Abstract: The embodiments provide a system and method for interpreting predictions from machine learning systems using natural language. The system includes a prediction module and an explanation module. The prediction module includes a machine learning model to make predictions for quantities such as recommended coverage amounts for insurance policies. The explanation module includes an impact analyzer that calculates impact values, which represent the degree of influence that each attribute has on predicted values. The explanation module also includes a natural language processing system for transforming generating natural language explanations indicating how the user attributes have influenced the predicted value.

Abstract: An authentication and payment system for paying a supplier via a network may include a non-transitory computer readable storage medium having program logic stored thereon having a code for an authentication and payment application executable by a processor of a computing device. The code may be programmed to perform a step of receiving a first set of specific data elements from a payer regarding an authorized purchase by the payer from the supplier. The code may be further programed to receive a second set of specific data elements from the supplier. If the first set of specific data elements matches at least a portion of second set of specific data elements, the code may be programed to generate and send a single-use number to the supplier. The single-use number may be usable as a credit card number for only one purchase.

Abstract: Embodiments of the invention are directed to techniques for detecting anomalous values in data streams using forecasting models. In some embodiments, a computer can receive a value of a data stream comprising a plurality of data values, where the received value corresponds to a time interval and previously received values each correspond to a previous time interval. Models can be selected based on the time interval, where each of the models has a different periodicity. For each of the selected models, the computer may generate a score by generating a prediction value based on the model and generating the score based on the prediction value and the received value. A final score can then be generated based on the scores. Next, a score threshold can be generated. If the final score exceeds the score threshold, the computer may generate a notification that indicates that the data value is an anomaly.

Abstract: Conventional fork-type autonomous mobile robots (AMRs) have been suited to handle pallets and are typically designed with two forks. Such AMRs are very bulky in nature and specifically designed for a cart handling application, and usually have large openings and less suitable for lifting roller carts. Present disclosure provides a fork assembly for AMRs/Autonomous Guided Vehicles (AGVs) for transporting roller cages/carts within warehouses. The fork assembly when integrated with AMR enables performing various tasks. More specifically, the fork assembly includes a first plate and a second plate. The fork assembly further include roller movement enabler blocks that are driven by respective fork motors. Movement of the roller movement enabler blocks enable rolling of rollers on respective roller guides within tapered region thereby enable lowering and rising of the second plate with reference to the first plate for lifting a payload and movement thereof to a desired location.

Abstract: An aerosol-generating system may include an electrical power supply, an electrical heater connected to the electrical power supply, and a substrate container. The substrate container may define a blind cavity containing an aerosol-forming substrate in the form of a gel that is a solid at room temperature. The electrical heater is external to the substrate container and configured to heat the substrate container to generate a vapor without contacting the aerosol-forming substrate. As result, the possibility of unwanted materials building up on the electrical heater can be reduced, thereby providing a more reliable and consistent performance.

Abstract: Systems, methods, and computer-readable media for providing accurate images of cards. An application executing on a mobile device may receive authentication credentials for an account, determine a card associated with the account, and determine a card manufacturer identifier (CMID) of the card. The application may receive, from a digital asset management system based on the CMID, a plurality of metadata attributes of the card and one or more images of the card. The application may determine, based on the metadata attributes and attributes of the mobile device, a first orientation of the card to enable near-field communication (NFC) data transfer between the card and the mobile device. The application may select a first image of the card. The application may generate a graphical user interface (GUI) including the first image and an indication specifying to tap the card to the mobile device according to the first orientation.

Abstract: An authentication and payment system for paying a supplier via a network may include a non-transitory computer readable storage medium having program logic stored thereon having a code for an authentication and payment application executable by a processor of a computing device. The code may be programmed to perform a step of receiving a first set of specific data elements from a payer regarding an authorized purchase by the payer from the supplier. The code may be further programmed to receive a second set of specific data elements from the supplier. If the first set of specific data elements matches at least a portion of second set of specific data elements, the code may be programmed to generate and send a single-use number to the supplier. The single-use number may be usable as a credit card number for only one purchase.

Abstract: This application relates to a method and a system for compressing a captured image of an item such as a mailpiece or parcel. The system may include a memory configured to store images of a plurality of items captured while the items are being transported and a processor in data communication with the memory. The processor may be configured to receive or retrieve one or more of the captured images, perform a wavelet scattering transform on the one or more captured images, perform deep learning on the wavelet scattering transformed images to classify the wavelet scattering transformed images and compress the classified wavelet scattering transformed images. Various embodiments can significantly improve a compression efficiency, a communication efficiency of compressed data and save a memory space so that the functionality of computing devices is significantly improved.

Abstract: Embodiments of a dry mix for producing a concrete composition are provided. The dry mix includes aggregate, cement, and bed ash. The bed ash contains the combustion product of a fluidized bed coal combustion reaction. Additionally, embodiments of a method of preparing the dry mix and embodiments of a method of preparing a concrete composition are provided. The dry mix is also suitable for repairing soil slips, and embodiments of a method of repairing a soil slip are also provided.

