Dialogue state tracking using a global-local encoder

    公开(公告)号:US11836451B2

    公开(公告)日:2023-12-05

    申请号:US17179933

    申请日:2021-02-19

    Abstract: A method for maintaining a dialogue state associated with a dialogue between a user and a digital system includes receiving, by a dialogue state tracker associated with the digital system, a representation of a user communication, updating, by the dialogue state tracker, the dialogue state and providing a system response based on the updated dialogue state. The dialogue state is updated by evaluating, based on the representation of the user communication, a plurality of member scores corresponding to a plurality of ontology members of an ontology set, and selecting, based on the plurality of member scores, zero or more of the plurality of ontology members to add to or remove from the dialogue state. The dialogue state tracker includes a global-local encoder that includes a global branch and a local branch, the global branch having global trained parameters that are shared among the plurality of ontology members and the local branch having local trained parameters that are determined separately for each of the plurality of ontology members.

    Aggregate component for parallel browser-initiated actions

    公开(公告)号:US11797638B2

    公开(公告)日:2023-10-24

    申请号:US17249240

    申请日:2021-02-24

    CPC classification number: G06F16/958 G06F3/0482 G06F8/34 G06F8/38 G06F3/0486

    Abstract: Database systems and methods are provided for visual design and customization of processes for web applications. A method involves identifying a plurality of configured actions added to an aggregate action component region on a graphical user interface (GUI) display, automatically generating configured code for each of the configured actions in accordance with a respective user configuration of the respective configured action, identifying a user configuration associated with the aggregate action component, automatically generating aggregate action component code configurable to cause a browser application to initiate the configured code for each of the configured actions in parallel in accordance with the user configuration, and automatically generating process flow code including a reference to the aggregate action component code in accordance with a sequence defined for a web page.

    MACHINE LEARNING MODEL LAYER
    23.
    发明公开

    公开(公告)号:US20230333901A1

    公开(公告)日:2023-10-19

    申请号:US17659775

    申请日:2022-04-19

    CPC classification number: G06F9/5044 G06F9/5055 G06N20/00

    Abstract: Techniques are disclosed that pertain to facilitating the execution of machine learning (ML) models. A computer system may implement an ML model layer that permits ML models built using any of a plurality of different ML model frameworks to be submitted without a submitting entity having to define execution logic for a submitted ML model. The computer system may receive, via the ML model layer, configuration metadata for a particular ML model. The computer system may then receive a prediction request from a user to produce a prediction based on the particular ML model. The computer system may produce a prediction based on the particular ML model. As a part of producing that prediction, the computer system may select, in accordance with the received configuration metadata, one of a plurality of types of hardware resources on which to load the particular ML model.

    MACHINE LEARNING BASED SUPERVISED USER EXPERIENCE FOR AN APPLICATION MONITORED BY MULTIPLE SECONDARY APPLICATIONS

    公开(公告)号:US20230300091A1

    公开(公告)日:2023-09-21

    申请号:US18321667

    申请日:2023-05-22

    CPC classification number: H04L51/02 G06N20/00 H04L51/222 G06F3/0482

    Abstract: Disclosed is a system for managing content generated by bots for presentation to a user in association with a chat application. The system receives content items generated by bots monitoring a chat application for display to a user at a user interface (UI). The system provides input based on the received one or more content items and associated contextual information to a trained machine learning (ML) model, and receives, from the trained ML model, for each of the content items, at least one score value based on at least one predicted user response associated with potentially displaying the content item to the user at the UI. The system selects a subset of content items from the received content items based on the received score values and causes a display of this selected subset of content items in addition to a display of content generated by the chat application.

    NON-OBTRUSIVE MARKUP AUGMENTATION FOR WEBSITE LOCALIZATION

    公开(公告)号:US20230297764A1

    公开(公告)日:2023-09-21

    申请号:US17695107

    申请日:2022-03-15

    Inventor: Marcus Lenzner

    CPC classification number: G06F40/143 G06F40/117 G06F40/106 G06F40/40

    Abstract: Disclosed herein are system, method, and computer program product embodiments for providing a non-obtrusive markup augmentation of websites during website localization. A method includes generating a localization descriptor for a localization element on a website, encoding the localization descriptor into unparsed syntax in code for the website, adding pseudo tag notations to the localization descriptor encoded in the unparsed syntax, extracting text that is being localized from the localization descriptor by identifying the pseudo tag notations, and creating a visual indicator for display as an overlay outside a visible area of the website to indicate a change in appearance of the website after localization, wherein a size of the visual indicator is scaled to match a size of the text as displayed in the visible area of the website when the code is rendered.

    AUTOMATIC GRAPHQL DATA SOURCE FIELD CHANGE HANDLER

    公开(公告)号:US20230289339A1

    公开(公告)日:2023-09-14

    申请号:US17689601

    申请日:2022-03-08

    Inventor: Gaurav SAGAR

    CPC classification number: G06F16/2448 G06F16/252

    Abstract: A GraphQL server stores schemas for GraphQL APIs in a schema store, the schemas defining operations, types and fields of the APIs. Field resolvers are associated with the fields of the APIs to specify which of the data source is to be used to fulfill an API query request. When an event is detected that makes changes to a schema of any of the data sources, the schema store is accessed and an old field name is automatically replaced with a new field name in both the retrieved schemas and the field resolvers mapped to the retrieved schemas. Responsive to receiving an API request, it is determined whether the request has any field names that do not match the corresponding schema, and if so, the old field name is replaced with the new field name in the request.

Patent Agency Ranking