Abstract: Systems and methods for automated, degradation-resistant, tuning of machine-learning (“ML”) models are provided. The systems and methods may identify an inoperative input utterance, retrieve a feature set associated with the inoperative input utterance, and generate an updated utterance-feature-intent (“UFI”) mapping based on the retrieved feature set. The systems and methods may retrain the ML model using the updated UFI mapping, and compare the accuracy of the system after the retraining and before the retraining. In a scenario where the accuracy of the system does not improve, the systems and methods may amplify the updated UFI mapping. In a scenario where the accuracy of the system does improve, the systems and methods may deploy the updated UFI mapping.

Abstract: A system and method for simplifying the creation of documentation, and especially software documentation. A software application, referred to as a source metadata tagger and document compiler is used to add metadata to a final output document. This metadata contains identifiers that are associated with various source files. In this way, the system can easily determine which source file is being reviewed and/or flagged by the reviewer. This information can be used by the ticketing/notification system to create a work item for the appropriate developer or development group. This is vastly simpler than the current system, where human intervention is required to determine which source file is being flagged.

Abstract: Embodiments are disclosed for generating an intelligent change summary are described. In some embodiments, a method of generating an intelligent change summary includes obtaining a representation of a plurality of versions of a document, determining a distance score based on a comparison of a first of version of the document and a second version of the document, the distance score representing a magnitude of changes made from the first version of the document to the second version of the document, and generating a change summary of the document based on the distance score.

Abstract: A channel graphical user interface (GUI) of a content sharing platform is provided for presentation to a channel owner. A request is received, via the channel GUI, to enable an online community option that (i) facilitates an online conversation between the channel owner and a subset of viewers of the channel of the channel owner on the content sharing platform, and (ii) allows the channel owner to control viewer responses that are permitted during the online conversation. In response to the request of the channel owner to enable the online community option, the online community option is associated with the channel in a data store, and the channel GUI is modified, the modified channel GUI presenting an online community GUI element representing the online community option.

Abstract: An electronic device having a display and a touch-sensitive surface displays a mediate item representation corresponding to a sequence of images in a first display mode of user-selectable display modes, where the media item is part of a collection of ordered media items such that prior media item(s) precede the first media item and subsequent media item(s) follow the first media item in the order. In response to detecting touch input including movement in a direction, the device: upon determining that a subsequent media item corresponding to a second sequence of images is designated as being displayed in the first display mode, displays the subsequent media item in the first display mode, and upon determining that the subsequent media item is designated as being displayed in a second display mode different from the first display mode, displays the subsequent media item in the second display mode.

Abstract: Methods and systems for artificial intelligence (AI)-assisted document annotation and training of machine learning-based models for document data extraction are described. The methods and systems described herein take advantage of a continuous machine learning approach to create document processing pipelines that provide accurate and efficient data extraction from documents that include structured text, semi-structured text, unstructured text, or any combination thereof.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed to optimizing engagement with a display during digital assistant-performed operations in response to a received command. The digital assistant generates an overlay having user interface elements that present information determined to be relevant to a user based on the received command and contextual data. The overlay is presented over the underlying operations performed on corresponding applications to mask the visible steps of the operations being performed. In this way, the digital assistant optimizes display resources that are typically rendered useless during the processing of digital assistant-performed operations.

Abstract: Systems, apparatuses, methods, and computer program products are disclosed for at least semi-automated generation of model documentation. In an example embodiment, a template corresponding to a model is selected. User input populating portions of a build notebook is received. The build notebook comprises a model identifier identifying the model and executable code. The build notebook is executed using an execution engine to cause an intermediary file encoded in a type-setting system to be generated. Executing the build notebook causes cross-references of elements of the model documentation to be mapped to appropriate locations within the intermediary file. The elements comprise graphical representations of results of model tests executed on the model and/or extracted or user-defined artifacts required for the model documentation.

Abstract: Systems and methods for efficiently identifying and extracting machine-actionable structured data from web documents are provided. The technology employs neural network architectures which process the raw HTML content of a set of seed websites to create transferable models regarding information of interest. These models can then be applied to the raw HTML of other websites to identify similar information of interest. Data can thus be extracted across multiple websites in a functional, structured form that allows it to be used further by a processing system.

Abstract: Systems and methods enable a computing system to recognize a sequence of repeated actions and offer to automatically repeat any such recognized actions. An example method includes determining a current sequence of user actions is similar to previous sequence of user actions, determining whether the previous sequence is reproducible and, when reproducible, initiating display of a prompt that requests approval for completing the current sequence based on the previous sequence and, responsive to receiving an indication of approval, completing the previous sequence. Another example method included determining that a first current sequence of user interactions is complete and is not similar to any saved sequence of user interactions, saving the first current sequence as a previous sequence, identifying a second sequence as satisfying a similarity threshold with the previous sequence, and initiating display of a prompt that requests approval for saving the previous sequence as a shortcut.

