AutoAI Problem Statement Generation via Dataset Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AutoAI/AutoML systems assume users have sufficient domain knowledge to formulate computable data science problems, leading to potential mistakes in problem formulation, such as using meaningless features or inappropriate algorithms, due to the lack of automated formulation of data science problems based on input datasets.
Innovation Solution
A method that automatically formulates data science problem statements by identifying matching prior datasets, generating natural language problem statements, and configuring AutoAI/AutoML settings using machine learning techniques, enabling the system to learn and improve its problem formulation based on the dataset itself.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If AutoAI systems assume users have sufficient domain knowledge to formulate data science problems, then the system complexity is reduced, but the reliability of problem formulation deteriorates due to potential mistakes in problem formulation
Solution Approach 1:
The system automatically generates problem statements, feature lists, and model configurations by analyzing the input dataset itself, rather than relying on user expertise. The AutoAI system serves itself by autonomously formulating computable data science problems based on dataset characteristics, eliminating the need for users to have domain knowledge while maintaining high reliability through automated analysis
Solution Approach 2:
The system performs preliminary analysis of the dataset to identify its characteristics, relationships, and suitable modeling approaches before the actual model building process. By pre-generating problem statements and feature lists based on dataset exploration, the system ensures reliable problem formulation is established before users interact with the system
2Reliability
If automated formulation of data science problems is implemented, then the reliability of problem formulation is improved, but the device complexity increases due to additional automation components
Solution Approach 1:
The system employs a unified dataset analysis engine that performs multiple functions: exploring data characteristics, generating problem statements, identifying features, and suggesting model configurations. This multi-functional approach achieves reliable automated problem formulation without proportionally increasing system complexity, as the same core components serve multiple purposes throughout the AutoAI pipeline
3Ease of operation
If users manually formulate data science problems, then the ease of operation is maintained, but the manufacturing precision of model generation deteriorates due to mistakes in problem formulation
Solution Approach 1:
The system automatically generates problem statements, feature lists, and model configurations by analyzing the input dataset itself, rather than relying on user expertise. The AutoAI system serves itself by autonomously formulating computable data science problems based on dataset characteristics, eliminating the need for users to have domain knowledge while maintaining high reliability through automated analysis
Solution Approach 2:
The system incorporates user feedback loops where generated problem statements and configurations can be reviewed, corrected, and refined. This feedback mechanism maintains ease of operation by allowing user oversight while improving manufacturing precision through iterative refinement based on user input and system learning
Data Source
AI summary
A method comprises receiving a new data set; identifying at least one prior data set of a plurality of prior data sets that matches the new data set; generating a natural language data science problem statement for the new data set based on information associated with the at least prior one data set that matches the new data set; outputting the generated natural language data science problem statement for user verification; and in response to receiving user input verifying the natural language generated data science problem statement, generating one or more AutoAI configuration settings for the new data set based on one or more AutoAI configuration settings associated with the at least one prior data set that matches the new data set.