Abstract: A method for providing document category recommendations may include training a machine learning algorithm; receiving, from a user, a selection of an anchor document; retrieving a user co-viewing sequence; generating, via the trained machine learning algorithm, a sequence embeddings set based on the anchor document and the user co-viewing sequence; comparing the generated embeddings set to respective embeddings for a plurality of candidate sets; determining the candidate set closest to the generated embeddings; and presenting at least one category from the closest candidate set.

Abstract: Elastic privileges in a secure access service edge, including: identifying, based on one or more access policies, an application accessible to a user; determining, for the user, an access pattern of the application; and restricting, without modifying the one or more access policies, access to the application by the user based on the access pattern.

Abstract: Certain aspects involve a machine-learning query system that uses a dual deep learning network to service queries and other requests. In one example, a machine-learning query system services a query received from a client computing system. A dual deep learning network included in the machine-learning query system matches an unstructured input data object, received from the client computing system, to an unstructured reference data object. The matching may include generating an input feature vector by an embedding subnetwork, based on the unstructured input data object. The matching may also include generating an output probability by a relationship subnetwork, based on the input feature vector and a relationship feature vector that is based on the unstructured reference data object. The machine-learning query system may transmit a responsive message to the client system.

Abstract: Detecting anomalous behavior using a browser extension, including: gathering first information describing activity associated with a user and generated by a browser extension on a user device; gathering second information describing activity associated with the user and generated by an application executed on the user device; and determining, based on the first information and the second information, whether the user has deviated from normal activity.

Abstract: A system of performing boundary scan test on pin through test point and a method thereof are disclosed. When an under-test pin of a target connector is determined to be unable to perform a boundary scan test, a test point connected to and closest to the under-test pin is searched, a test signal is transmitted to a target connector, a result signal from the target connector in response to the test signal is received, an expected result and the result signal are compared to generate a test result, so that a boundary scan function can be applied to test a connector of a computer product, to achieve the technical effect of providing a better test range and a better test coverage to improve test efficiency and reduce test cost, compared to conventional boundary scan test.

Abstract: Generating user-specific polygraphs for network activity, including: gathering information describing network activity associated with a user and generating, based on the information, a user-specific polygraph that includes one or more destinations associated with the network activity.

Abstract: Embodiments of the present disclosure provide a datum selection method for minimizing hole position errors in group hole machining of large components, comprising: 1) determining a type of a computer numerical control (CNC) machine tool and establishing a topological structure of the CNC machine tool; 2) establishing a theoretical postural model of a tool center point during a motion; 3) establishing a hole position error model; 4) establishing an average error model of hole positions in group hole machining; and 5) obtaining a machining datum for group holes of corresponding components. For the skeleton and skinned group hole machining of aircraft components, different principles of datum selection are provided respectively, which can effectively improve the positional accuracy of the skeleton or skinned group hole machining, at the same time provides a more scientific and reasonable approach for the datum selection in group hole machining of large components.

Abstract: Detecting deviations from typical user behavior, including: identifying a geographic location of a device that is associated with a user; determining device activity associated with the user; and detecting, based on a profile associated with the user, that the device activity associated with the user deviates from normal activity for the user.

Abstract: Detecting anomalous behavior of a device, including: generating, using information describing historical activity associated with a user device, a trained model for detecting normal activity for the user device; gathering information describing current activity associated with the user device; and determining, by using the information describing current activity associated with the user device as input to the trained model, whether the user device has deviated from normal activity.

Abstract: Certain aspects involve a machine-learning query system that uses a dual deep learning network to service queries and other requests. In one example, a machine-learning query system services a query received from a client computing system. A dual deep learning network included in the machine-learning query system matches an unstructured input data object, received from the client computing system, to an unstructured reference data object. The matching may include generating an input feature vector by an embedding subnetwork, based on the unstructured input data object. The matching may also include generating an output probability by a relationship subnetwork, based on the input feature vector and a relationship feature vector that is based on the unstructured reference data object. The machine-learning query system may transmit a responsive message to the client system.

