Patents by Inventor Terrence Kunstek

Terrence Kunstek has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Technologies for unsupervised data classification with topological methods
    Patent number: 11797578
    Abstract: Technologies for unsupervised data classification include a computing device that generates a high dimensional profile for each column of a data store that is indicative of topological features of the column. The computing device also analyzes each column using multiple predetermined profiling rules and generates an initial classification for the columns based on profiling rule results. The initial classification may be applied to related columns based on clustering of the columns using the high dimensional profile. The computing device may train a machine learning model based on the initial classification and the high dimensional profile. Input features for training may include the profiling rule results and the high dimensional profile. Classifications may be provided to client devices via a predetermined interface. Other embodiments are described and claimed.
    Type: Grant
    Filed: November 3, 2021
    Date of Patent: October 24, 2023
    Assignee: Second Sight Data Discovery, LLC
    Inventors: Reuben Vandeventer, Terrence Kunstek
  • TECHNOLOGIES FOR UNSUPERVISED DATA CLASSIFICATION WITH TOPOLOGICAL METHODS
    Publication number: 20220138229
    Abstract: Technologies for unsupervised data classification include a computing device that generates a high dimensional profile for each column of a data store that is indicative of topological features of the column. The computing device also analyzes each column using multiple predetermined profiling rules and generates an initial classification for the columns based on profiling rule results. The initial classification may be applied to related columns based on clustering of the columns using the high dimensional profile. The computing device may train a machine learning model based on the initial classification and the high dimensional profile. Input features for training may include the profiling rule results and the high dimensional profile. Classifications may be provided to client devices via a predetermined interface. Other embodiments are described and claimed.
    Type: Application
    Filed: November 3, 2021
    Publication date: May 5, 2022
    Inventors: Reuben Vandeventer, Terrence Kunstek