Coverage analysis with event clustering

Patent metadata

Field	Value
Publication number	US-11640421-B2
Application number	US-201916411542-A
Country	US
Kind code	B2
Filing date	May 14, 2019
Priority date	May 14, 2019
Publication date	May 2, 2023
Grant date	May 2, 2023

Abstract

A computer that receives a set of names of coverage events. The computer creates, by a machine learning-based technique, groups from the set of received names of the coverage events based on the set of names of the coverage events. The computer generates a cross product coverage model from the created groups and identifies subgroups of uncovered events for each of the created groups.

First claim

What is claimed is:

1. A processor-implemented method for event clustering, the method comprising:
receiving a set of names of coverage events;
creating, by a machine learning-based technique, groups from the set of received names of the coverage events based on the set of names of the coverage events, wherein the machine learning-based technique uses a K-means algorithm that considers a location of each word in the set of names for clustering the set of names in the created groups and maps the clustered set of names into cross product spaces;
determining a cross product density for the cross product spaces using a ratio between a size of a cross product space and a number of events in the cross product space;
generating a cross product coverage model from the created groups, wherein generating the cross product coverage model comprises:
determining one or more anchors of a cross product space based on a word with a certain location in an event name that is common to all event names in a cluster;
determining one or more attributes of the cross product space based on determining that locations in the event names have more than one possible word in the cluster; and
generating the cross product coverage model based on the one or more anchors and the one or more attributes; and

identifying subgroups of uncovered events for each of the created groups.

2. The method of claim 1 further comprising:
displaying the identified subgroups of uncovered events in a table comprising a unit name, a total number of events, a number of total clusters, a number of large clusters, and a number of events in clusters, wherein the number of events in cluster represents a percentage value of the events in clusters to the number of total events for each unit.

3. The method of claim 1 , wherein the machine learning-based technique is further based on a Latent Dirichlet Allocation (LDA) algorithm.

4. The method of claim 1 , wherein generating the cross product coverage model from the created groups further comprises combining close cross products.

5. The method of claim 4 , wherein the cross product density is a ratio between a size of the cross-product space and a number of events in the cross-product space.

6. A computer system for event clustering, the computer system comprising:
one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising:
receiving a set of names of coverage events;
creating, by a machine learning-based technique, groups from the set of received names of the coverage events based on the set of names of the coverage events, wherein the machine learning-based technique uses a K-means algorithm that considers a location of each word in the set of names for clustering the set of names in the created groups and maps the clustered set of names into cross product spaces;
determining a cross product density for the cross product spaces using a ratio between a size of a cross product space and a number of events in the cross product space;
generating a cross product coverage model from the created groups, wherein generating the cross product coverage model comprises:
determining one or more anchors of a cross product space based on a word with a certain location in an event name that is common to all event names in a cluster;
determining one or more attributes of the cross product space based on determining that locations in the event names have more than one possible word in the cluster; and
generating the cross product coverage model based on the one or more anchors and the one or more attributes; and

identifying subgroups of uncovered events for each of the created groups.

7. The computer system of claim 6 , further comprising displaying the identified subgroups of uncovered events in a table comprising a unit name, a total number of events, a number of total clusters, a number of large clusters, and a number of events in clusters, wherein the number of events in cluster represents a percentage value of the events in clusters to the number of total events for each unit.

8. The computer system of claim 6 , wherein the machine learning-based technique is further based on a Latent Dirichlet Allocation (LDA) algorithm.

9. The computer system of claim 6 , wherein generating the cross product coverage model from the created groups further comprises combining close cross products.

10. The computer system of claim 9 , wherein the cross product density is a ratio between a size of the cross-product space and a number of events in the cross-product space.

11. A computer program product for event clustering, the computer program product comprising:
one or more computer-readable tangible storage medium and program instructions stored on at least one of the one or more tangible storage medium, the program instructions executable by a processor, the program instructions comprising:
program instructions to receive a set of names of coverage events;
program instructions to create, by a machine learning-based technique, groups from the set of received names of the coverage events based on the set of names of the coverage events, wherein the machine learning-based technique uses a K-means algorithm that considers a location of each word in the set of names for clustering the set of names in the created groups and maps the clustered set of names into cross product spaces;
program instructions to determine a cross product density for the cross product spaces using a ratio between a size of a cross product space and a number of events in the cross product space;
program instructions to generate a cross product coverage model from the created groups, wherein generating the cross product coverage model comprises:
determining one or more anchors of a cross product space based on a word with a certain location in an event name that is common to all event names in a cluster;
determining one or more attributes of the cross product space based on determining that locations in the event names have more than one possible word in the cluster; and
generating the cross product coverage model based on the one or more anchors and the one or more attributes; and

program instructions to identify subgroups of uncovered events for each of the created groups.

12. The computer program product of claim 11 further comprising program instructions to display the identified subgroups of uncovered events in a table comprising a unit name, a total number of events, a number of total clusters, a number of large clusters, and a number of events in clusters, wherein the number of events in cluster represents a percentage value of the events in clusters to the number of total events for each unit.

13. The computer program product of claim 11 , wherein the machine learning-based technique is further based on a Latent Dirichlet Allocation (LDA) algorithm.

14. The computer program product of claim 11 , wherein program instructions to generate the cross product coverage model from the created groups further comprises program instructions to combine close cross products.

15. The computer program product of claim 14 , wherein the cross product density is a ratio between a size of the cross-product space and a number of events in the cross-product space.

Assignees

• IBM

Inventors

• Gal Raviv
• Ziv Avi
• Simchoni Giora

Classifications

• G06F40/284

  Lexical analysis, e.g. tokenisation or collocates · CPC title

• G06F16/355 Primary

  Creation or modification of classes or clusters · CPC title

• G06F11/3676

  for coverage analysis · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F40/30

  Semantic analysis · CPC title