Abstract: Computer implemented systems and methods are provided for securing data. In some embodiments, a system for securing data may comprise one or more processors configured to receive a request for data over a network. The one or more processors may be configured to identify one or more confidential portions of data within the requested data. The one or more processors may be further configured to transmit the confidential portions of data to a hardware device configured to secure the confidential portions of data, and receive the secured data from the hardware device.

Abstract: Embodiments are directed to a first token vault managed by a token service provider and a second token vault managed by a second entity. The first token vault is continuously synchronized with the second token vault in real-time. That is, the data in the second token vault is updated simultaneously with or immediately after the data in the first token vault is updated. In some embodiments, the first token vault may be a master token vault and the second token vault may be a mirror token vault. The mirror token vault may store the same information as the master token vault. Alternatively, the mirror token vault may store a subset of the information stored at the master token vault. In other embodiments, the information stored at the first token vault and the information stored at the second token vault may be mutually exclusive.

Abstract: Techniques for enhancing the security of a communication device when conducting a transaction using the communication device may include using a limited-use key (LUK) to generate a transaction cryptogram, and using a signature key to generate a signature. The transaction can be an offline data authentication transaction, and access can be granted based on authentication of the signature prior to verifying the transaction cryptogram.

Abstract: A device may include a processor, a wireless transceiver in communication with the processor, and a non-transitory memory. The memory may store instructions that, when executed by the processor, cause the processor to perform processing. The processing may include sending, by the wireless transceiver, a request to share a transaction to at least one external user device in communication range of the device. The processing may include receiving, at the wireless transceiver, at least one sharing confirmation from the at least one external user device. The processing may include generating a transaction request. The transaction request may include data describing the transaction and a transaction portion to be paid by an account associated with the device. The processing may include sending the transaction request to a transaction service for fulfillment.

Abstract: In some implementations, a transaction card includes a first card body portion, a second card body portion, at least one of an integrated circuit chip or a magnetic stripe accessible from at least one of the first card body portion or the second card body portion, and a third card body portion that includes a series of accordion pleats. The series of accordion pleats may be between the first card body portion and the second card body portion. The series of accordion pleats may be configured to contract the third card body portion into a contracted configuration, and to expand the third card body portion into an expanded configuration. The first card body portion may be nearer to the second card body portion with the third card body portion in the contracted configuration than with the third card body portion in the expanded configuration.

Abstract: A label stack assembly and method and system for preparing the label stack assembly is disclosed herein. The label stack assembly may be applied to a product in order to provide care instructions in multiple different languages. Individual pages of the label stack assembly may have separate content, such as translations of care instructions into multiple languages, and may be marked by tabs of various shapes, sizes, or styles that may protrude from an edge of the label stack assembly. Tabs may be marked with country codes indicating the translations or other information. The pages of the label stack assembly may be printed on both a front and a back side, such as may be desired, in order to minimize the number of pages used while maximizing the number of care instruction translations or other content pages.

Abstract: A method of providing sentiment analysis includes aggregating, by a processor, a plurality of text-based comments, classifying, by the processor, the plurality of text-based comments as being associated with a polarity of sentiment, and generating, by the processor, a plurality of phrases from the plurality of text-based comments. The method also includes identifying, by the processor, a predetermined number of most common phrases for a particular polarity of sentiment from the plurality of phrases and outputting, by the processor, a graphic that includes the predetermined number of most common phrases for the particular polarity of sentiment.

Abstract: Methods and systems are described herein for determining resource usage metrics of cloud computing systems. The system may generate access tokens for an account associated with a certain resource class of resources. The system may receive requests comprising a corresponding access token request and a corresponding resource class identifier. The system may determine the corresponding resource class for each request. The system may generate aggregated resource data based on a corresponding quantity of each resource within the resource class and may determine, based on the aggregated resource data, a resource usage metric indicating a total amount of resources used by the cloud computing system.

Abstract: A system may include an interface configured to couple to a network, and includes a processor and a memory accessible to the processor. The memory may be configured to store instructions that, when executed, cause the processor to process search results corresponding to multiple data owners to selectively filter personally identifiable information (PII) associated with one or more consumers from the set of search results according to data sharing permissions for each of the data owners to produce filtered results. The instructions may further cause the processor to provide the filtered results to a user device through the network.

Abstract: There is a need for more effective and efficient predictive data analysis. This need can be addressed by, for example, solutions for performing cross-temporal predictive data analysis. In one example, a method includes determining a time-adjusted encoding for each temporal unit of a group of temporal units, processing each time-adjusted encoding using a cross-temporal encoding machine learning model to generate a cross-temporal encoding of the group of temporal units, and performing one or more prediction-based actions based at least in part on the cross-temporal encoding.