on range values for the predictor variables, wherein the predictor variable values are continuous predictor variables.

4. The computer-implemented method of claim 1 wherein constraints of the combination of predictor variable values are based on a region of interest, wherein the predictor variable values are discrete predictor variable values.

5. The computer-implemented method of claim 1 wherein the at least one a-priori, pre-designed experimental design matrix is generated based on one of a space-filling design and an optimal experimental design.

6. The computer-implemented method of claim 1 wherein processing further comprises dynamically updating a visualization time window of the streaming data.

7. A system for identifying information in high dimensional data streams having dynamically evolving data patterns, the system comprises:
one or more processors;
a memory coupled to the one or more computer processors and comprising instructions, which when performed by the one or more computer processors, cause the one or more processors to perform operations to:
load at least one a-priori, pre-designed experimental design matrix and at least one modeling operation into memory;
filter streaming data based on event data continuously to identify diagnostic data points by comparing the event data with the at least one a-priori, pre-designed experimental design matrix; and

perform the modeling operation using the identified diagnostic data points to identify current and emerging patterns of relationships between at least one outcome variable and predictor variables,
wherein the at least one a-priori, pre-designed experimental design matrix is generated based on combinations of the predictor variables, wherein the combinations are based on an outcome variable, wherein the at least one a-priori, pre-designed experimental design matrix is generated further based on at least one of:
main effects of predictor variable values;
limitations of the combinations of predictor variable values;
constraints of the combination of predictor variable values; and
interaction effects between selected predictor variables.

8. The system of claim 7 wherein the modeling operation is one of a prediction modeling operation and a clustering modeling operation.

9. The system of claim 7 wherein limitations of the predictor variable values are determined based on range values for the predictor variables, wherein the predictor variable values are continuous predictor variables.

10. The system of claim 7 wherein constraints of the combination of predictor variable values are based on a region of interest, wherein the predictor variable values are discrete predictor variable values.

11. The system of claim 7 wherein the at least one a-priori, pre-designed experimental design matrix is generated based on one of a space-filling design and an optimal experimental design.

12. The system of claim 7 wherein the instructions further cause the at least one processor to perform operations to dynamically update a visualization time window of the streaming data.

13. At least one non-transitory computer readable medium comprising instructions for identifying information in high dimensional streaming data having dynamically evolving data patterns, when executed by at least one processor, cause the at least one processor to perform operations to:
load at least one a-priori, pre-designed experimental design matrix and at least one modeling operation into memory;
filter, continuously and real-time, streaming data based on event data to identify diagnostic data points by comparing the event data with the at least one a-priori, pre-designed experimental design matrix; and
perform, continuously and in real-time, the modeling operation using the identified diagnostic data points to identify current and emerging patterns of relationships between at least one outcome variable and predictor variables,
wherein the at least one a-priori, pre-designed experimental design matrix is generated based on combinations of the predictor variables, wherein the combinations are based on an outcome variable, wherein the at least one a-priori, pre-designed experimental design matrix is generated further based on at least one of:
main effects of predictor variable values;
limitations of the combinations of predictor variable values;
constraints of the combination of predictor variable values; and
interaction effects between selected predictor variables.

14. The at least one non-transitory computer readable medium of claim 13 wherein the modeling operation is one of a prediction modeling operation and a clustering modeling operation.

Assignees

• Tibco Software Inc

Inventors

• Hill Thomas
• O'Connell Michael
• Rope Daniel J

Classifications

• G06F16/2474

  Sequence data queries, e.g. querying versioned data · CPC title

• G06F16/26

  Visual data mining; Browsing structured data · CPC title

• G06F16/285

  Clustering or classification · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F16/9035

  Filtering based on additional data, e.g. user or group profiles · CPC title