Patents by Inventor Kevin Andrew PERKINS

Kevin Andrew PERKINS 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).

  • Patent number: 11817994
    Abstract: One or more computing devices, systems, and/or methods for time series trend root cause identification are provided. In particular, an overall trend of multi-dimensional time series data and element trends for measured elements of dimensions within the multi-dimensional time series data is identified. Weighted correlations between the element trends of the measured elements and the overall trend are calculated. The weighted correlations of the measured elements and aggregate weighted correlations of measured element combinations are evaluated to identify a set of measured elements having a threshold correlation to the trend.
    Type: Grant
    Filed: January 25, 2021
    Date of Patent: November 14, 2023
    Assignee: YAHOO ASSETS LLC
    Inventors: Jifu Zhao, Kevin Andrew Perkins, Mithilesh Nanjamanaidu Srinivasan Rangavadivel, Matthew Robert Ahrens
  • Publication number: 20230153987
    Abstract: One or more computing devices, systems, and/or methods for defect detection are provided. An image, depicting an object for evaluation to determine whether the object has a defect, is inputted into a segmentation model to identify an object region of interest of the object. An object region area of the object region of interest is calculated. A convex hull area of a convex hull encompassing the object region of interest is calculated. A ratio of the object region area to the convex hull area is determined. The ratio is compared to a threshold to determine whether the object has the defect or does not have the defect.
    Type: Application
    Filed: November 17, 2021
    Publication date: May 18, 2023
    Inventors: Kevin Andrew PERKINS, Ashiwini Kumar Kounduri
  • Publication number: 20230088722
    Abstract: One or more computing devices, systems, and/or methods are provided. In an example, a dataset associated with a plurality of dimensions and a plurality of metrics is identified. A subset of dimensions of the plurality of dimensions and a subset of metrics of the plurality of metrics are selected based upon historical dataset queries. A plurality of sets of results is generated, using the dataset, based upon the subset of dimensions and the subset of metrics. A plurality of significance scores is determined based upon the plurality of sets of results. One or more first sets of results of the plurality of sets of results are selected based upon the plurality of significance scores.
    Type: Application
    Filed: September 23, 2021
    Publication date: March 23, 2023
    Inventors: Matthew Robert AHRENS, Jifu ZHAO, Kevin Andrew PERKINS, Ashwini Kumar KOUNDURI, Mithilesh Nanjamanaidu Srinivasan Rangavadi
  • Patent number: 11580325
    Abstract: One or more computing devices, systems, and/or methods for hyper parameter optimization for machine learning ensemble generation are provided. For example, one or more base models are trained using diverse sets of hyper parameters, wherein different sets of hyper parameters (e.g., hyper parameters with different values) are used to train different base models. A matrix, populated with predictions from the set of base models, is generated. A machine learning ensemble is generated by processing the matrix utilizing a meta learner.
    Type: Grant
    Filed: January 25, 2019
    Date of Patent: February 14, 2023
    Assignee: YAHOO ASSETS LLC
    Inventor: Kevin Andrew Perkins
  • Publication number: 20220253647
    Abstract: One or more computing devices, systems, and/or methods are provided. Machine learning model training may be performed using first training data to generate a first machine learning model. Inference-training data may be generated, wherein the inference-training data may include a plurality of sets of training data of the first training data, a plurality of sets of inference data, and/or target information indicative of the plurality of sets of training data being associated with a first classification and the plurality of sets of inference data being associated with a second classification. Machine learning model training may be performed using the inference-training data to generate a second machine learning model. Predictions associated with one or more sets of data may be determined using the second machine learning model. An evaluation of the first machine learning model and the one or more sets of data may be generated based upon the predictions.
    Type: Application
    Filed: February 5, 2021
    Publication date: August 11, 2022
    Inventor: Kevin Andrew Perkins
  • Publication number: 20220239549
    Abstract: One or more computing devices, systems, and/or methods for time series trend root cause identification are provided. In particular, an overall trend of multi-dimensional time series data and element trends for measured elements of dimensions within the multi-dimensional time series data is identified. Weighted correlations between the element trends of the measured elements and the overall trend are calculated. The weighted correlations of the measured elements and aggregate weighted correlations of measured element combinations are evaluated to identify a set of measured elements having a threshold correlation to the trend.
    Type: Application
    Filed: January 25, 2021
    Publication date: July 28, 2022
    Inventors: Jifu Zhao, Kevin Andrew Perkins, Mithilesh Nanjamanaidu Srinivasan Rangavadivel, Matthew Robert Ahrens
  • Patent number: 10862782
    Abstract: One or more computing devices, systems, and/or methods are provided. Activity of one or more client devices may be analyzed to detect one or more sets of network traffic. A set of network traffic may comprise transmission of data by a client device to one or more first hosts and/or reception of data by the client device from one or more second hosts. The one or more sets of network traffic may be analyzed to generate a set of network traffic information associated with a first application. The set of network traffic information may be indicative of a first set of hosts associated with the first application. It may be determined that first network traffic associated with a client device is associated with the first application based upon the first network traffic and the set of network traffic information associated with the first application.
    Type: Grant
    Filed: April 2, 2019
    Date of Patent: December 8, 2020
    Assignee: Oath Inc.
    Inventors: Kevin Andrew Perkins, Mithilesh Nanjamanaidu Sriniva, Aaron John Klish, Matthew Robert Ahrens
  • Publication number: 20200322240
    Abstract: One or more computing devices, systems, and/or methods are provided. Activity of one or more client devices may be analyzed to detect one or more sets of network traffic. A set of network traffic may comprise transmission of data by a client device to one or more first hosts and/or reception of data by the client device from one or more second hosts. The one or more sets of network traffic may be analyzed to generate a set of network traffic information associated with a first application. The set of network traffic information may be indicative of a first set of hosts associated with the first application. It may be determined that first network traffic associated with a client device is associated with the first application based upon the first network traffic and the set of network traffic information associated with the first application.
    Type: Application
    Filed: April 2, 2019
    Publication date: October 8, 2020
    Inventors: Kevin Andrew Perkins, Mithilesh Nanjamanaidu Sriniva, Aaron John Klish, Matthew Robert Ahrens
  • Publication number: 20200242400
    Abstract: One or more computing devices, systems, and/or methods for hyper parameter optimization for machine learning ensemble generation are provided. For example, one or more base models are trained using diverse sets of hyper parameters, wherein different sets of hyper parameters (e.g., hyper parameters with different values) are used to train different base models. A matrix, populated with predictions from the set of base models, is generated.
    Type: Application
    Filed: January 25, 2019
    Publication date: July 30, 2020
    Inventor: Kevin Andrew PERKINS