ML-Based UX Test Analysis System for Unstructured Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
User experience (UX) testing faces inefficiencies due to the need for manual analysis of unstructured data from UX tests, which can lead to sub-optimal product design choices and resource misallocation, whether conducted in-house or by third-party service providers.
Innovation Solution
A system leveraging machine learning to automate the selection, normalization, and synthesis of UX test results, using a themer to predict themes and a selector to choose representative quotations, thereby enhancing the scalability and insightfulness of UX test data processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis of unstructured UX test data is performed, then detailed qualitative insights can be obtained, but the process is time-consuming and inefficient
Solution Approach 1:
The patent introduces machine learning models as an intermediary between raw unstructured UX test data and human analysts. The ML models automatically process and structure the unstructured data, extracting key insights and organizing them in a standardized format that analysts can efficiently review, thereby reducing manual analysis time while maintaining insight quality
Solution Approach 2:
The patent replaces the mechanical manual analysis process with automated machine learning-based processing. The ML models perform the initial data processing, pattern recognition, and insight extraction tasks that would otherwise require manual human effort, significantly reducing the time required while preserving the depth of qualitative insights
2Reliability
If third-party service providers are used for UX testing, then expertise is leveraged, but the analysis process becomes expensive and cumbersome
Solution Approach 1:
The patent creates a universal automated analysis system that can handle multiple types of UX test data and produce standardized outputs applicable to various products and services. This multi-functional system replaces the need for specialized third-party providers for each specific testing engagement, reducing process complexity while maintaining expertise through the ML models
Solution Approach 2:
The patent enables organizations to perform their own UX test analysis using the automated ML-based system, eliminating the need to outsource to third-party providers. The system is designed to be self-sufficient, requiring minimal external assistance while providing expert-level analysis capabilities through the trained ML models
3Loss of information
If comprehensive UX test data is collected, then complete user experience understanding is achieved, but data processing complexity increases
Solution Approach 1:
The patent extracts only the most relevant and valuable information from comprehensive UX test data using ML models. The system identifies and extracts key patterns, themes, and insights while filtering out redundant or less important data, maintaining information completeness for critical aspects while reducing overall processing complexity
Solution Approach 2:
The patent segments the comprehensive UX test data into structured categories and components using ML processing. By dividing the large volume of unstructured data into manageable segments with standardized formats, the system maintains complete information while making the data much easier to process and analyze
Data Source
AI summary
Techniques are described herein for selecting, curating, normalizing, enriching, and synthesizing the results of user experience (UX) tests. In some embodiments, a system identifies a set of expectation elements associated with one or more UX tests. An expectation element may specify, using unstructured data that does not conform to a schema, an expectation for a user experience and a respective outcome for the user experience. A themer model may generate predictions that map the respective expectation elements to a theme from a theme schema, which may include a plurality of themes. A selector model may generate selection scores for the expectation elements. The predicted themes and selection scores may be used to render user interfaces and/or trigger other actions directed to optimizing a product's design.