Abstract: Methods and systems are provided for predicting reward liability data of reward programs. A method includes accessing, by a server system, historical reward related data associated with one or more reward programs administered by a reward program provider of reward program providers. The historical reward related data includes past redeemed reward points for each reward program aggregated on a particular time basis. Method includes identifying first seasonality patterns and second seasonality patterns associated with the historical reward related data. Method includes training a reward liability prediction model based on first and second seasonality patterns, wherein the trained time-series prediction model is configured to predict future reward liability data associated with the one or more reward programs.

Abstract: The disclosure relates to AI-based machine-learning and natural language modeling to identify semantic similarities between sets of content having natural language text. For example, a system may generate a relevance classification that indicates whether content such as articles are non-specifically relevant to charities without identifying a particular charity. If the content is non-specifically relevant to charities, the system may apply a natural language model to generate sentence embeddings based on the content and determine a level similarity between the sentence embeddings and a query embedding generated from a charity query. The charity query may itself be generated from a full description of the charity through an encoder-decoder architecture with reinforcement learning.

Abstract: A classifier is trained to classify business supplier relationships using synthetic training data samples. Real training data samples are collected and transformed into sample encodings using an encoder. The real training data samples include feature data associated with health class indicators indicative of relationships between suppliers and service providers. A set of synthetic training data samples is generated from the sample encodings using a generator and discrimination feedback data is generated using a discriminator based on the real training data samples and the synthetic training data samples. The discrimination feedback data is used to train the generator. A classifier model is trained to classify suppliers with health class indicators using the set of synthetic training data samples. The use of the encoder, generator, and discriminator enables the generation of accurate synthetic training data that represents the source distribution of real data which are often partially observed.

Abstract: Aspects of the disclosure provide a computerized method and system that utilizes reference expense reports to build and train one or more neural network learning models that intelligently determine the riskiness of to-be-determined expense reports submitted for reimbursement. In examples, a determined riskiness may inform a reimbursement entity manager when determining whether to approve, reject, and/or flag for further review a to-be-determined expense report. In instances, computerized expense report resolution systems and methods may be further automated in order to omit user interactions with to-be-determined expense reports, such that an intelligent computer determines whether to approve, reject, and/or flag a to-be-determined expense report based on the intelligently determined riskiness of the to-be-determined expense report.

Abstract: A method for optimizing invoice payments according to supplier and buyer controls includes: receiving one or more received data message including invoice data, a buyer identification value, a supplier identification value, and a plurality of buyer optimization priorities, wherein the invoice data is associated with an invoice and includes an invoice amount and due date; identifying a plurality of supplier controls associated with the supplier identification value; identifying one or more buyer preferences associated with the buyer identification value; determining an optimal payment schedule for one or more payment transactions for the invoice based on the invoice data, the buyer optimization priorities, the plurality of supplier controls, and the one or more buyer preferences; transmitting a transmitted data message including the determined optimal payment schedule.

Abstract: Systems and computer-implemented methods are described for modeling card inactivity. For example, hierarchical modeling may be used in which a first level classifier may be trained and validated to predict whether a card will be inactive. For cards predicted to become inactive by the first level classifier, a second level classifier may be trained and validated to predict when the card will become inactive. The first level classifier may include a binary classifier that generates two probabilities that respectively predict that the card will and will not become inactive. The second level classifier may include a multi-class classifier that generates a first probability that the card will become inactive at a first time period (such as one or more months in the future) and a second probability that the card will become inactive at a second time period. The multi-class classifier may generate other probabilities corresponding to other time periods.

Abstract: Embodiments provide methods and systems for identifying a re-routed transaction. Method performed by processor includes retrieving a plurality of transaction windows from a transaction database. Each transaction window includes a transaction declined under a restricted MCC. The method includes accessing a plurality of features associated with each transaction of each transaction window from the transaction database. The method includes predicting an output dataset of a plurality of reconstructed transaction windows based on feeding the input dataset to a trained neural network model. The method includes computing a corresponding reconstruction loss value for each transaction of each transaction window. The method includes comparing the corresponding reconstruction loss value for each transaction with a pre-determined threshold value.

Abstract: The disclosure relates to AI-based machine-learning and natural language modeling to identify semantic similarities between sets of content having natural language text. For example, a system may generate a relevance classification that indicates whether content such as articles are non-specifically relevant to charities without identifying a particular charity. If the content is non-specifically relevant to charities, the system may apply a natural language model to generate sentence embeddings based on the content and determine a level similarity between the sentence embeddings and a query embedding generated from a charity query. The charity query may itself be generated from a full description of the charity through an encoder-decoder architecture with reinforcement learning.

Abstract: Embodiments provide methods and systems for processing unstructured and unlabelled data. A method includes generating, by a processor, a structured and unlabelled training dataset from an unstructured and unlabelled dataset. The method includes categorizing the structured and unlabelled training dataset into a plurality of clusters by executing an unsupervised algorithm. Each cluster of a selected set of clusters from the plurality of clusters is labelled with an applicable label from a set of labels. The method includes executing a supervised algorithm to generate a trained supervised model using a labelled training dataset including the set of labels and an input dataset generated from plurality of datapoints present in each cluster of the selected set of clusters. The method includes generating a Labelled Data1 (LD1) by executing the trained supervised model configured to assign applicable label from the set of labels to each datapoint of the structured and unlabelled training dataset.

Abstract: Systems and computer-implemented methods are described for modeling card inactivity. For example, hierarchical modeling may be used in which a first level classifier may be trained and validated to predict whether a card will be inactive. For cards predicted to become inactive by the first level classifier, a second level classifier may be trained and validated to predict when the card will become inactive. The first level classifier may include a binary classifier that generates two probabilities that respectively predict that the card will and will not become inactive. The second level classifier may include a multi-class classifier that generates a first probability that the card will become inactive at a first time period (such as one or more months in the future) and a second probability that the card will become inactive at a second time period. The multi-class classifier may generate other probabilities corresponding to other time periods.

Abstract: Aspects of the disclosure provide a computerized method and system that utilizes reference expense reports to build and train one or more neural network learning models that intelligently determine the riskiness of to-be-determined expense reports submitted for reimbursement. In examples, a determined riskiness may inform a reimbursement entity manager when determining whether to approve, reject, and/or flag for further review a to-be-determined expense report. In instances, computerized expense report resolution systems and methods may be further automated in order to omit user interactions with to-be-determined expense reports, such that an intelligent computer determines whether to approve, reject, and/or flag a to-be-determined expense report based on the intelligently determined riskiness of the to-be-determined expense report.

Abstract: Disclosed are exemplary embodiments of systems and methods for use in geolocation analysis. In one exemplary method, a computing device accesses boundary definitions associated with regions. For each region, the computing device determines a centroid of the region, having a location including a latitude and a longitude, based on an associated boundary definition, truncates the centroid location by deleting the latitude or longitude, and stores the truncated location as a partition key for the region. The computing device also accesses a transaction record, including a merchant point location having a latitude and longitude. The computing device truncates the point location by deleting the latitude or longitude, and identifies regions associated with the partition keys based on a comparison of the truncated point location to the partition keys. The computing device then determines whether the point location is included in one of the identified regions.