Abstract: A method and apparatus for stimulating neural activity in the brain of a user of an apparatus with a display screen by causing at least one portion of the display screen to flicker in a controlled manner and utilizing the apparatus to measure an effect on a user exposed to the flicker for a time.

Abstract: Systems, methods, apparatuses and non-transitory, computer-readable storage mediums are disclosed for a media editing application with an anchored timeline for editing captions and subtitles. In an embodiment, a method comprises: causing to display on a display device, by one or more hardware processors executing a media editing application, a graphical user interface (GUI) including a timeline; causing to display on the timeline, by the one or more hardware processors, a first lane including a media object; causing to display on the timeline, by the one or more hardware processors, a second lane including a caption or subtitle object; and anchoring, by the one or more hardware processors, the media object to the caption or subtitle object in the timeline.

Abstract: A computer navigation system that represents categories and sub-categories of material with functionally labeled objects connected in a hierarchical organization. Descending nodes represent more detailed subsets of material. Hovering the computer mouse over a node displays thumbnails of the items available under it, enabling the user to search for items without opening the respective pages, transforming a literal or verbal data filing and retrieval process into a more intuitive and easy to use visual one. Once the user hovers the computer mouse over the node with the desired item and sees its thumbnail the user clicks on the node, opening the screen from which the user can access that item (or upload a similar one), or clicks on the thumbnail to access the item directly. In addition, the system employs algorithms to group items into various categories and sub-categories and assigns them to the appropriate nodes.

Abstract: An Artificial Intelligence (AI)-based data transformation system receives an input package and enables automatic execution of one or more processes in a robotic process automation system (RPA). The input package includes a plurality of documents and metadata required for the execution of the automated processes. The plurality of documents are categorized into a domain. Entities with their corresponding name-value pairs and entity relationships are extracted from the plurality of documents. An ontology is selected based on the domain. The entities are mapped to output fields identified from the selected ontology. The mappings thus generated are transmitted to the RPA system which employs the mappings to automatically execute the one or more processes.

Abstract: A system and method of displaying complementary content on one or more linked machines are disclosed. In some embodiments, the system and method may include a non-transitory, computer-readable medium storing computer-executable instructions and one or more processors in communication with the non-transitory, computer readable medium. When the computer-executable instructions are executed, the one or more processors may be configured to receive a linking instruction to link a display of second content of a website on a second machine to a selection of a portion of first content of the website by a first machine, cause a display of the first content on the first machine, receive the selection of the portion of the first content displayed on the first machine, and based on the portion of the first content being selected, cause the display of the second content on the second machine based on the linking instruction.

Abstract: Systems, methods and computer-readable storage media including an interactive surface, a plurality of cameras, a network circuit, and a processing circuit configured to receive an indication of a customer device. The processing circuit is further configured to authenticate the customer. The processing circuit is further configured to determine that a document associated with the customer is on or proximate the interactive surface. The processing circuit is further configured to receive a plurality of digital images. The processing circuit is further configured to create a master digital image. The processing circuit is further configured to save the master digital image. The processing circuit is further configured to update the account of the customer. The processing circuit is further configured to update the account of the customer generate and provide a graphical representation.

Abstract: A computer system document according to one embodiment includes a generation unit and an update unit. The generation unit performs first machine learning on a target document, as input data, to which a correct path in a tree structure where each node indicates a document category is given, and thereby generates a classification model indicating a right path to a leaf node for the target document. The update unit performs second machine learning that applies, to the classification model, the target document to which the correct path is not given, and when a path from an Nth level node to an (N+1)th level node is different from the correct path, updates the classification model by setting a modified path from the (N+1)th level node to an (N+2)th level node different from a child node of the (N+1)th level node based on the correct path.

Abstract: A method of automated multi-source website hybridization using streaming data, and performed by first server, includes extracting, from a communication received from a remote device, an identifier of the remote device and at least a portion of a website including a first display generation element, a second display generation element and at least a reference link identifying the first server, each configuring a remote device to provide to a user a user-navigable display combining a visual element and a user-selectable navigation link, generating a first location descriptor, retrieving at least a data stream using the first location descriptor, and remotely generating a hybrid website on the remote device by creating at least a modified second display generation element configuring the remote device to display a second user-navigable display combining the a first user-navigable display with a data stream received from the first server, combined with the first display generation element.

Abstract: A user can create a customizable visual code that is associated with a network address of a page. The customizable visual data can be integrated with one or more user selected images and data marks. When the customizable visual code is scanned, one or more pages may be displayed on the scanning device.

Abstract: A method and apparatus for stimulating neural activity in the brain of a user of an apparatus with a display screen by causing at least one portion of the display screen to flicker in a controlled manner and utilizing the apparatus to measure an effect on a user exposed to the flicker for a time.