Abstract: Systems and methods of optimizing cash flow are disclosed. A system obtains bill information regarding a plurality of bills and invoice information regarding a plurality of invoices, and the system pairs one or more bills to one or more invoices. Pairing the one or more bills includes, for each bill, generating one or more potential pairs of the bill to an invoice. For each potential pair, the system calculates a matching score associated with the potential pair based on the bill information of the bill and the invoice information of the invoice, identifies a subset of potential pairs of the one or more potential pairs associated with a threshold matching score, and selects a pair of a paired invoice to the bill from the subset of potential pairs. The system generates instructions to automatically pay the one or more bills, with payment scheduled based on the pairings.

Abstract: A computer-implemented method and system are provided to utilize machine learning technology to process user financial transaction data to predict a personalized payment screen architecture. A plurality of feature datasets associated with transaction data of a plurality of electronic invoices are obtained by a computing device. Each feature dataset comprises a plurality of features, a payment screen and a payment method configured to be presented on at least one payment screen. The computing device is configured to train a machine learning model with the feature datasets to produce a probability matrix with probabilities of each payment method used to pay the invoices through each payment screen. The computing device may weigh the probability matrix to generate a recommendation matrix and determine a prediction of a payment screen based on the recommendation matrix.

Abstract: Systems and methods for optimally formatting item identifiers (ID) are disclosed. An example method is performed by one or more processors of a system and includes obtaining descriptions of items, identifying, for each item, one or more attributes of the item described in the item's description, extracting a value for each of the identified attributes, identifying a set of common attributes among the identified attributes for which values were extracted for more than a threshold ratio of the items, assigning a priority weight to each of the common attributes using an optimization algorithm, identifying a set of optimum attributes among the set of common attributes based on the priority weights, mapping an optimum code to each unique value extracted for the optimum attributes, and generating an optimum ID format that provides, for each item, a unique ID including the optimum codes mapped to the values of the item's optimum attributes.

Abstract: Certain aspects of the present disclosure provide techniques for detecting errors in account numbers. One example method generally includes receiving, from a user device, an entered number associated with a user and determining, based on a first portion of the entered number, an entity associated with the entered number. The method further includes obtaining, from an account number database, a plurality of account numbers associated with the entity and generating, from the plurality of account numbers, an account number matrix. The method further includes attempting to solve a multiplication equation of the account number matrix, wherein a solution of the multiplication equation is a vector of constants, upon determining a solution to the multiplication equation, determining whether the entered vector is a valid number for the entity and upon determining the entered vector is a valid number for the entity, storing the entered number in the account number database.

Abstract: Transactions include text fields, such as description fields. Transactions are extracted from financial institutions using web-scraping extraction. In the process of extracting transactions, errors can be introduced into text fields, such as the inclusion of a space within a word or the removal of a space between words. A statistical approach is applied to the text fields. When two alternative text fields are presented, the alternative that statistically includes more common tokens, such as unigrams and bigrams, is chosen as the correct alternative. The incorrect alternative is replaced by the correct alternative in the text field.

Abstract: A method may include receiving, for a package, shipment details including attributes, obtaining, for a subset of the attributes, logistic preferences, applying the logistic preferences to the shipment details to obtain modified shipment details, training a classifier using shipment transactions each including values for the attributes and labeled with a vendor logistic service, generating, by applying the classifier to the modified shipment details, scores for vendor logistic services, and recommending a vendor logistic service from the vendor logistic services using the scores.

Abstract: Certain aspects of the present disclosure provide techniques for a personalized reporting service. Software applications can provide relevant reports that aggregate time series data to users that meet certain baseline values, including a threshold, timeframe, and cadence. The baseline values can be determined by a trained machine-learning model that identifies “interesting” trend components in the time series data. The reporting service can receive configurations from the user including feedback that are utilized by the machine-learning model to update the baseline values and provide relevant reports to the user.

Abstract: Certain aspects of the present disclosure provide techniques for training decision trees representing users of a software application. An example method generally includes generating, from a transaction history data set for a plurality of users of a software application, a plurality of grouped data sets including transactions grouped by counterparty. A plurality of feature vectors are generated from the plurality of grouped data sets. Each feature vector generally corresponds to a user of the plurality of users and includes a plurality of features describing relationships between the user and a plurality of counterparties in a transaction history associated with the user. A decision tree is trained based on the plurality of feature vectors. The decision tree generally includes a plurality of paths terminating in a similar or different classification, and the plurality of paths distinguishes a user associated with the decision tree from other users of the software application.

Abstract: A method collects word-based data corresponding to a first identifier. A first phrase vector is generated for the first identifier by extracting frequency data from the word-based data. A similarity metric is generated corresponding to the first identifier and a second identifier by comparing the first phrase vector of the first identifier to a second phrase vector of the second identifier. A tuple is generated that includes the first identifier and the second identifier using the similarity metric. A machine learning model is trained with the tuple to generate an embedded vector corresponding to the first identifier.

