Personal Taste Assessment System Using Preference Dependency Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for recommending consumer products, such as beverages, lack the ability to accurately assess whether a user will enjoy recommended items and fail to suggest unrelated items that the user may enjoy.
Innovation Solution
A system that predicts a person's preference for an item by accessing a user profile, identifying candidate items with matching characteristics, and processing these characteristics with a preference model to generate a predicted rating, which includes analyzing dependency patterns in the preference model to determine consistent, contrasting, or polarizing patterns of dependency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current recommendation systems use past purchase data or ratings to identify items, then they can provide basic recommendations, but they fail to accurately assess whether the user will truly enjoy the recommended items
Solution Approach 1:
The patent segments the preference assessment into multiple independent components: item characteristics (sensory attributes), user preference patterns (derived from rated items), and prediction models. This segmentation allows each component to be developed and optimized independently, improving overall assessment accuracy without proportionally increasing system complexity.
Solution Approach 2:
The patent transforms the recommendation approach by changing from simple item matching to multi-parameter preference prediction. It uses sensory characteristics (sweetness, acidity, bitterness, etc.) and user preference patterns as parameters to generate predictions, thereby improving measurement precision through comprehensive parameter analysis.
2Adaptability or versatility
If recommendation systems recommend items with direct relationship to previously purchased items, then they can provide relevant recommendations, but they fail to find and recommend apparently unrelated items that the consumer is likely to enjoy
Solution Approach 1:
The patent creates a universal preference prediction model that can assess user preferences across different item types (beverages, foods, etc.) using the same sensory characteristic framework. This universality allows the system to recommend both related and apparently unrelated items with consistent prediction accuracy, enhancing recommendation diversity without sacrificing precision.
Solution Approach 2:
The patent introduces sensory characteristics (sweetness, acidity, bitterness, etc.) as intermediary parameters that bridge between different item types. These intermediaries allow the system to find connections between apparently unrelated items by comparing their sensory profiles against user preferences, thereby improving adaptability while maintaining prediction accuracy.
3Loss of information
If the system analyzes dependency patterns in the preference model, then it can learn how users prefer items, but this increases the computational processing requirements
Solution Approach 1:
The patent extracts and focuses on specific dependency patterns (consistent, contrasting, polarizing) from the preference model rather than analyzing all possible patterns. This selective extraction reduces computational energy consumption while still capturing the essential preference information needed for accurate predictions.
Solution Approach 2:
The patent applies partial analysis by focusing on the most significant dependency patterns that drive user preferences. Rather than exhaustively analyzing all possible relationships, it identifies and processes the key patterns (consistent, contrasting, polarizing) that provide sufficient preference information with reduced computational overhead.
Data Source
AI summary
A personal taste assessment system recommends and predicts a person's preference for a consumable or other item. The system accesses a user profile for a person. The user profile includes a preference model representing associations between the person's ratings of items and a set of item characteristics. The system also accesses a database of characteristic values for a group of items, uses the identifying information to identify a candidate item having characteristic values whose properties match characteristics associated with the rated items that the person found to be appealing, and processes the characteristic values of the candidate item with the user profile to generate a predicted rating as a prediction of how the person would rate the identified candidate item. The system then causes an electronic device to output an identification of the candidate item and the predicted rating.