Test Scenario Template Generation from Cross-Organization Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software testing across different organizations often results in redundant efforts, as each organization develops its own testing suite and fails to leverage cumulative testing experience from others, leading to inefficiencies in selecting and generating appropriate tests from a large body of testing data.
Innovation Solution
A system and method for generating and suggesting test scenario templates based on data collected from multiple organizations, allowing for the clustering of similar test runs and the creation of templates that can be tailored to specific user profiles, facilitating the sharing and utilization of testing data across organizations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If each organization develops its own testing suite independently, then each organization can customize tests to its specific business processes, but redundant testing efforts occur and cumulative testing experience is not shared across organizations
Solution Approach 1:
The patent creates a universal test scenario template that can be applied across multiple organizations. The template captures common testing patterns and procedures that are reusable across different organizational contexts, allowing the same template to serve multiple organizations while maintaining adaptability to their specific needs through parameter customization.
Solution Approach 2:
The patent merges testing experiences from multiple organizations by collecting test scenario data from crowd users across different organizations, analyzing common patterns, and consolidating them into a single unified template. This combining approach eliminates redundant testing efforts while preserving organization-specific requirements.
2Loss of information
If testing data is collected from multiple organizations, then cumulative testing experience can be leveraged, but it becomes problematic to select and generate appropriate tests from the large body of available testing data
Solution Approach 1:
The patent introduces an intermediary processing system that acts as a mediator between the large body of collected testing data and the final template generation. This intermediary system analyzes test scenario data from multiple organizations, identifies common patterns, and filters out organization-specific variations, thereby simplifying the selection and generation process while preserving valuable testing experience.
Solution Approach 2:
The patent extracts essential testing patterns and procedures from the large body of collected testing data by identifying common elements across different organizations. The extraction process separates universal testing requirements from organization-specific details, creating a streamlined template that captures the essential testing experience without the complexity of the full dataset.
3Ease of operation
If test scenario templates are generated based on clustered test runs from multiple organizations, then templates can be tailored to specific user profiles, but the process of clustering and tailoring increases system complexity
Solution Approach 1:
The patent performs preliminary clustering of test runs from multiple organizations before template generation. By pre-grouping similar test scenarios and identifying common patterns in advance, the system reduces the complexity of subsequent template customization. The preliminary analysis creates a structured foundation that makes the final template generation and user-specific tailoring more straightforward.
Data Source
AI summary
System, method, and non-transitory medium for generating a test scenario template from a cluster of similar partial runs of test scenarios involves executing the following: identifying runs of test scenarios run by users on software systems that belong to different organizations; generating partial runs of the test scenarios by extracting from each run of a test scenario data pertaining to a proper subset of the plurality of test steps described in the run; clustering the partial runs of the test scenarios into clusters of similar partial runs; receiving a profile of a user; selecting, based on the profile, from among the clusters a certain cluster suitable for the user; and generating a test scenario template based on partial runs belonging to the certain cluster.