Meal Recommendation System Using Transaction Data to Reduce User Input

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing meal recommendation systems require extensive user input, leading to irrelevant, incomplete, or inaccurate results, and often fail to customize meals based on individual user needs and available ingredients.

Innovation Solution

A system utilizing multiple machine learning models to generate meal recommendations by analyzing user input and transaction data to create a food profile, estimate available ingredients, and provide cooking instructions or dining options based on ingredient availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If meal recommendation systems require extensive user input to determine available ingredients and preferences, then the system can provide more accurate recommendations, but the time and effort required from users increases significantly

Engineering Contradiction:

Improveaccuracy of meal recommendationsVSAvoidtime required for user input

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by automatically analyzing transaction data from the user's account to identify ingredients they have available, eliminating the need for users to manually input this information. The system proactively generates ingredient lists and meal recommendations based on past purchases, so users don't need to spend time entering data before receiving recommendations

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system enables self-service by automatically tracking the user's ingredient inventory through transaction data analysis and autonomously generating meal recommendations without requiring user intervention. The system serves itself by continuously monitoring the user's account, updating ingredient availability, and providing recommendations based on automated data processing

Inventive Principle:

Principle #25Self-service

2Adaptability or versatility

If meal recommendation systems rely heavily on user input data, then they can capture user preferences and dietary restrictions, but the quality and relevance of recommendations decrease when user input is incomplete or inaccurate

Engineering Contradiction:

Improvecustomization to user needsVSAvoidaccuracy of ingredient availability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system implements feedback by continuously monitoring the user's account transaction data to verify ingredient availability and update recommendations. The system cross-references purchase history with recipe requirements, providing ongoing validation and adjustment of meal suggestions based on actual ingredient possession, which improves both reliability and adaptability

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system achieves universality by using transaction data analysis as a multi-functional solution that simultaneously captures ingredient availability, infers user preferences through purchase patterns, and validates dietary restrictions. This single data source serves multiple purposes, reducing reliance on potentially incomplete user input while maintaining customization

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Ease of operation

If the system provides detailed meal recommendations based on extensive analysis, then the helpfulness and relevance increase, but the complexity of the system increases

Engineering Contradiction:

Improvehelpfulness of recommendationsVSAvoidsystem complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The system extracts only the essential information needed for recommendations from the user's transaction data, such as ingredient availability and purchase patterns. By pulling out specific relevant data points rather than processing all available information, the system reduces computational complexity while maintaining recommendation quality and ease of use

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS12288147B2Systems and methods for generating meal recommendations

Publication Date: 2025.04.29 CAPITAL ONE SERVICES LLC

AI summary

Disclosed embodiments may include a method for generating meal recommendations by receiving first data and user input data. The system may generate a food profile and estimate of ingredients based on the first data and the user input data. The system may generate, based on the estimate of ingredients, one or more cooking instructions or an output that the user does not have sufficient ingredients. If cooking instructions are generated, the system may generate and transmit a graphical user interface comprising the one or more cooking instructions based on the estimate of ingredients for display. If it is determined the user does not have sufficient ingredients, the system may generate and transmit a graphical user interface comprising one or more dining options for the user for display.