Machine Learning Virtual Tables for Software Version Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face challenges in determining whether a newer software version meets their needs without having to staff a consultant or watch irrelevant product videos, as they often need to observe the new version's functionality on data not relevant to their own system.
Innovation Solution
A system utilizing machine learning techniques to create virtual tables that connect to a user's own data system, allowing comparison of existing and target software versions using the same underlying data, with a graphical user interface displaying both versions side-by-side, highlighting feature changes and allowing users to sample how the target version operates with their own data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If customers staff a consultant or watch product videos to evaluate new software versions, then they can observe functionality, but they cannot evaluate the new version on their own relevant data and incur additional time and cost
Solution Approach 1:
The system creates a simulated environment that copies the target software version's interface and functionality, allowing customers to evaluate it locally on their own data without needing external consultants or video demonstrations. The simulation replicates the software experience directly in the customer's environment.
Solution Approach 2:
A simulation engine acts as an intermediary between the customer's data and the target software version, enabling evaluation by translating the target version's functionality into a simulatable format that operates on the customer's local data without requiring direct access to the actual new software.
2Ease of operation
If customers use consultants to evaluate new software versions, then they can get expert guidance, but they incur additional costs and cannot independently assess the software
Solution Approach 1:
The system enables customers to independently evaluate new software versions through automated simulation technology. Customers can autonomously configure simulations, input their own data, and assess functionality without requiring consultant intervention, making the evaluation process self-service and eliminating dependency on external experts.
Solution Approach 2:
The system performs preliminary actions by pre-configuring the simulation environment, pre-processing customer data, and pre-establishing evaluation frameworks before the customer begins assessment. This preparation work simplifies the customer's task and reduces the complexity of independent evaluation.
3Adaptability or versatility
If customers evaluate software on generic demo data, then they can see functionality, but they cannot assess how the software will perform on their specific data
Solution Approach 1:
The system applies local quality by customizing the simulation to use the customer's specific local data rather than generic demo data. The evaluation is tailored to the customer's particular datasets, business logic, and operational context, ensuring that the assessment reflects real-world performance on their actual data.
Solution Approach 2:
The system changes parameters by allowing customers to input their own data characteristics, data formats, and business rules into the simulation. The simulation adapts its behavior based on these parameter changes, enabling evaluation that reflects the customer's specific data environment rather than fixed demo conditions.
Data Source
AI summary
In an example embodiment, machine learning techniques are utilized to create virtual tables that connect to actual tables in a user's own system. The virtual table predicts how the user's data can be used to populate fields in newer versions of software that the user already runs, even when those fields are not present in the version that the user already runs. These tables may then be used in a specialized tool, which displays in one area of the display a screen of the version of the software that the user is currently running (“the existing version”) and displays in another area of the display a screen of the version of the software that the user is comparing to the existing version. Both display areas display the same screen, as rendered by their respective different versions of the software, using the same underlying base data.