Abstract: Embodiments of the present disclosure provide a datum selection method for minimizing hole position errors in group hole machining of large components, comprising: 1) determining a type of a computer numerical control (CNC) machine tool and establishing a topological structure of the CNC machine tool; 2) establishing a theoretical postural model of a tool center point during a motion; 3) establishing a hole position error model; 4) establishing an average error model of hole positions in group hole machining; and 5) obtaining a machining datum for group holes of corresponding components. For the skeleton and skinned group hole machining of aircraft components, different principles of datum selection are provided respectively, which can effectively improve the positional accuracy of the skeleton or skinned group hole machining, at the same time provides a more scientific and reasonable approach for the datum selection in group hole machining of large components.

Abstract: Detecting anomalous behavior using a browser extension, including: gathering first information describing activity associated with a user and generated by a browser extension on a user device; gathering second information describing activity associated with the user and generated by an application executed on the user device; and determining, based on the first information and the second information, whether the user has deviated from normal activity.

Abstract: Methods, systems, and products for offline workflows in an edge-based data platform, including: accessing log data describing activity associated with a user; generating, based on the log data, one or more alerts; and initiating, based on the one or more alerts, a workflow to acknowledge the one or more alerts by the user.

Abstract: A method for training a machine learning model includes receiving a randomly-initialized first version of a machine learning model, conducting first training on the machine learning model first version using first training data, the first training data comprising a first type of information respective of a plurality of documents, adding a layer to the machine learning model first version after conducting the first training to create a machine learning model second version, and conducting second training on the machine learning model second version using second training data, the second training data comprising a second type of information respective of the plurality of documents.

Abstract: The invention provides devices and methods for non-invasive monitoring and measuring of intraocular pressure (IOP) of a subject. Embodiments include a lens that is adapted to fit on the subject's eye, a microstructure disposed in or on the lens, the microstructure having at least one feature that exhibits a change in shape and/or geometry and/or position on the lens in response to a change in curvature of the lens. When the curvature of the lens changes in response to a change in IOP, a corresponding change in shape and/or geometry and/or position of the feature may be used to determine the change in IOP. The change in the feature is detectable in digital images of the lens taken with a mobile electronic device such as a smartphone.

Abstract: A method for predicting a next user selection in an electronic user interface, such as a website, includes training a machine learning model according to a training data set, the training data set including a plurality of token sets, each token set representative of a respective document accessible through the interface, each token set including a plurality of words, each word describing a characteristic of the document, to create a trained model. The method further includes receiving, from a user, a sequence of selections of documents, inputting the sequence of selections to the trained model, and outputting to the user, in response to the sequence of selections, a predicted next document selection according to an output of the trained model.

Abstract: An antenna structure and a wireless communication device having the antenna structure are provided, the antenna structure includes a metal frame, a feeding portion, a first ground portion, and a second ground portion. The metal frame defines a first gap, a second gap, and a third gap, the metal frame between the first gap and the second gap forms a first radiating portion, the metal frame between the first gap and the third gap and the metal frame on a side of the third gap cooperatively form a second radiating portion, the metal frame on a side of the second gap forms a third radiating portion. The feeding portion is connected to the first radiating portion. The first ground portion is apart from the feeding portion and connected to the first radiating portion. The second ground portion closes to the second gap and is connected to the third radiating portion.

Abstract: Generating user-specific polygraphs for network activity, including: gathering information describing network activity associated with a user and generating, based on the information, a user-specific polygraph that includes one or more destinations associated with the network activity.

Abstract: Detecting anomalous behavior of a device, including: generating, using information describing historical activity associated with a user device, a trained model for detecting normal activity for the user device; gathering information describing current activity associated with the user device; and determining, by using the information describing current activity associated with the user device as input to the trained model, whether the user device has deviated from normal activity.

Abstract: This application provides a path switching method and a related apparatus. In the path switching process, an ECMP table includes a first ECMP index of a first path group and a second ECMP index of a second path group; when all paths in the first path group have failed, a path group identifier corresponding to the first ECMP index may be changed from valid to invalid in the ECMP table, indicating that all paths in the first path group are unavailable. In this case, a path group identifier corresponding to the second ECMP index is valid, indicating that the second path group includes an available path, thereby completing switching between an active path and a standby path. In this process, only an identifier corresponding to the ECMP index needs to be updated, and therefore, the time consumed is relatively short, and the path switching delay can be effectively reduced.