Abstract: Disclosed are various embodiments for using decision trees for machine-learning when data is missing from a data set. A first attribute for splitting a plurality of records is identified. Then, the plurality of records are split into a first subset of records and a second subset of records. The first subset of records can include each of the plurality of records where a value is present for the first attribute and the second subset of records comprising each of the plurality of records where the value is absent for the first attribute. Finally, a node can be added to a decision tree that reflects the split of the plurality of records into the first subset of records and the second subset of records.

Abstract: A computing device (e.g., a server, a cloud-based device, a network device, an item analysis device, etc.) may receive location information indicating a geographic location of a user device and image data captured by the user device of a listing (e.g., a menu, an event ticket, a content source, etc.) for a selectable item. The computing device may determine an entity offering the selectable item based on the location information. An intent indicator for the selectable item may be determined based on the entity and the information for the selectable item. An instruction to the user device may cause modification of the image data to indicate the intent indicator based on the intent indicator satisfying an intent threshold that defines a level at which an intent is classified as a positive intent.

Abstract: Disclosed are various embodiments facilitating a holistic engagement with a user across multiple communication channels of an organization or an enterprise based at least in part on a determined user intent. As users interact with various services associated with the organization through one or more communication channels, interaction data can be captured and stored. In various examples, the interaction data that is stored by the various services can be obtained and organized according to a predefined schema. The organized interaction data can be applied to a trained intent model that outputs a user intent based at least in part on observations of other users with similar histories. The predicted intent can be provided to the different services such that subsequent interactions between the user and the organization can be based at least in part on the intent in a consistent manner, regardless of the communication channel associated with the interaction.

Abstract: The system is configured to monitor and operate a hybrid reward system with a multi-channel point of sale terminal. The system may perform operations including receiving, by a computer based system and via a first connection from a point of sale terminal, reward program information and transaction information in response to a reward program instrument being presented at the point of sale terminal. A transaction authorization request associated with the reward program information may be provided by the point of sale terminal, via a second connection from the point of sale terminal, to a transaction authorization system to authorize a transaction. The transaction may be initiated by a transaction instrument. The reward program instrument is at least one of a transaction instrument and a reward program indicia.

Abstract: Disclosed are various embodiments for using a distributed ledger to track the evolution or deployment of feature sets for machine-learning. A registration request is received from a first node of a distributed ledger, the registration request comprising a code hash representing a feature set generator and a data hash representing a data set. The registration request is then relayed to a second node of the distributed ledger for approval by the second node. Next, an approval for the registration request is received from the second node. Subsequently, an entry in the distributed ledger comprising the code hash and the data hash is created.

Abstract: Disclosed herein are system, method, and computer program product embodiments for individually deploying batch functions in a cluster computing environment. A batch function system may receive a scheduled batch job for execution at a particular timing. Rather than deploying the corresponding batch functions as a bundled monolith, the batch function system may individually deploy corresponding batch functions. To individually deploy the batch functions, the batch function system may use a verticle data structure for defining a batch job. The batch function system may then execute a batch function corresponding to the batch job on a cluster computing platform by deploying the batch function to a pod executing on computing nodes. Upon individual completion of execution, the batch function system may release the pod and/or computing nodes to make the computing resources available. The batch function system may also preserve and pass data between pods executing batch functions.

Abstract: A transaction card that includes a card body that can include a ceramic material. The card body can include a primary surface and a secondary surface, a laser marked feature disposed on the card body and a laser etched feature disposed on the card body. A method of making a transaction card can include forming a ceramic material slurry comprising a ceramic material and a binder, forming a green body from the ceramic material slurry, firing the green body at a firing temperature to create a fired ceramic body, grinding the fired ceramic body into a card body, and polishing a primary surface of the card body.

Abstract: Systems and methods for automated detection of website errors during a sequence of device interactions with a website. In one example, a computing device is configured to receive a website navigation sequence of a series of web page interactions between a client device and a website. The computing device can determine a predicted completion time for a next measurement of the client device executing the website navigation sequence. An actual completion time for the next measurement of the client device executing the website navigation sequence can be determined. Then, the computing device can determine an anomaly website event based on the actual competition time failing to meet a boundary threshold. The anomaly website event is determined to be a website error based on a second machine learning model being trained with a plurality of previous website errors identified from a plurality of previous website navigation sequences.

Abstract: A machine-learning system which utilizes neural networks and other machine learning concepts to automatically categorize a stream of data. Known financial data can be analyzed and trained as part of a machine learning system. Then, a new set of a financial data for a particular user can be analyzed and automatic determinations can be made whether the user is a small business or consumer, and what industry the user is in.

