Abstract: An assembly is used to connect an equalizer bar to a walking beam of a reciprocating pump unit. The walking beam has a bearing shaft having cylindrical ends. Brackets attached to the equalizer bar define slots, each having a hooked section and having a base section. The hooked sections are configured to rest against the cylindrical ends of the bearing shaft. The base sections of the slots oppose the hooked sections at a gap from the cylindrical ends of the bearing shaft. Wedges are configured to position in the slots between the bases and the cylindrical ends of the bearing shaft.

Abstract: A system and method for detecting a recurring charge that is unknown to the customer and/or the result of fraud or deceit is disclosed. A customer complaint initiates an analysis of a particular transaction. In response to the complaint, information relating to the transaction is extracted and used to identify additional information from a variety of different sources. Such information may include information relating to the customer, information relating to the merchant, as well as information relating to past transactions or related transactions. This information is provided to a machine learning algorithm, both for training purposes and for analysis purposes. The machine learning algorithm analyzes a transaction in order to determine the likelihood that the transaction is a recurring transaction, as well as the likelihood that it is a result of a bad actor, fraud or deceit.

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: Provided are systems that include a processor to receive a training dataset including a plurality of data records, calculate feature projection errors and classification scores for the plurality of data records using a machine learning model, determine a distribution of features according to feature projection error, apply a downscaling function to each feature value of a feature in a false positive classification distribution to provide a downscaled set of feature values, apply an upscaling function to each feature value of a feature in a false negative classification distribution to provide an upscaled set of feature values, combine the downscaled set of feature values and the upscaled set of feature values with the training dataset to provide an updated training dataset, and train the machine learning model using the updated training dataset to provide an updated trained machine learning model. Methods and computer program products are provided.

Abstract: The disclosure relates generally to assortment of products in a retail scenario and more particularly to variety optimization in hyper-localized assortment. In retail scenario, due to availability of limited space in the store, retailers have limited options for displaying a category based on the attribute values of important attributes of the category. In addition, due to increased customer expectations, the attribute values are increasing each day leading to challenges in variety optimization in assortment. The state-of-art techniques address the problem of identification of key attributes for a category, however identifying optimal attribute values among the numerous attribute values specific to location of stores is still challenging.

Abstract: Systems and methods for propagating labels through a sparsely labeled dataset using a supervised projection of a semantic embedding. The system may receive the dataset and a labeling task. It may determine a semantic graph by embedding the dataset using a language model. It may process the semantic graph in an artificial intelligence model, as well as determine: an optimal set of dimensions for the labeling task based on the semantic graph; a projected graph based on the optimal set of dimensions; a first distance between a first projection and a second projection in the projected graph; a first likelihood the first projection has a first label; and a second likelihood the first projection has a second label based in part on the first distance to the second projection. It may compare the first likelihood to the second likelihood and generate a recommendation to stop the labeling task.

Abstract: Systems and methods for improving training efficiency while reducing total labeling cost for downstream modeling using entropy scores for labeling predictions. The system may receive a first text string, process the first text string in an artificial intelligence model, and determine: a first likelihood that the first text string has a first label component; a second likelihood that the first text string has a second label component; a composite likelihood for the first text string based on the first likelihood and the second likelihood; and a first entropy score from the composite likelihood. It may compare the first entropy score to a first threshold entropy score and generate a recommendation to review a first label being assigned to the first text string.

Abstract: Disclosed are methods, systems, and non-transitory computer-readable medium for targeting bid and position for a keyword. For instance, the method may include obtaining information about the keyword, the information about the keyword including observations of value with respect to position for the keyword. The method may further include applying a Gaussian Process Model on the observations to obtain a prediction function and associated uncertainties, the prediction function and the associated uncertainties relating positions to expected values; applying a Thompson sampling reinforcement learning model on the expected values and the positions to obtain a target position; and applying a bid model to the target position to obtain bid information for the keyword. The method may also include transmitting a bid message to a search engine, the bid message including the bid information.

Abstract: Systems and methods for improving label quality in datasets using a quality filter based on a consistency score. The system may receive a first given label for a first text string. It may process the first text string in an artificial intelligence model, as well as determine: a first predicted label for the first text string; and a first consistency score for the first text string based on a comparison of the first predicted label and the first given label. The system may compare the first consistency score to a first threshold consistency score and, in response to comparing the first consistency score to the first threshold consistency score, filter the first text string to a first group. The system may generate a recommendation to use the first group for a supervised learning task.

Abstract: Methods and systems disclosed herein may detect a malicious user by scanning the messaging platform with the members for any malicious content using sentiment detection. Sentiment detection may be performed using a machine learning model or another type of sentiment detection. For example, the machine learning model may detect a malicious comment and then determine whether the user is malicious using a variety of behavior patterns. For example, the machine learning model may use factors like reactions from the community, sentiment analysis, and the number of posts to determine whether the user is malicious. By doing so, the system is able to search constantly for users that may be malicious.

