20240013089. Sequential Synthesis and Selection for Feature Engineering simplified abstract (Capital One Services, LLC)
Sequential Synthesis and Selection for Feature Engineering
Organization Name
Inventor(s)
Michael Langford of Plano TX (US)
Sequential Synthesis and Selection for Feature Engineering - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240013089 titled 'Sequential Synthesis and Selection for Feature Engineering
Simplified Explanation
The patent application describes systems and methods for sequential synthesis and selection for feature engineering in machine learning. It involves applying operations to existing features of a dataset to generate potential features. A feature importance algorithm is then applied to each potential feature along with the existing features to determine their importance values. These values are used to sort the potential features. Additionally, the correlation between each potential feature and the existing features is evaluated to ensure it is below a certain threshold level, avoiding heavily correlated new features.
- Sequential synthesis and selection for feature engineering in machine learning
- Applying operations to existing features of a dataset to generate potential features
- Applying a feature importance algorithm to determine the importance values of potential features
- Sorting potential features based on their importance values
- Evaluating the correlation between potential features and existing features to avoid heavily correlated new features
Potential Applications
- Machine learning and data analysis tasks
- Feature engineering in various domains such as image recognition, natural language processing, and predictive modeling
Problems Solved
- Enhancing the feature engineering process in machine learning
- Avoiding heavily correlated new features that may negatively impact model performance
Benefits
- Improved accuracy and performance of machine learning models
- Efficient and automated feature engineering process
- Reduction of manual effort and time required for feature selection and synthesis
Original Abstract Submitted
systems and methods, as described herein, relate to sequential synthesis and selection for feature engineering. a dataset may be associated with a label defining a machine-learning target attribute and a received operation that can be applied to at least one of the existing features of the dataset. one or more potential features may be generated by applying the operation to one or more existing features. for each of the one or more potential features, a feature importance algorithm may be applied to the respective feature along with the one or more existing features, generating a respective feature importance value. respective feature importance values may be generated for each of the one or more existing features based on applying the feature importance algorithm and used to sort the potential features. a level of correlation to each of the one or more existing features may be determined to make sure it is under a threshold level to avoid new features heavily correlated to existing ones.
