Food Sample Prediction Model Using Bayesian Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Predicting characteristic values for food samples is challenging due to complex interactions between ingredients, requiring numerous experiments and being computationally intensive, while conventional sensory panels for evaluating food analogs are impractical and inefficient.
Innovation Solution
A method that determines prototype characteristic values, trains a prediction model using Bayesian optimization, and recommends sample compositions to match target dairy fat melt profiles, utilizing intermediate products and weighting temperature-dependent characteristics for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If numerous experiments are conducted to predict characteristic values for food samples, then prediction accuracy is improved, but experimental time and computational resources are increased
Solution Approach 1:
The patent performs preliminary actions by measuring characteristic values for a comprehensive set of prototype samples and storing them in a database before actual predictions are needed. This pre-computed database allows rapid queries without repeating all experiments, thus improving prediction speed while maintaining accuracy.
Solution Approach 2:
The patent creates a virtual copy of the experimental data through a database that stores characteristic values for multiple prototype samples. Instead of physically conducting all experiments repeatedly, the system queries pre-stored data, effectively copying the experimental results for rapid prediction without additional experimental time.
2Measurement precision
If conventional sensory panels are used to evaluate food analogs, then evaluation is performed, but the process is impractical and inefficient
Solution Approach 1:
The patent replaces the mechanical/conventional sensory panel evaluation system with an automated computer-based prediction system. The system uses algorithms and pre-stored data to automatically predict characteristic values, eliminating the need for manual sensory panels and significantly improving evaluation efficiency while maintaining or enhancing accuracy.
Solution Approach 2:
The prediction system performs self-service by automatically querying its own database and calculating predictions without requiring external human evaluators. The system serves itself by using its pre-computed data and algorithms to provide evaluations, making the process efficient and scalable.
3Measurement precision
If complex interactions between ingredients are accounted for, then prediction accuracy is improved, but computational intensity increases
Solution Approach 1:
The patent performs preliminary computational work by pre-calculating and storing the effects of complex ingredient interactions in the database during the prototype development phase. During actual predictions, the system simply queries pre-computed results rather than performing complex calculations in real-time, thus reducing computational resource requirements while maintaining accuracy.
Data Source
AI summary
In variants, the method for determining a sample recommendation can include: determining characteristic values for a sample, determining target characteristic values, determining a similarity score for the sample based on the characteristic values for the sample and the target characteristic values, training a prediction model, and determining a sample recommendation.