ARTIFICIAL INTELLIGENCE (AI) TRAINED DATA MODEL SELECTION

    公开(公告)号:WO2022115659A1

    公开(公告)日:2022-06-02

    申请号:PCT/US2021/060895

    申请日:2021-11-26

    摘要: This disclosure describes techniques for continuous improvement of machine learning models (also called data models) in a Content Management System (CMS). In one example, a CMS may store a set of data models for each application such as plate number recognition, facial recognition, a determination of likelihood of assault to a law enforcement officer in a traffic violation or robbery scenario, and car identification. In an example embodiment, a predictive model may be used to select a data model from the plurality of data models. The selected data model may be further improved or trained to a new sample of data features to generate an output pattern (e.g., likelihood of assault to a law enforcement officer).

    HIERARCHICAL DATA INGESTION IN A UNIVERSAL SCHEMA

    公开(公告)号:WO2022115658A1

    公开(公告)日:2022-06-02

    申请号:PCT/US2021/060894

    申请日:2021-11-26

    摘要: This disclosure describes techniques for creating a universal schema with default fields that support sensor formats of different devices. In one example, the universal schema supports substantial equivalents between data fields in different sensor formats. Further, a sensor format may be configured to support inheritance and aggregation of sensor formats in prior devices. Accordingly, the mapping of sensor formats that supports inheritance and aggregation in the universal schema may provide several advantages such as capturing a mapping of substantive equivalents between the fields in different sensor formats.

    NETWORK OPERATING CENTER (NOC) WORKSPACE INTEROPERABILITY

    公开(公告)号:WO2022115657A1

    公开(公告)日:2022-06-02

    申请号:PCT/US2021/060893

    申请日:2021-11-26

    发明人: GUZIK, Thomas

    摘要: This disclosure describes techniques for a consumer application that integrates a model-controller-view (MCV) design pattern with an event streaming platform such as an Apache KafkaTM in a network operation center (NOC) server to support NOC workspace interoperability. The MCV design pattern may include a pattern that divides an application into three main logical components (e.g., model component, controller component, and view component) to handle specific aspects of the application. In one example, the model component decouples the telemetry data streams from an event stream platform, and the controller component filters a queried set of decoupled telemetry data streams to dynamically control views to be rendered in the view component.

    DATA SOURCE CORRELATION TECHNIQUES FOR MACHINE LEARNING AND CONVOLUTIONAL NEURAL MODELS

    公开(公告)号:WO2022115656A1

    公开(公告)日:2022-06-02

    申请号:PCT/US2021/060892

    申请日:2021-11-26

    IPC分类号: G06N3/04 G06N3/08 G06N20/20

    摘要: A data model computing device receives a first data model with a first set of attributes, a first margin of error, a first set of predictions, and an underlying data set. Subsequently, the data model computing device receives a second data model with a second set of attributes, as the test data for a machine learning module. Based on the first and second data model, the machine learning function generates a second set of predictions and a second margin of error. The data model computing device performs a statistical analysis on the first and second set of predictions and the first and second margin of error to determine if the second set of predictions converge with the first set of predictions and second margin of error is narrower than the first margin of error, to determine if the second data model improves the prediction results of the machine learning module.

    DATA AGGREGATION WITH SELF-CONFIGURING DRIVERS

    公开(公告)号:WO2022115655A1

    公开(公告)日:2022-06-02

    申请号:PCT/US2021/060890

    申请日:2021-11-26

    发明人: GUZIK, Thomas

    摘要: A data aggregation implementation includes self-configuring drivers. From the viewpoint of a Network Operation Center (NOC), a plurality of heterogenous content sources provide content that may be of a variety of different types and formats. All of this content must be ingested and stored for retrieval and reporting, analysis, and/or presentation despite many differences in their collection, format, transmission, and quality. In some embodiments, the NOC includes or cooperates with one or more servers to, among other functions, receive content from content sources, request object reflection by the driver of each content source, receive driver attributes in response, and map the metadata of the content for each content source to a universal schema, thereby self-configuring the driver.

    DEVICE CONTROL USING ENTITY IDENTIFIERS
    9.
    发明申请

    公开(公告)号:WO2018200758A1

    公开(公告)日:2018-11-01

    申请号:PCT/US2018/029485

    申请日:2018-04-25

    发明人: GUZIK, Thomas

    IPC分类号: G08B21/02 G08B25/14 G08B27/00

    摘要: Techniques for efficiently and automatically activating portable devices include a device nomenclature schema used to associate devices with particular entities, including a person, a group of persons, a role, a location, a type, etc. A first device can be configured to activate one or more other devices according to assigned entity names when certain conditions occur. When a recording event occurs, metadata associated with one or more recording devices is stored, including entities that were activated during the recording event. A reviewer of the event recording can thereby easily determine other devices that might have stored data related to the event.

    OPTIMIZING CONTINUOUS MEDIA COLLECTION
    10.
    发明申请

    公开(公告)号:WO2023287646A1

    公开(公告)日:2023-01-19

    申请号:PCT/US2022/036444

    申请日:2022-07-08

    摘要: Described herein are techniques that may be used to identify a portion of media data to be prioritized. Such techniques may comprise receiving, from a media collection device, media information that includes a first media data and at least one of trigger data or sensor data, determining, based on one or more of the trigger data or the sensor data, that a portion of the first media data is to be prioritized, identifying, based on one or more of the trigger data or the sensor data, a beginning and end time to be associated with a second media data that includes the portion of the first media data, and generating the second media data from the received first media data based on the beginning and ending time, the second media data including less than the entirety of the first media data.