Spreadsheet Formula Recommendations Using Learned Table Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-technical users face challenges in authoring complex formulas in spreadsheets due to the need to understand function syntax and parameters, leading to low accuracy in existing formula prediction methods.
Innovation Solution
A deep learning-based approach that represents spreadsheets and regions as images, using machine learning models to identify similar spreadsheets and regions, and adapt formulas from reference cells to the target cell, enhancing accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If non-technical users attempt to author complex formulas manually, then they can create custom calculations, but they face challenges due to needing to understand function syntax and parameters
Solution Approach 1:
The system enables formulas to serve themselves by automatically generating and completing formula expressions based on user actions and spreadsheet context, eliminating the need for users to manually type or understand complex syntax
Solution Approach 2:
The patent replaces the mechanical process of manual formula typing and syntax memorization with an automated machine learning-based formula completion system that generates formulas intelligently based on contextual analysis
2Measurement precision
If existing formula prediction methods are used, then users receive automated suggestions, but accuracy remains low
Solution Approach 1:
The system transitions from traditional text-based formula prediction to a multi-dimensional approach that represents spreadsheets as images and uses spatial relationships, enabling more accurate prediction by capturing structural and contextual information beyond simple text patterns
Solution Approach 2:
The patent changes the fundamental parameters of formula prediction by using deep learning models trained on spreadsheet image representations, allowing the system to learn complex patterns and relationships that improve prediction accuracy while maintaining efficiency
3Measurement precision
If machine learning models use more training examples, then formula prediction quality improves, but training data generation becomes more complex
Solution Approach 1:
The system generates its own training data automatically by analyzing spreadsheet images and extracting formula patterns without requiring manual annotation, enabling the model to continuously improve through self-generated training examples
Solution Approach 2:
The patent performs preliminary actions by pre-processing spreadsheet images into structured representations and pre-generating training data before model training, simplifying the overall training process while enabling the use of large datasets
Data Source
AI summary
The present disclosure relates to methods and systems that automatically identify similar spreadsheets to target spreadsheets. The methods and systems automatically predict formulas that users want to author in a target cell of a target spreadsheet by identifying a reference formula from a similar region in a similar reference sheet that is similar to the target cell. The methods and systems generate a predicted formula by adapting parameters of a reference formula to a context of the target cell. The methods and systems provide an output with the predicted formula in the target cell of the target spreadsheet.