Abstract: The embodiments of present disclosure address unresolved problems of label inconsistency, where outputs of different levels create impossible combinations, and error propagation from previous level outputs can significantly impact its performance. Embodiments provide a method and system for a Progressive Multi-level Training framework with a Logit-masking strategy (PMTL) for a retail taxonomy classification. PMTL enables neural network models to be trained separately for each level to reduce error propagation problems. To further enhance the model's performance at each level and get the label-wise constraint from the previous level, the global representation from model of previous level is augmented. Further, a logit masking strategy is used to restrict model(s) to learning only relevant classes through part of final classification layer, thereby addressing label inconsistency issue, and incorporating benefit of parent node-based local classifier.

Abstract: A computer-implemented method for performing a data processing and feature recognition processes. The method includes processing user transaction data associated with user transactional behaviors in a multi-layered data storage system, providing a user transactional behavior dataset at a processor, and collecting a plurality of user transactional behavior data associated with the user transactional behavior dataset.

Abstract: A method of treating bacterial vaginosis in a subject is disclosed. The method comprises administering to the subject a therapeutically effective amount of a composition which comprises between three and twenty species of bacteria or a secretion thereof, wherein at least 70% of the bacteria of the composition are of the species Lactobacillus crispatus, thereby treating the bacterial vaginosis.

Abstract: Complexity of voice based control of tele-robots increases when used in collaborative multi-presence systems, as tele-robot fail to differentiate between operator and observers, and hence may end up executing command from wrong person. Further, the tele-robot ends up losing energy as it listens continuously to speech from multiple people, which may or may not contain commands. Embodiments disclosed herein provide a telepresence system in which an Automatic Speech Recognition (ASR) component hosted on a cloud receives and processes speech signals, infers commands, and only the inferred commands are transmitted to the tele-robot, thereby causing energy saving at the tele-robot end as it doesn't have to listen to all the speech from all the operator and observers for the execution of the voice based navigation control command. Further, a dedicated channel is established between the operator at any instance and the tele-robot, enabling direct communication between the operator and the tele-robot.

Abstract: Systems and methods are provided for determining fabrics of a geological sample using three-dimensional segmentation. An example method can include receiving three-dimensional (3D) image of a geological sample, adjusting an initial size of the 3D image of the geological sample, and partitioning the resized 3D image of the geological sample into cubes. The example method can include, for each cube, generating orthogonal planes based on a center of mass of each cube and extracting, for the orthogonal planes associated with each cube, one or more features to represent texture of the geological sample. The example method can further include grouping the cubes into one or more clusters based on the one or more features and constructing a volume of the resized 3D image of the geological sample based on the one or more clusters for a texture analysis of the geological sample.

Abstract: Some embodiments can include methods and related systems to project coarse representations of natural language interactions into fine-grained representations using contrastive learning projections. Some embodiments can maximize a first set of distances between dissimilar points and anchor points in the fine-grained representations, and minimize a second set of distances between similar points and the anchor points in the fine-grained representations. Some embodiments can then propagate labels from labeled projections to unlabeled projections based on a similarity metric.

Abstract: Apparatus and method for controlling flow back through a conduit. The apparatus includes a selective sleeve disposed within the conduit, a bottom sleeve coupled to the selective sleeve, the conduit comprising a check valve, and an inflow control device fluidically coupled to the conduit. A shearable element locks the bottom sleeve to the conduit. The check valve is covered by the selective sleeve when the apparatus is introduced into a wellbore. The selective sleeve shifts to close the check valve and open the inflow control device. The selective sleeve also shifts to close the check valve and the inflow control device.

Abstract: A downhole sleeve includes a cylindrical housing having a hollow interior, the cylindrical housing configured to slide around a component. The downhole sleeve also includes a plurality of protrusions positioned around the cylindrical housing, each protrusion configured to: receive a radially-inward force applied to an exterior surface of the respective protrusion; and transfer the radially-inward force through an interior surface of the respective protrusion to hold the component in position.

Abstract: A computer-implemented method to assess risk of gas breakout in a well includes determining a rate of migration of a gas injected into a well containing a section of cement. The method also includes predicting, based on the rate of migration, an amount of time for a breakout to occur, wherein the breakout occurs when the gas penetrates the section of the cement to a location of interest. The method further includes determining an effective diffusivity of the gas when the gas achieves breakout. The method further includes assessing a risk of a gas breakout based on the rate of migration and the effective diffusivity of the gas.

Abstract: A horsehead is used on a walking beam of a reciprocating pump unit. The horsehead includes side plates, a face, tabs, and brace plates. The side plates are spaced apart, and the face affixed to front, arced edges of the side plates. The face defines a split therein extending from a bottom end toward a top end, and the tabs are affixed to the face across the split. The face thereby defines two runner surfaces. A fixture at the top end of the face can be used for connection to a bridle of the pump unit. The brace plates are connected in the space between the side plates and are arranged in a lattice structure. Preferably, the brace plates include cutouts to reduce weight.

