Abstract: A customer assistance guidance system includes an interface for communicating information to a customer relationship management (CRM) system. The system receives state information from the CRM system that defines a current interaction state between an agent of the CRM system and a customer. The system may predict, based on the state information, a next interaction state between an agent of the CRM system and the customer. The system may communicate the predicted next interaction state to the CRM system to facilitate the suggestion of a next course of action for the agent to perform.

Abstract: A system may parse a group of manuscripts to extract manuscript properties and textual data included in each respective manuscript. The system may perform computer based identification of a predetermined category in which each respective manuscript resides and generate a categorized inventory of log entries, which include the manuscript properties and the textual data as respective field values. Each of the log entries may be representative of one of the manuscripts and may be categorized in the categorized inventory of log entries according to the predetermined category. The system may generate derived field values in at least some of the log entries. The derived field values may be indicative of at least some of the respective field values. The system may generate an architectural computer based dimensional mapping of a categorized inventory of log entries based on the relational association among the respective field values and the derived field values.

Abstract: Implementations include determining a set of routes for delivery of assets within a delivery network based on realized and forecasted demand for a period, during at least partial execution of the set of routes, receiving real-time data representative of delivery of assets, and updated demand forecasts, determining, by an optimization controller, a time-to-trigger, and a cut-off time, and an updated demand forecast, the time-to-trigger defining an interval, at which the set of routes for delivery of assets is to be reoptimized, in response to the time-to-trigger being achieved, re-optimizing, by a route optimization module of the continuous delivery platform, the set of routes to provide a revised set of routes, and conveying at least one route of the revised set of routes to a delivery personnel, the at least one route being different from any route in the set of routes, and being executed during the period.

Abstract: Examples of a process optimization system are provided. The system may obtain a query from a user and process data associated with the query from a plurality of data sources. The system may identify a plurality of factors and a target variable from the process data. The system may identify a factor range for the plurality of factors. The system may deconstruct the factor range to identify a plurality of data partitions. The system may identify a data pruning activator based on the plurality of data partitions and a preponderant data partition therefrom. The system may identify a plurality of clusters associated with the preponderant data partition. The system may identify a preponderant cluster from the plurality of clusters. The system may identify a confidence score associated with the preponderant cluster. The system may generate a process optimization result based on the preponderant cluster and the confidence score.

Abstract: A domain-specific question answering system receives an end-user's question, identifies relevant information from domain-specific documents, composes a natural-language based response including the relevant information to be provided to the end-user. A knowledge base includes a domain-specific ontology and is built from data structures configured to preserve the hierarchy of the informational elements within the domain-specific documents. The question received from the end-user is analyzed to extract the features and the key phrases which are used to identify the relevant data structures. A pre-trained model that is further trained on domain-specific data is used to identify the sentences that include information responsive to the question. A response to the end-user's question is composed from the sentences. The question answering system also enables validation of the answers provided to the users' questions.

Abstract: A system includes a security controller located within a central-zone in communication with a provisioned agent operating on an edge-zone device. The security controller may implement controller-blind security control on the edge-zone device by making security authorization determinations using a secure-operation confidence analysis performed by the provisioned agent. The provisioned agent, operating within the edge-zone, may have access to un-sanitized data, which may be unavailable to the security controller in the central-zone. The security controller may provide condition-descriptors via a secure-operation confidence assignment matrix. Using the condition-descriptors, the provisioned agent determine feedback, which may include a secure-operation confidence coefficient, based on the un-sanitized data to which the security controller is blind.

Abstract: In some implementations, a system may generate information that identifies a passphrase to be used as a biometric input. The system may receive a voice input of a user speaking the passphrase. The system may generate one or more cryptographic keys based on the voice input. The system may generate a digital identifier based on the one or more cryptographic keys. The system may generate one or more biometric templates for the user. The system may encrypt the one or more biometric templates using the one or more cryptographic keys and to generate one or more encrypted biometric templates. The system may store in a secure storage associated with the user, at least one of the digital identifier, a public key of the one or more cryptographic keys, a phone number associated with the user, or the one or more encrypted biometric templates. Numerous other aspects are provided.

Abstract: A system may facilitate distributed ledger technology (DLT) record based (for example, blockchain-based) voting. A voter may distribute vote-value to answers using committed tokens that bind the voter to a particular vote-value without divulging the particular vote value while in a cryptographic form. The voter may distribute committed tokens to multiple answers. In some cases, the distribution of the committed tokens to multiple answers may frustrate attempts to determine the one or more targets to which the voter delivers a non-null vote-value.

Abstract: A system may receive a request to perform a token exchange between a first account and a second account. A selected token for the token exchange may be associated with the first account. A hash of the selected token may be stored on a furnisher blockchain of a first distributed ledger. The system may generate a new token and a hash of the new token. The system may generate a shielded exchange instruction, the shielded exchange instruction comprising the hash of the new token and a zero-knowledge proof, wherein the zero-knowledge proof is indicative of the selected token being associated with the first account. The system may submit the shielded exchange instruction to a distributed ledger technology platform to cause execution of a furnisher smart contract and a receiver smart contract. Each of the smart contracts are configured to separately verify the zero-knowledge proof and update corresponding blockchains to consummate and record the token exchange.

Abstract: A robotic operations control system enables seamless distribution of tasks among a blended workforce that includes human employees and robot assistants. The robot assistants include an intelligent billing assistant (IBA), an intelligent collections assistant (iNCA), a dispute resolution assistant (DRA) and a cash processor. Each of the assistants includes a workflow manager that receives work items to be processed by the robot assistant based on the priority levels assigned to the work items. The IBA includes a dispute prediction model that predicts the likelihood of dispute for a work item and automatically generates a customized checklist. The iNCA includes an AI prioritizing assistant that identifies priorities for work items to be processed. The DRA includes an AI assisted investigator that expedites investigations of disputes by proactively gathering information for a human employee to review and mark the dispute as valid, invalid or for query.