Abstract: Systems and methods are described related to using flexible circuitry in a payment device. In one embodiment, a payment device comprises a computing device and a memory. The computing device is configured to store a biometric identifier associated with a transaction account. The computing device is configured to at least receive, by the antenna, an interrogation signal from a point of sale (POS) device in order to activate the payment device for a payment transaction. The computing device is configured to determine a scanned fingerprint identifier of a user based at least in part on the interrogation signal activating the payment device and authenticate the scanned fingerprint identifier by matching the scanned fingerprint identifier to the biometric identifier. The transaction account is transmitted to the POS device for processing the payment transaction based at least in part on the authentication of the scanned fingerprint identifier.

Abstract: The present disclosure presents systems and related methods for creating real-time predictions. One such method comprises receiving, by a computing device, a first set of data and a second set of data, wherein the first set of data comprises a plurality of items available from a first source for a first set of users and the second set of data comprises transaction purchase data for a second set of users that have reward accounts, utilizing a predictive data model that determines a propensity score for a user from only behavior data that is not attributed to the user; receiving a third set of data from a third source comprising social media channel data for a third set of users; and updating the predictive data model to determine the propensity score for the user based at least in part on the third set of data.

Abstract: Embodiments for intelligently matching a user to an agent for a communication session are disclosed. In one example, a system that comprises a computing device with a processor and a memory. The system also includes machine-readable instructions stored in the memory, that when executed, cause the computing device to at least receive a request from a client device to communicate with an agent. The request is associated with a user account. Interaction data associated with the user account is identified. User scores for the user account are determined based on the interaction data. A user segment for the user account is determined based on the user scores. The system identifies an agent to participate in the communication session based on the agent being associated with the user segment and establishes the communication session between an agent client device of the agent and the client device.

Abstract: At least some embodiments are directed to an entity matching detection system. The entity matching detection system includes a latent similarity identification machine learning model that receives one or more data records and generates a final similarity score indicative of a latent similarity between the one or more data records and a second data record. The entity matching detection system can identify lexical and semantic similarities between attribute values and can analyze and compute similarity scores for direct-linked attribute values and cross-linked attribute values extracted from different data records.

Abstract: A system and method for generating personalized content for a user device. In one example, a system can cause a computing device to generate a merchant score for a plurality of merchants based at least in part on merchant statistics and display a user interface on a user device of a user. The user interface includes a subset of the merchants that have been identified for a user based at least in part on the merchant score. Clickstream data can be received from the user interface displayed on the user device. An updated merchant score can be generated based at least in part on the user preference. A sorting habit for the user is determined based at least in part on a user-specified sort received from the user interface. A modified subset of the merchants is displayed on the user device based on the sorting habit.

Abstract: A BLUETOOTH low energy mobile payments system may comprise a BLE beacon that is deployed at a merchant location. The BLE beacon may be associated with a specific merchant terminal. The BLE beacon may be configured to broadcast a unique beacon ID that is detectable by a user device. The user device may be configured to communicate the beacon ID to a payment system In response to receiving the beacon ID, the user device may initiate a payment between a user and a merchant, via a user device and a merchant terminal.

Abstract: The systems and methods disclosed herein comprise computer-based platforms configured for automated early-stage application security monitoring for allowing users (e.g., application developers) to make decisions at the early stage of the application development.

Abstract: A hybrid identity service system is disclosed. The system may receive a transaction request. The system may generate a first passcode and a second passcode. The system may decrypt a first encrypted private key with the first passcode to recover a private key. The system may sign the transaction request with the private key and may encrypt the private key with the second passcode to generate a second encrypted private key.

Abstract: Systems and methods for automated data extraction and adaptation are disclosed. The system may receive a data input from an external source using various different input channels. The system may determine a data quality of the data input by comparing data fields of the data input to known metadata in the system. The system may reformat the data input based on the comparison to a format consumable by downstream applications and services. The system may apply various machine learning operations on the data input including a descriptive analytics analysis, a predictive learning analysis, and/or a prescriptive intelligence analysis.

Abstract: A system and method for automated remote payments between a vehicle and a refueling station is disclosed. The system may enable the vehicle to initiate an automated remote payment to the refueling station without needing the user to manually input transaction account information, or to manually prepay for the transaction. The refueling station may detect when the vehicle is in proximity, and the vehicle and the refueling station may open communications to transmit data. The vehicle may transmit vehicle identifying data to the refueling station, and the refueling station may communicate the vehicle identifying data to a payment network to authorize the transaction. In response to authorizing the transaction, the vehicle may proceed with refueling at the refueling station.

Abstract: Disclosed herein are method, system, and computer product embodiments for generating a graphical user interface (GUI) to visually represent a programmatic process with various process objects. This may also include interacting with interfaces of the GUI to update the programmatic process. Process objects may contain a hierarchy of functions. Through interaction with the GUI registration interface a process object may be modified such that the hierarchy of functions is changed. Additional process objects may be defined based on a hierarchy of functions. The GUI may also include an analysis interface which may display timing data associated with execution of the programmatic process.