Automated Batch Job Testing via Natural Language Scenario Graphs

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional testing tools are not suitable for batch jobs, requiring significant manual effort and system administrator intervention to design and execute test cases, and lack metadata to identify relevant data fields, making batch data testing resource-intensive and inefficient.

Innovation Solution

An intelligent batch job testing system that uses machine learning to independently generate and execute test cases by extracting keywords from natural language requests, creating scenario graphs, and mapping batch job files using optimized file layouts generated with metadata from sample files, allowing for automated test case execution and continuous improvement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional testing tools are used for batch jobs, then user interface-based testing can be performed, but the tools are not suitable for batch jobs and require significant manual intervention

Engineering Contradiction:

Improveease of test case executionVSAvoidautomation of test case generation and execution

Core Design Contradiction:

Ease of operationVSExtent of automation

Solution Approach 1:

The system performs self-service by automatically generating test cases from natural language requests without requiring manual intervention. The batch job testing system independently identifies data sources, extracts relevant fields, creates test scenarios, and executes test cases autonomously, eliminating the need for system administrator intervention that plagues conventional tools.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical processes with automated intelligent systems. Instead of manually designing test cases and identifying data fields, the system uses natural language processing, scenario graph generation, and machine learning algorithms to automatically generate and execute test cases, substituting human effort with automated intelligence.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If manual design of test scenarios is performed, then full coverage of batch job can be achieved, but significant investment of resources and time is required

Engineering Contradiction:

Improvecoverage of batch job testingVSAvoidtime for test case design and execution

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-generating comprehensive test scenarios from natural language requests before actual batch job execution. The scenario graph generation process creates all necessary test cases in advance, identifying data sources and fields beforehand, which eliminates the need for time-consuming manual test design during production testing.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms where test results are automatically analyzed and used to improve future test case generation. The machine learning components learn from executed test results to refine scenario graphs and optimize test coverage, continuously improving reliability while reducing the time required for manual test design iterations.

Inventive Principle:

Principle #23Feedback

3Productivity

If batch job files are processed without metadata, then file processing can be performed, but the system cannot identify relevant data fields

Engineering Contradiction:

Improvespeed of batch job processingVSAvoiddifficulty of identifying data fields

Core Design Contradiction:

ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The system introduces an intermediary natural language request that bridges the gap between batch job files without metadata and the testing system. The natural language description serves as a mediator that provides the necessary field identification information, allowing the system to process files efficiently while accurately identifying relevant data fields through the scenario graph generation process.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS10678682B2Intelligent batch job testing

Publication Date: 2020.06.09 BANK OF AMERICA CORP

AI summary

Apparatus and methods for an intelligent batch job testing system are provided. The system may receive a natural language request for batch job testing. The system may generate a graphical representation of a test scenario, based in part on keywords extracted from the natural language request. The graphical representation may include interconnected nodes. The graphical representation may include finite states for one or more of the nodes. The system may generate test cases based on the graphical representation. The system may access batch job data and map the data files using an optimized file layout that corresponds to the batch job data file. Feedback based on test results may be applied to modify test cases and testing protocols to improve testing accuracy.