Abstract: A method for making a smokeless tobacco article can include combining thermoplastic polymer particles with tobacco particles to form a mixture of particles, compressing the mixture of particles within a pressing apparatus to form the mixture into a predetermined shape, separating the mixture from the pressing apparatus such that it retains the predetermined shape, and heating the mixture outside of the pressing apparatus to at least partially melt at least some of the thermoplastic polymer particles and form a matrix of polymer and tobacco particles.

Abstract: Methods, systems, and apparatuses are described herein for improving computer authentication processes through computer-based authentication in a manner that uses knowledge of former devices. A computing device may train a machine learning model to output an indication of device reliability data associated with a particular device. The computing device may receive a request for access to an account from a user. The computing device may receive account data and provide the account data to the trained machine learning model. The computing device may receive data indicating device reliability for a set of devices from the machine learning model. The computing device may generate a modified set of device choices for the user by excluding devices having reliability levels below a threshold value. An authentication question may be generated, and access to the account may be provided based on a response to the authentication question.

Abstract: A system may include a router that may receive a plurality of data packets from one or more devices that communicatively couples to the router. The system may also include at least one processor that identifies an identity of a device based on a data packet received by the router from the device, generate an insurance policy that includes the device in response to identifying the identity of the device, and sends a notification indicative of the insurance policy to a computing device in response to generating the insurance policy.

Abstract: A system and method for using an augmented intelligence guide system to assist users, customers, members, and other users of a service (for example, a financial service or an insurance service) in completing a task or process to improve the success rate of the accurate completion of the steps necessary to execute the task or process is described. The system includes a process detection module, an experience determination module, an assistance classification module, and a guidance engine.

Abstract: Provided is a downhole tool, a method for sealing within a well system, and a well system. The downhole tool, in at least one aspect, includes a tubular, and a collection of individual separate chunks of expandable metal positioned about the tubular, the collection of individual separate chunks of expandable metal comprising a metal configured to expand in response to hydrolysis.

Abstract: A method for minimizing over pressurization during a pressure test of a fluid distribution system comprises a plurality of pumping units in fluid communication with a wellhead via a manifold. The method comprises pressurizing the fluid distribution system with a fluid using an initial number of pumping units to an initial predetermined pressure, and further pressurizing the fluid distribution system with the fluid using a final number of pumping units to a final predetermined pressure. The final number of pumping units is less than the initial number of pumping units.

Abstract: A thermocouple cable is formed from a tubing and a plurality of thermocouple conductors bundled within the tubing, wherein each thermocouple conductor forms a junction with a shared thermocouple conductor to form a thermocouple junction, and each thermocouple junction is attached to a support cable in a thermocouple bundle. The cable is formed by pulling the thermocouple bundle into the tubing.

Abstract: Multidimensional MRI-based methods permit identification and categorization of brain specimens to identify sub-voxel tissue components that are specific to traumatic axon injury or other lesions. Lower dimensional MR spectral data is acquired and processed to provide multidimensional MR data of higher dimensions. One or more spectral ranges are selected that define signatures for brain injury and evaluation of the multidimensional MR data in these ranges is used to locate voxels associated with brain injury. For example, partial one dimensional data sets such as T1, T2, and mean diffusion coefficient (MD) data sets can be combined to provide two dimensional data sets such as T1-T2, MD-T2, and MD-T1 data sets. Using the spectral signatures, a specimen image can be produced showing areas of lesser or greater injury.

Abstract: Disclosed are systems and methods for developing a web-based application using micro frontends. A web application architecture system may include a plurality of web servers and a plurality of single page applications. Each one of the plurality of single page applications may be hosted on a single respective one of the plurality of web servers. The web application architecture system may further include one or more shared libraries that are accessed by each of the plurality of single page applications. The one or more shared libraries may include shared interface components for each of the plurality of single page applications such that the plurality of single page applications are a part of a single web-based application. Additionally, the web application architecture system may include a load balancer configured to route traffic to each of the plurality of single page applications.

Abstract: A method and a system of classifying textual data blocks are claimed. The method includes receiving at least one textual data block in an original version, including a plurality of textual data elements; performing a preprocessing procedure on the at least one textual data block in the original version, wherein the preprocessing procedure includes replacing the textual data elements characterized by pertinence to at least one specific part-of-speech (POS) category with a respective POS token, thereby obtaining the at least one textual data block in a preprocessed version; inferring a pretrained ML-based model on the at least one textual data block in the preprocessed version, to classify the at least one textual data block by pertinence to the at least one class.