Dynamic Digital Component Segmentation for Design Space Exploration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data collection methods struggle to address bias and representativeness issues in unevenly distributed populations, leading to unrepresentative data sets that affect the accuracy of models and product design processes.
Innovation Solution
A method that dynamically alters digital components to solicit user feedback from underrepresented segments, using machine learning and AI to generate tasks and modify product designs based on user demographics, thereby improving data quality and exploring design spaces more effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data collection uses conventional methods without targeted segmentation, then the data collection process is simple, but the data set becomes unrepresentative and biased toward overrepresented populations
Solution Approach 1:
The patent segments the population into distinct groups based on demographic characteristics and data representation levels. It identifies underrepresented segments and targets them specifically for data collection, transforming a uniform data collection approach into a segmented, targeted strategy that improves representativeness while managing complexity through systematic division of the population base.
Solution Approach 2:
The system performs preliminary analysis to identify which population segments are underrepresented before initiating targeted data collection. By pre-segmenting the population and determining representation gaps in advance, the system can then focus resources on specific segments that need more data, rather than using a blanket collection approach.
2Measurement precision
If the system dynamically alters digital components to target underrepresented segments, then data quality improves, but the complexity of the data collection system increases
Solution Approach 1:
The system continuously monitors data collection results to identify segments that remain underrepresented despite previous collection efforts. This feedback loop allows the system to dynamically adjust its targeting strategy, altering digital components and collection approaches based on real-time assessment of representation gaps, thereby maintaining high data quality while adapting to changing needs.
Solution Approach 2:
The patent implements dynamic alteration of digital components based on identified representation gaps. The system adapts its data collection strategy in real-time, changing the nature and targeting of digital components deployed to specific segments based on current needs, rather than using a static, one-size-fits-all approach.
3Reliability
If conventional data collection methods are used, then the system is easier to operate, but bias in data sets cannot be effectively reduced
Solution Approach 1:
The system automatically performs segmentation, identification of underrepresented segments, and dynamic alteration of digital components without requiring manual intervention. The automated nature of these processes reduces the operational burden on users while maintaining the sophisticated functionality needed for bias reduction, making the complex system as easy to operate as conventional methods.
4Productivity
If the system segments visual representation and targets specific segments, then exploration of design spaces improves, but the complexity of processing increases
Solution Approach 1:
The patent segments the visual representation of design spaces into distinct regions corresponding to different design factor combinations. By identifying which segments lack sufficient data points, the system can target exploration efforts to specific regions of the design space, improving overall exploration efficiency while managing processing complexity through focused rather than exhaustive analysis.
Data Source
AI summary
Methods, systems, and computer readable medium include receiving, from a user device, a request for a digital component, receiving a data set of user-provided information regarding a particular product design, generating, based on the data set, a visual representation mapping design factors to potential product design geometry, segmenting the visual representation based on the design factor values, selecting a segment that contains less than a threshold amount of data points, selecting a digital component, dynamically altering, based on the selected segment, a presentation of the digital component that solicits information from the user about the segment, distributing, for presentation at the user device, the dynamically-altered digital component, obtaining, from the user device through a feedback mechanism, feedback information regarding the segment that contains less than the threshold amount of data points, and modifying a design factor of the particular product design based, at least in part, on the feedback information.