Intel corporation (20240134884). METHODS AND APPARATUS FOR RECOMMENDATION SYSTEMS WITH ANONYMIZED DATASETS simplified abstract

METHODS AND APPARATUS FOR RECOMMENDATION SYSTEMS WITH ANONYMIZED DATASETS

Organization Name

intel corporation

Inventor(s)

Chendi Xue of Austin TX (US)

Jian Zhang of Shanghai (CN)

Poovaiah Manavattira Palangappa of San Jose CA (US)

Rita Brugarolas Brufau of Hillsboro OR (US)

Ke Ding of Saratoga CA (US)

Ravi H. Motwani of Fremont CA (US)

Xinyao Wang of Shanghai (CN)

Yu Zhou of Shanghai (CN)

Aasavari Dhananjay Kakne of Santa Clara CA (US)

METHODS AND APPARATUS FOR RECOMMENDATION SYSTEMS WITH ANONYMIZED DATASETS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240134884 titled 'METHODS AND APPARATUS FOR RECOMMENDATION SYSTEMS WITH ANONYMIZED DATASETS

Simplified Explanation

The patent application describes a system and method for preserving privacy in a user dataset by selecting a subset of user data features for a feature selection training model.

• Interface circuitry, machine readable instructions, and programmable circuitry are used to determine data usage types for user data features.
• The data usage types are classified into feature categories, and feature engineering mechanisms are applied to these categories.
• A subset of user data features is selected based on the application of feature engineering, and a second dataset with fewer features is output for the training model.

Potential Applications

This technology could be applied in industries where privacy of user data is crucial, such as healthcare, finance, and marketing.

Problems Solved

This technology addresses the challenge of preserving privacy in large user datasets while still maintaining the utility of the data for training models and analysis.

Benefits

The system allows for the selection of a subset of user data features that balance privacy concerns with the need for accurate and effective training models.

Potential Commercial Applications

"Privacy-Preserving Feature Selection Model for User Datasets"

Possible Prior Art

There may be prior art related to feature selection techniques in machine learning and data privacy preservation methods.

How does this technology compare to existing methods for feature selection in user datasets?

This technology focuses specifically on preserving privacy while selecting features for training models, which may differentiate it from traditional feature selection techniques.

What are the potential limitations or challenges of implementing this system in real-world applications?

One potential challenge could be ensuring that the selected subset of user data features still provides enough information for accurate model training without compromising privacy.

Original Abstract Submitted

systems, apparatus, articles of manufacture, and methods are disclosed to preserve privacy in a user dataset including interface circuitry, machine readable instructions, and programmable circuitry to determine a data usage type for each one of a plurality of user data features in a first dataset, classify the data usage type associated with each user data feature of the plurality of user data feature into a feature category, apply at least one feature engineering mechanism to feature categories of the data usage types of the plurality of user data features, select, based on application of feature engineering, a subset of the plurality of user data features for a feature selection training model, and output a second dataset based on the subset of the plurality of user data for the feature selection training model, the second dataset to include fewer user data features than the first dataset.