Abstract: A method, a system, and a computer program product for executing transactions using contactless cards. A near-field communication (NFC) exchange is executed between a first device and a second device upon the first device being detected by the second device to be located within a predetermined distance of the second device. The first device is associated with an electronic deposit account capable of receiving an electronic transfer of funds. At least one identification data associated with the first device is received from the first device based on the NFC exchange and sent to a server. A response from the server is processed and includes an authentication data corresponding to the first device and associated with the sent identification data. A request for electronically transferring predetermined electronic funds to the electronic deposit account is received.

Abstract: Provided are approaches for providing card information in a highly secure and convenient manner. In some approaches, a transaction card may include a body having a first main side opposite a second main side, wherein a first outer layer extends along the first main side, and an internal layer adjacent the first outer layer, wherein one or more visual features are provided on the internal layer. The transaction card may further include a plurality of beads within a carrier, wherein the plurality of beads and the carrier are positioned between the internal layer and the first outer layer, wherein a force applied to the first main side of the body displaces one or more beads of the plurality of beads.

Abstract: The present disclosure relates to systems and methods for backing up a distributed database provided as a database instance across servers within a first geographic region. In one example, such a system may include at least one processor configured to: generate and transmit a command to duplicate the distributed database in object databases on first region servers; generate and transmit a command to duplicate the object databases on servers in at least a second geographic region independent from servers in the first region; and, when an error is detected with at least one server within the first region: generate and transmit a command to initiate a new distributed database on second region servers, generate and transmit a command to populate the new distributed database using the object databases on second region servers, and generate and transmit a command to re-route traffic from the distributed database to the new distributed database.

Abstract: The method includes establishing a retention medium, connecting a plurality of elements together to form an aerosol-generating article, the plurality of elements including a retention medium directly abutting a heat diffuser, the heat diffuser including a porous body on a distal end of the aerosol-generating article, the porous body being configured to absorb heat such that the porous body heats air drawn through the porous body and the retention medium during an operational use of the aerosol-generating article, providing at least one frangible capsule that releasably contains a substrate, and inserting the at least one frangible capsule within the retention medium, the retention medium being configured to absorb and retain the substrate after the at least one frangible capsule is ruptured by an application of an external force, the retention medium being devoid of the substrate prior to a release of the substrate from the at least one frangible capsule.

Abstract: Aspects described herein may relate to a transaction exchange platform using a streaming data platform (SDP) and microservices to process transactions in accordance with corresponding workflows. The transaction exchange platform may receive transactions from origination sources, which may be added to the SDP as transaction objects. Microservices on the transaction exchange platform may interact with the transaction objects based on configured workflows associated with the transactions. Further, the microservices may leverage machine-learning models to determine whether transaction objects may be more effectively processed using alternative or secondary workflows. Processing on the transaction exchange platform may facilitate clearing and settlement of transactions. Some aspects may provide for dynamic and flexible reconfiguration of workflows.

Abstract: Methods and kits for assembling a flow cage assembly are provided. The flow cage assembly may be used in a traveling valve, a standing valve, or a top plunger adapter of a downhole reciprocating pump. In some embodiments, the method comprises assembling a body, a tubular insert, and a retaining element such that the retaining element forms an interference fit with the body and thereby securely retains the tubular insert within the body. Top plunger adapters are also provided, assembled from embodiments of the methods and kits.

Abstract: System and methods are described for enabling a first client device to authorize a second client device to use a payment token for contactless payments, in which the payment token is authorized to be used by the second device with restrictions set by the first device. In one embodiment, a system comprises a computing device that includes a secure element. The system also includes machine-readable instructions that, when executed by the processor, cause the computing device to at least receive an indication for generating a sub-payment token and receive a selection of a key restriction for the sub-payment token. An application stored in the secure element is called to generate the sub-payment token based on the key restriction. The sub-payment token is received and comprises a sub-payment key and an application transaction counter range. The sub-payment token is sent through the wireless communication session to a second computing device.

Abstract: At least one processor is configured for defining a plurality of mutually exclusive customer treatment groups, including in accordance with first and second algorithms, to receive content items. Content items are respectively provided to a random customer treatment group as well as first and second algorithm customer treatment groups, and metrics representing at least engagement by each of the customers are determined and analyzed. A selection of the first or the second algorithm is made. The at least one processor is configured to provide, to at least some of the plurality of customers, content items in accordance with the selected algorithm.

Abstract: One or more identifiers are received based on an electronic scan of computer-readable indicia. The computer-readable indicia being coupled to an asset. The asset is associated with one or more shipping operations. Based at least in part on the one or more identifiers, it is determined that the asset is requested for return. Based at least in part on the one or more identifiers, a shipping label corresponding to a destination location for the asset to be returned to is automatically generated or a location for the asset to be delivered to is automatically changed.