Abstract: Described herein are example implementations for generating a perturbation seed for the perturbation of electronic data. A system obtains a plurality of datapoints (with one or more statistics calculated from the plurality of datapoints to be perturbed based on a perturbation seed). The system calculates one or more metrics from the plurality of datapoints. The system also generates, for each of the one or more metrics, a rounded metric by rounding the metric. The system further generates the perturbation seed. Generating the perturbation seed includes hashing the one or more rounded metrics. Rounding a metric may be to a defined place value (such as the second most significant place value), and a binary output of hashing the one or more rounded metrics may be converted to a number. The system may perturb one or more statistics based on the perturbation seed and output the one or more perturbed statistics.

Abstract: Techniques for detecting fraud may include mapping routing numbers of one or more financial institutions with geolocation data of the financial institutions; obtaining a geolocation of a user based on the user's internet protocol (IP) address; obtaining a first user input from the user indicating a first financial institution; generating a match score for each of the one or more financial institutions that indicates a level of match between the first user input and the respective financial institution; boosting the match score for each financial institution based on its location with respect to the geolocation of the user; generating a list of financial institutions having the boosted match score above a threshold; obtaining a second user input from the user indicating at least one second financial institution; and presenting search results to the user based on the second user input, wherein the search results are boosted.

Abstract: A processor may receive an image of a user and an avatar representing the user within a computing environment. The processor may generate a score for the avatar on the basis of its resemblance to the image using a machine learning (ML) process. The processor may configure at least one option for action by the user within the computing environment according to the score.

Abstract: Methods and systems for efficiently generating tagged training data for machine learning models. In conventional systems, all of the raw data (e.g., each sentence) has to be manually tagged. Instead, the methods and systems generate a representative sample for multiple portions of raw data, e.g., a representative sentence for multiple, similar sentences. Only the representative sample is tagged and used for training, thereby realizing a significant efficiency in both tagging the data and training the machine learning models.

Abstract: Knowledge about a user is used to determine whether one or more messages received by the user are malicious. The knowledge about the user may be based on the user's financial history such as transaction records. Particularly, a classifier model is trained on a supervised approach using a dataset containing, for example, a categorization of incoming messages (e.g., password change message), the user's aggregated transaction records, message attributes, user attributes, and corresponding classification labels. After the training, the classifier model is deployed to determine whether an incoming message is malicious.

Abstract: Systems and methods for k-anonymizing a corpus of documents using linguistic similarities and embeddings distances between words. For instance, a word pair is selected based on linguistic similarity (e.g., belonging to the same part of speech) and small embeddings distance. For the selected word pair, a plurality of words is retrieved, also based on linguistic similarity to, and embeddings distances from, the selected word pair. Out of the plurality of words, a third word is identified that has a closer linguistic similarity to the word pair and also has smaller embeddings distances from the word pair. Each word in the word pair is then replaced by the third word. The process is repeated until k-anonymity is achieved.

Abstract: Described herein are example implementations for handling of phishing attempts. A system receives a request to perform an electronic transaction, with the request including information regarding a user account. The system generates one or more probabilities of the request being valid based on the request and processing of a plurality of electronic transactions associated with one or more user accounts, identifies whether the request is valid based on the one or more probabilities, and in response to identifying that the request is not valid, provides an indication that the request is not valid.

Abstract: A method including embedding, by a trained issue MLM (machine learning model), a new natural language issue statement into an issue vector. An inner product of the issue vector with an actions matrix is calculated. The actions matrix includes centroid-vectors calculated using a clustering method from a second output of a trained action MLM which embedded prior actions expressed in natural language action statements taken as a result of prior natural issue statements. Calculating the inner product results in probabilities associated with the prior actions. Each of the probabilities represents a corresponding estimate that a corresponding prior action is relevant to the issue vector. A list of proposed actions relevant to the issue vector is generated by comparing the probabilities to a threshold value and selecting a subset of the prior actions with corresponding probabilities above the threshold. The list of proposed actions is transmitted to a user device.

Abstract: Systems and methods are provided to structure event description data.

Abstract: Aspects of the present disclosure provide techniques for automated text amendment. Embodiments include identifying a first plurality of n-grams in first text associated with a domain. Embodiments include identifying a second plurality of n-grams in second text associated with the domain. Embodiments include identifying a third plurality of n-grams in third text that is not associated with the domain. Embodiments include determining candidate n-grams that are overexpressed in the second plurality of n-grams compared to the third plurality of n-grams. Embodiments include determining a match between a candidate n-gram of the candidate n-grams and a given n-gram of the first plurality of n-grams based on one or more matching factors. Embodiments include amending the first text based on the match between the candidate n-gram and the given n-gram.

Abstract: A system includes a categorization application including a personal categorization model for execution on a user device, the personal categorization model configured to assign a personal assigned category and a personal assigned categorization confidence value to transaction record. The system also includes a global categorization model executing on a central server. The global categorization model is configured to receive the transaction record from the user device, and process the transaction record through the global categorization model to obtain a global assigned category and a global assigned categorization confidence value. The system is configured to select a transaction category from the personal assigned category and the global assigned category using the personal assigned categorization confidence value and the global assigned categorization confidence value, and categorize the transaction with the transaction category.