Abstract: A system can receive textual input from a client device via a chatbot interface. The system can classify, by machine learning models, the textual input into categories, where the categories correspond to content-based categories and attribute categories. The system can determine, by the machine learning models, a polarity score for the textual input based on the categories. The system can detect, via a payroll processing system, a payroll-related event based on the polarity score and the profile. The system can identify, using the machine learning models, a churning event based on the payroll-related event. The system can generate, by the machine learning models, a command to adjust access control policies associated with the profile, where the command modifies access to a subset of payroll records maintained in a database.

Abstract: A system can receive textual input from a client device via a chatbot interface. The system can identify a profile data structure based on an identifier of the client device. The system can classify, by machine learning models, the textual input into categories, where the categories correspond to content-based categories and attribute categories. The system can determine, by the machine learning models, a polarity score for the textual input based on the categories. The system can input the categories, the polarity score, and the identifier into a payroll processing system to identify a payroll-related event and extract data corresponding to the payroll-related event. The system can provide the extracted data to the machine learning models to generate an output response data structure. The system can transmit the output response data structure to the client device to parse the output response data structure and present a corresponding message dialog.

Abstract: A method receives an electronic image and uses the image as an input to a neural network. Based on a determination that the image represents a document, the method uses the image as an input to another neural network to identify a portion of the document containing an identifier. The method extracts the identifier by performing character recognition on the identified portion and determines whether the identifier is valid by using a validation API to determine whether the identifier is associated with a valid account at an institution. Based on a determination that the identifier is associated with a valid account, the method authorizes a transaction associated with the identifier. Based on a determination that the identifier is not associated with a valid account, the method denies the transaction. The first neural network classifies the electronic image into one of multiple valid document types and an invalid document type.

Abstract: A method for fact retrieval includes receiving an input from a user, said input including a string of text that represents a user intent. The method includes performing natural language processing on the input to generate an embedding that corresponds to a semantic representation of the string of text, and based on the generated embedding, identifying an action that is associated with the user intent. The method includes executing the identified action, such that the executed action returns a result associated with the user intent, and providing the result to the user.

Abstract: A method receives an electronic image and uses the image as an input to a neural network. Based on a determination that the image represents a document, the method uses the image as an input to another neural network to identify a portion of the document containing an identifier. The method extracts the identifier by performing character recognition on the identified portion and determines whether the identifier is valid by using a validation API to determine whether the identifier is associated with a valid account at an institution. Based on a determination that the identifier is associated with a valid account, the method authorizes a transaction associated with the identifier. Based on a determination that the identifier is not associated with a valid account, the method denies the transaction. The first neural network classifies the electronic image into one of multiple valid document types and an invalid document type.

Abstract: Systems, methods, and computer-readable storage media for recommending computational commands based on a weighted, ranked hierarchical graph. The system converts a map of intended user actions to canonical actions (commands) into a graph representation. Then, via a chatbot algorithm, the system receives an initial intent as a request by a user of the chatbot algorithm. Based on the popularity of alternative canonical actions, the system makes recommendations of alternative actions which are similar to the initial intent provided by the user.

Abstract: A method for predicting customer churn includes receiving a graph data structure storing data associated with activity of a user, the graph data structure having multiple nodes, including a user input node associated with the user. The method includes updating at least the user input node with a vector representation of a received user input, and using historical user input, training a sentiment model to classify user input according to one of multiple sentiments. The method includes using the trained sentiment model to classify the received user input as a particular sentiment, and adding to the graph data structure a sentiment node that is associated with the particular sentiment and that is connected to the user input node. The method includes, using the graph data structure, training a churn model to estimate customer churn probability, and using the trained churned model to estimate a particular churn probability for the user.

Abstract: Aspects extract, from payroll data of employees of an organization, data historically associated to previous instances of certified tax credit eligibility; normalize the extracted data with respect to data type and data value; generate from the normalized extracted data via a neural network classifier multi-class outputs for each employee that indicate strengths of likelihood that each employee is currently eligible for each of a plurality of different tax credits; filter the normalized extracted data by removing portions associated to employees indicated within the multi-class outputs as having no currently eligible likelihood for the different tax credits, thereby generating a remainder set of normalized extracted data associated to remainder eligible ones of the employees; and prioritize application for the tax credits for the remainder eligible employees as a function of respective values and likelihoods of eligibility within the remainder set of normalized extracted data.

Abstract: Systems, methods, and computer-readable storage media for identifying and generating insights using machine learning and serverless computing systems. When events are detected, data from those events is collected and formatted into predefined formats. That formatted data is then used within an instance, where a machine learning algorithm is executed by a serverless computing system using the data within the instance as input. The result from the machine learning algorithm is an insight into the event and event data, which can be presented to users for interpretation.

Abstract: Systems, methods, and computer-readable storage media for responding to a query using a neural network and natural language processing. If necessary, the system can request disambiguation, then parse the query using a trained machine-learning classifier, resulting in at least one of an identified subject or an identified domain of the text query. The system can determine if the user is authorized to retrieve answers to the query and, if so, retrieve factual data associated with the query. The system can then retrieve a response template, and fill in the template with the retrieved facts. The system can then determine, by executing a machine comprehension model on the filled response template, a probable readability token, portion of text, of at least a portion of the filled response template and, upon identifying that the probable readability is above a threshold, reply to the text query with the at least a portion of the filled response template.

