Non-transitive Multicomponent Score Item Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing matchmaking algorithms, such as the Bradley-Terry model, exhibit a transitive property, which limits their effectiveness in selecting items based on multicomponent scores, as they do not allow for nontransitive relations between items.
Innovation Solution
A method and device for selecting items based on multicomponent scores that compute a value representing a relative ranking relation between items, using a score function that enables non-transitive relations, allowing for the refinement of selection and ordering on a non-transitive basis, with components that are non-discriminative and can vary over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If transitive matchmaking algorithms (Bradley-Terry model) are used to rank items, then the ranking process is simple and deterministic, but the selection accuracy is limited because it cannot capture nontransitive relations between items
Solution Approach 1:
The patent segments the score into multiple independent components (S1, S2, ..., Sm) rather than using a single transitive score. Each component can capture different aspects of item quality, allowing nontransitive relations to emerge from the combination of these segmented components. This segmentation enables the system to represent complex preferences that a single score cannot capture.
Solution Approach 2:
The patent transitions from a one-dimensional transitive score to a multi-dimensional score space with m components. By adding these additional dimensions, the system can represent items in a higher-dimensional space where nontransitive relations become possible and meaningful, allowing for more accurate selection while maintaining computational tractability.
2Adaptability or versatility
If a single transitive score is assigned to each item, then the computation is efficient and straightforward, but the system cannot handle multicomponent criteria that vary over time
Solution Approach 1:
The patent introduces dynamics by allowing the score components to vary over time and by enabling the weight vector w to be adjusted based on different selection criteria. This dynamic approach allows the system to adapt to changing conditions and preferences while maintaining efficient computation through the structured form of the multicomponent score.
Solution Approach 2:
The patent changes parameters by introducing multiple score components and weights that can be adjusted independently. This allows the system to adapt to different criteria and time-varying conditions by modifying the weight vector and component values, providing versatility without sacrificing computational efficiency due to the structured nature of the calculations.
3Measurement precision
If nontransitive relations are introduced to improve selection accuracy, then the matchmaking process becomes more sophisticated, but the transitive property is lost which simplifies the original ranking structure
Solution Approach 1:
By segmenting the score into multiple independent components, the patent enables nontransitive relations to emerge naturally from the combination of these components. Each component maintains its own stability and meaning, while their combination allows for sophisticated ranking that captures nontransitive preferences without creating instability in the individual components.
Data Source
AI summary
The provided solution makes it possible to select one item from a set of items, a multicomponent score being associated with each item of the set of items. After having computed a value characterising a relation ranking between a first item of the set of items and a second item of the set of items, the computed value being computed as a function of a plurality of score components associated with the first and second item of the set of items, respectively, the first item or of the second item is selected based on the computed value.