# US 2025/0036554 A1 · Creating an optimal test suite

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0036554 A1
- Document ID: US20250036554A1
- Status: Published application
- Kind: A1
- Application number: 18/225,350
- Filed: 2023-07-24
- Published: 2025-01-30
- Earliest claimed: 2023-07-24
- Assignees: Webomates LLC
- Applicants: Webomates Inc.
- Inventors: Aseem BAKSHI; Mark SAWERS; Ruchika GUPTA
- Web page: https://patentgrep.com/patent/US20250036554A1

## CPC

- G06F 11/3676: {for coverage analysis}
- G06F 11/3688: {for test execution, e.g. scheduling of test suites}
- G06F 11/3684: {for test design, e.g. generating new test cases}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A system and a method for creating an optimal test suite. The system may receive code coverage data from a set of testing channels. Further, a combined dataset is created by merging the code coverage data received from the set of testing channels. The combined dataset is analyzed to identify a line of code being executed by two or more test cases corresponding to two or more testing channels of the set of testing channels. Further, at least one test case from the two or more test cases having less efficiency is eliminated to create an optimal test suite. The efficiency is determined based on at least one of execution time, execution cost, and resources required to execute the line of code.

## Claims

### Claim 1

A method to create an optimal test suite comprising:
receiving, by a processor, code coverage data from two or more testing channels, wherein the code coverage data is obtained by executing a software build with a test suite;
merging, by the processor, the code coverage data received from the two or more testing channels to create a combined dataset of the code coverage data;
analyzing, by the processor, the combined dataset to identify a line of code being executed by two or more test cases corresponding to the two or more testing channels;
eliminating, by the processor, at least one test case from the two or more test cases having lesser efficiency to create the optimal test suite, wherein the efficiency is determined based on at least one of an execution time, an execution cost, and resources utilized to execute the line of code;
executing, by the processor, the optimal test suite to generate master code coverage comprising code coverage data from the two or more testing channels; and
generating, by the processor, a set of test cases to modify the optimal test suite when at least one of:
the master code coverage is less than a defined threshold, and
critical code paths are unexecuted;
wherein the set of test cases is generated using an Artificially Intelligent (AI) model and Natural Language Processing (NLP) techniques, wherein the AI model is trained using a training dataset comprising test cases along with their corresponding code coverage information, and wherein the AI model is continuously trained based on feedback received regarding the generated set of test cases.

### Claim 2 (depends on 1)

The method of claim 1, wherein the two or more testing channels comprises at least two of a unit testing channel, an automation testing channel, a manual testing channel, and a crowdsource testing channel.

### Claim 3

(canceled)

### Claim 4

(canceled)

### Claim 5

(canceled)

### Claim 6 (depends on 1)

The method of claim 1, wherein the critical code paths are determined by using a code profiling tool.

### Claim 7 (depends on 1)

The method of claim 1, further comprising determining, by the processor, a testing channel having highest code coverage value from the two or more testing channels, wherein the testing channel is determined based on at least one of execution time, execution cost, and resources versus code coverage value.

### Claim 8

A system for creating an optimal test suite, the system comprising:
a memory; and
a processor coupled to the memory, wherein the processor is configured to execute program instructions stored in the memory to:
receive code coverage data from two or more testing channels, wherein the code coverage data is obtained by executing a software build with a test suite;
merge the code coverage data received from the two or more testing channels to create a combined dataset;
analyze the combined dataset to identify a line of code being executed by two or more test cases corresponding to two or more testing channels;
eliminate at least one test case from the two or more test cases having lesser efficiency to create the optimal test suite, wherein the efficiency is determined based on at least one of an execution time, an execution cost, and resources utilized to execute the line of code;
executing the optimal test suite to generate master code coverage comprising code coverage data from the two or more testing channels; and
generating, by the processor, a set of test cases to modify the optimal test suite when at least one of:
the master code coverage is less than a defined threshold, and
critical code paths are unexecuted;
wherein the set of test cases is generated using an Artificially Intelligent (AI) model and Natural Language Processing (NLP) techniques, wherein the AI model is trained using a training dataset comprising test cases along with their corresponding code coverage information, and wherein the AI model is continuously trained based on feedback received regarding the generated set of test cases.

### Claim 9 (depends on 8)

The system of claim 8, wherein the two or more testing channels comprises at least two of a unit testing channel, an automation testing channel, a manual testing channel, and a crowdsource testing channel.

### Claim 10

(canceled)

### Claim 11

(canceled)

### Claim 12

(canceled)

### Claim 13 (depends on 8)

The system of claim 8, wherein the critical code paths are determined by using a code profiling tool.

### Claim 14 (depends on 8)

The system of claim 8, wherein the processor is further configured to: determine a testing channel having highest code coverage value from the two or more testing channels, wherein the testing channel is determined based on at least one of execution time, execution cost, and resources versus code coverage value.

### Claim 15

A non-transitory computer program product having embodied thereon a computer program to create an optimal test suite, the non-transitory computer program product storing instructions for:
receiving code coverage data from two or more testing channels, wherein the code coverage data is obtained by executing a software build with a test suite;
merging the code coverage data received from the two or more testing channels to create a combined dataset of the code coverage data;
analyzing the combined dataset to identify a line of code being executed by two or more test cases corresponding to two or more testing channels;
eliminating at least one test case from the two or more test cases having lesser efficiency to create an optimal test suite, wherein the efficiency is determined based on at least one of an execution time, an execution cost, and resources utilized to execute the line of code;
executing the optimal test suite to generate master code coverage comprising code coverage data from the two or more testing channels; and
generating a set of test cases to modify the optimal test suite when at least one of:
the master code coverage is less than a defined threshold, and
critical code paths are unexecuted;
wherein the set of test cases is generated using an Artificially Intelligent (AI) model and Natural Language Processing (NLP) techniques, wherein the AI model is trained using a training dataset comprising test cases along with their corresponding code coverage information, and wherein the AI model is continuously trained based on feedback received regarding the generated set of test cases.

### Claim 16 (depends on 1)

The method of claim 1, wherein the set of test cases is generated for an unexecuted section of the code, and wherein the set of test cases is generated for an automation testing channel.

### Claim 17

(canceled)

### Claim 18 (depends on 1)

The method of claim 1, wherein the code coverage information includes binary labels indicating whether each line or section of the code was covered.

### Claim 19 (depends on 1)

The method of claim 1, wherein the AI model is at least one of an Amazon Comprehend (AWS Comprehend), Google Cloud Natural API, an Information Retrieval model, a Semantic Textual Similarity model, and a Paraphrase Identification.

### Claim 20 (depends on 8)

The system of claim 8, wherein the set of test cases is generated for an unexecuted section of the code, and wherein the set of test cases is generated for an automation testing channel.

### Claim 21 (depends on 8)

The system of claim 8, wherein the code coverage information includes binary labels indicating whether each line or section of the code was covered.

### Claim 22 (depends on 8)

The system of claim 8, wherein the AI model is at least one of an Amazon Comprehend (AWS Comprehend), Google Cloud Natural API, an Information Retrieval model, a Semantic Textual Similarity model, and a Paraphrase Identification.