Abstract: A method, computer system, and computer program product are provided for facilitating peer-to-peer micro-loan transactions. A micro-loan system determines integrity scores for a plurality of users. The integrity scores are based on human capital management information and peer-submitted feedback. These micro-loan system receives a loan transaction request from a borrower-user, and determines a risk score for the loan transaction based the borrower's integrity score. The micro-loan system identifies potential lenders based on the determined transaction risk and the lenders' integrity scores. The micro-loan system facilitates a negotiation between the borrower and the potential lenders to determine transaction terms. When terms are finalized, the loan transaction and the transaction terms are recorded in a distributed ledger, and loan funds are remitted to the borrower.

Abstract: Systems, methods, and computer-readable storage media for recommending computational commands based on a weighted, ranked hierarchical graph. The system converts a map of intended user actions to canonical actions (commands) into a graph representation. Then, via a chatbot algorithm, the system receives an initial intent as a request by a user of the chatbot algorithm. Based on the popularity of alternative canonical actions, the system makes recommendations of alternative actions which are similar to the initial intent provided by the user.

Abstract: A method, computer system, and computer program product are provided for managing time record events. Time record events are collected for a number of users. Each time record event includes a geolocation of one of a number of users. The time record events and geolocations for each of the number of users are models via machine learning. A current geolocation for a given user is identified. A suggested event is predicted based on the current geolocation and a current time. The suggested event is pushed to the user.

Abstract: Systems, methods, and computer-readable storage media for responding to a query using a neural network and natural language processing. If necessary, the system can request disambiguation, then parse the query using a trained machine-learning classifier, resulting in at least one of an identified subject or an identified domain of the text query. The system can determine if the user is authorized to retrieve answers to the query and, if so, retrieve factual data associated with the query. The system can then retrieve a response template, and fill in the template with the retrieved facts. The system can then determine, by executing a machine comprehension model on the filled response template, a probable readability token, portion of text, of at least a portion of the filled response template and, upon identifying that the probable readability is above a threshold, reply to the text query with the at least a portion of the filled response template.

Abstract: Systems, methods, and computer-readable storage media for identifying and generating insights using machine learning and serverless computing systems. When events are detected, data from those events is collected and formatted into predefined formats. That formatted data is then used within an instance, where a machine learning algorithm is executed by a serverless computing system using the data within the instance as input. The result from the machine learning algorithm is an insight into the event and event data, which can be presented to users for interpretation.

Abstract: Aspects extract, from payroll data of employees of an organization, data historically associated to previous instances of certified tax credit eligibility; normalize the extracted data with respect to data type and data value; generate from the normalized extracted data via a neural network classifier multi-class outputs for each employee that indicate strengths of likelihood that each employee is currently eligible for each of a plurality of different tax credits; filter the normalized extracted data by removing portions associated to employees indicated within the multi-class outputs as having no currently eligible likelihood for the different tax credits, thereby generating a remainder set of normalized extracted data associated to remainder eligible ones of the employees; and prioritize application for the tax credits for the remainder eligible employees as a function of respective values and likelihoods of eligibility within the remainder set of normalized extracted data.

Abstract: Systems, methods, and computer-readable storage media for responding to a query using a neural network and natural language processing. If necessary, the system can request disambiguation, then parse the query using a trained machine-learning classifier, resulting in at least one of an identified subject or an identified domain of the text query. The system can determine if the user is authorized to retrieve answers to the query and, if so, retrieve factual data associated with the query. The system can then retrieve a response template, and fill in the template with the retrieved facts. The system can then determine, by executing a machine comprehension model on the filled response template, a probable readability token, portion of text, of at least a portion of the filled response template and, upon identifying that the probable readability is above a threshold, reply to the text query with the at least a portion of the filled response template.

Abstract: A method receives an electronic image and uses the image as an input to a neural network. Based on a determination that the image represents a document, the method uses the image as an input to another neural network to identify a portion of the document containing an identifier. The method extracts the identifier by performing character recognition on the identified portion and determines whether the identifier is valid by using a validation API to determine whether the identifier is associated with a valid account at an institution. Based on a determination that the identifier is associated with a valid account, the method authorizes a transaction associated with the identifier. Based on a determination that the identifier is not associated with a valid account, the method denies the transaction. The first neural network classifies the electronic image into one of multiple valid document types and an invalid document type.

Abstract: A method, computer system, and computer program product are provided for managing time record events. Time record events are collected for a number of users. Each time record event includes a geolocation of one of a number of users. The time record events and geolocations for each of the number of users are models via machine learning. A current geolocation for a given user is identified. A suggested event is predicted based on the current geolocation and a current time. The suggested event is pushed to the user.

Abstract: Systems, methods, and computer-readable storage media for recommending computational commands based on a weighted, ranked hierarchical graph. The system converts a map of intended user actions to canonical actions (commands) into a graph representation. Then, via a chatbot algorithm, the system receives an initial intent as a request by a user of the chatbot algorithm. Based on the popularity of alternative canonical actions, the system makes recommendations of alternative actions which are similar to the initial intent provided by the user.