Abstract: In some examples, recruitment process graph based unsupervised anomaly detection may include obtaining log data associated with a recruitment process for a plurality of candidates, and generating knowledge graphs and graph embeddings. The graph embeddings may be trained to include a plurality of properties such that graph embeddings of genuine candidate hires and fraudulent candidate hires are appropriately spaced in a vector space. The trained graph embeddings may be clustered to generate a plurality of embedding clusters that include a genuine candidate cluster, and a fraudulent candidate cluster. For a new candidate graph embedding for a new candidate, a determination may be made as to whether the new candidate graph embedding belongs to the genuine candidate cluster, to the fraudulent candidate cluster, or to an anomalous cluster, and instructions may be generated to respectively retain or suspend the new candidate.

Abstract: In some examples, data classification and modelling based application compliance analysis may include generating, for application data for an application, and based on data laws, a knowledge graph schema that is used to generate an instantiated knowledge graph. Based on a plurality of data sources, domain knowledge and patterns, and the instantiated knowledge graph, classified data may be generated to generate an annotated knowledge graph. Based on the annotated knowledge graph, an indication of applicable sovereign laws and compliance related to the application data for the application may be generated. Based on the annotated knowledge graph and the applicable sovereign laws and compliance related to the application data for the application, a sovereign cloud strategy related to the application data for the application may be generated. Further, the sovereign cloud strategy may be applied to the application data for the application.

Abstract: Data conversion circuitry receives structured records including normalized data and performs a data conversion process on the structured records to generate unstructured records including unstructured syntax. The data conversion circuitry performs the data conversion process according to an unstructured syntax requirement including a syntax field mapping between the structured records and the generated unstructured records.

Abstract: A knowledge graph virtualization system (KGVS) implements data virtualization techniques using semantic knowledge when parsing a knowledge graph to return results to a query. The data virtualization using semantic knowledge enable end users and applications to describe desired output without detailed knowledge of underlying data storage, schema, and relationships.

Abstract: In some examples, artificial intelligence and machine learning based project management assistance may include ascertaining an inquiry by a user. The inquiry may be related to a project. An attribute associated with the user and an attribute associated with the project may be ascertained. The inquiry may be analyzed based on the ascertained attributes associated with the user and the project. A predictor category may be identified, based on the analyzed inquiry, from a plurality of predictor categories that include a performance predictor category, a quality predictor category, a retrospect predictor category, and a planning predictor category. A predictor from a plurality of predictors may be identified based on the identified predictor category. A response to the inquiry may be generated based on execution of the identified predictor. Further, a display responsive to the inquiry may be generated based on the generated response.

Abstract: A device may receive data associated with customers, a customer-related process, or customer-related operations of an organization. The device may process the data using a technique. The device may map the data to an organization operating model. The organization operating model may be used to perform an analysis to identify a deficiency related to the customers, the customer-related process, or the customer-related operations, or a first manner in which to improve a customer-channel combination, the customer-related process, or the customer-related operations. The device may perform the analysis of the data. The device may identify the customer-channel combination of the organization. The customer-channel combination may identify a channel via which the organization acquires, engages, or retains the customers.

Abstract: A device may provide, to user devices, a set of objects that are to be used by the user devices as part of a collaborative virtual environment. The collaborative virtual environment may support a virtual workshop. The set of objects may include objects capable of being used to solve a problem that is presented to users within the virtual workshop. The device may receive data describing a set of solutions. The device may curate the data describing the set of solutions using one or more natural language processing techniques. The device may rank the set of solutions that has been curated using a ranking technique or a machine learning technique. The device may provide data identifying a highest ranked solution or data identifying the set of solutions for display within the virtual workshop or to additional user devices that are accessible outside of the virtual workshop.

Abstract: A self-learning system for analytical attribute and clustering segmentation may be provided. A text classifier may identify a log description of a log entry in response to text of the log description being associated with indicators of a word model. A datafield classifier may generate a datafield metrics including an accuracy value of the attribute identifiers representing the datafield. A metafield classifier may generate a context metrics for the context of the log entry, the context metrics including an accuracy value of the attribute identifiers representing the metafields. A combination classifier may form a weighted classification set and select an attribute identifier as being representative of the datafield based on the weighted classification set. The combination classifier may further evaluate an attribute importance value of each attribute identifier, and select an attribute identifier having a top attribute importance value.

Abstract: A device receives occupational activity descriptions and occupational role attributes, and processes the occupational activity descriptions to generate estimated occupational activity attribute values. The device trains a neural network model based on the estimated occupational activity attribute values to generate a trained neural network model, and receives a new activity description for a new role in an organization. The device processes the new activity description, with the trained neural network model, to generate estimated new activity attribute values, and processes the estimated new activity attribute values, with the logistic regression model, to generate probabilities that the new role is suitable for different workforce types. The device determines a workforce recommendation for the new role based on the probabilities that the new role is suitable for the different workforce types.

Abstract: A device may receive first information that includes a first data element. The device may receive second information that includes a second data element. The first data element and the second data element may be associated with different area identifiers. The device may associate the first information and the second information with a geographic area. The device may determine, for the geographic area, a first data element score based on the first data element. The device may determine, for the geographic area, a second data element score based on the second data element. The device may determine, for the geographic area, a market index based on the first data element score and the second data element score. The device may provide information regarding the market index to permit and/or cause an action to be performed in association with the geographic area.