Abstract: Implementations include determining a set of components within the connected vehicle ecosystem, components within the set of components representing at least one layer within the connected vehicle ecosystem, for each component in the set of components: providing a set of facts representative of the respective component, and providing a component digital twin using the set of facts, defining a set of digital twins including digital twins of components in the set of components, generating, using the set of digital twins, at least one AAG representative of potential lateral movement between components of the at least one layer within the connected vehicle ecosystem, the at least one AAG representing a contextual digital twin of components operating within the connected vehicle ecosystem, and evaluating the connected vehicle ecosystem using the at least one AAG.

Abstract: Systems and methods for computing feature contribution and providing hum-interpretable reason codes for a Support Vector Machine (SVM) model are disclosed, A system computes, for each data point from amongst plurality of data points indicative of plurality of features, a feature contribution of each one of the plurality of features for a SVM model used for at least one of classification decision and a regression analysis. Further, the system provides human interpretable reason code for the interpretation corresponding to at least one of classification decision and the regression analysis from the SVM model. The system outputs to the user, a feature contribution output, and the human interpretable reason codes output. The feature contribution output and the human interpretable reason codes output are indicative of an acceptable decision to be taken by the user based on the classification decision and the regression analysis received from the SVM model.

Abstract: An Artificial Intelligence (AI) based document processing and validation system identifies anomalies such as errors, fraud, and duplicates of received documents and enables automatic actions for valid documents using machine learning (ML) techniques. The received documents are processed for determining probabilities for errors, fraud, and duplicates. A validation worklist is generated with the documents arranged in descending order of the probabilities and invalid documents with higher probabilities are flagged for review while the valid documents with lower probabilities are further processed for the execution of automatic actions. The feedback from the invalid document review is used to further train the models in determining the probabilities.

Abstract: A device may receive log data from application logs associated with applications, service logs associated with services, and server logs associated with server devices. The device may store the log data. The device may perform natural language processing on the log data to convert the log data into event data identifying events associated with categories. The device may process the event data, with a first machine learning model, to identify patterns in the event data and to generate an alert based on the patterns. The device may process the event data, with a second machine learning model, to generate a correlation matrix for the event data and to predict an event based on the correlation matrix. The device may process the event data, with a third machine learning model, to classify the event data based on the categories and to generate a recommendation based on classifying the event data.

Abstract: In some implementations, a device may receive prescription information associated with a medication in a container. The device may cause a camera to capture first image data associated with the medication while the medication is in the container and the container is positioned on a receptacle. The device may cause an adjusting device to reposition the container on the receptacle. The device may cause the camera to capture second image data associated with the medication while the medication is in the container. The device may process, via a neural network, the first image data and the second image data to identify the medication based on depictions of individual units of the medication included in the first image data and the second image data. The device may verify the medication based on the prescription information and an identifier of the medication provided by the neural network.

Abstract: This document describes a simulation system that simulates robots and other actors performing tasks in an area. In one aspect, a method includes obtaining a graph representing a physical area. The graph includes area nodes that represent regions of the area that are traversed by a set of actors that perform tasks in the area and terminal nodes that represent regions of the facility where the actors perform the tasks. A set of agents that each include a model corresponding to an actor is identified. At least a portion of the agents includes models for robots that perform tasks in the area. The model of an agent represents durations of time for traversing area nodes and performing tasks are terminal nodes during simulations. A sequence of tasks being performed in the area is simulated using the graph and the set of agents.

Abstract: A device may receive historic temporal data identifying events associated with a system, and may perform block bootstrapping of the hierarchical time series data, based on a hyperparameters, to generate blocks of data points of the historic time series data. The device may process the blocks of data points, with a plurality of different machine learning models, to calculate predictions, and may apply weights to the predictions to generate weighted predictions. The device may aggregate the weighted predictions to generate aggregated predictions, and may apply final weights to the aggregated predictions to generate weighted aggregated predictions. The device may aggregate the weighted aggregated predictions to generate a final prediction, and may perform one or more actions based on the final prediction.

Abstract: A device may receive network data, business data, and user configuration data associated with an entity that is a candidate for a private network and may process the business data and the user configuration data, with a classification machine learning model, to determine a network hardware equipment prediction. The device may process the network data and the business data, with a first linear regression machine learning model, to determine a business output prediction and may utilize a second linear regression machine learning model to determine a data consumption prediction based on the network hardware equipment prediction. The device may process the network hardware equipment prediction, the business output prediction, and the data consumption prediction, with a machine learning model, to determine a financial profitability prediction for the private network and may perform one or more actions based on the financial profitability prediction.

Abstract: A device may receive employee data associated with an employee, and may determine, from the employee data, skill data identifying skills of the employee. The device may process the employee data and the skill data, with a similarity model, to determine similarity scores between the skills of the skill data, and may add or remove one or more skills to or from the skill data for predefined target skill profile categories to generate modified skill data. The device may compare the similarity scores and a predefined threshold to determine anchor skill data from the modified skill data, and may group the anchor skill data for corresponding pluralities of the predefined target skill profile categories to generate groups of the anchor skill data. The device may process the groups and the similarity scores, with a clustering model, to generate clustered anchor skill data, and may perform actions based on the clustered anchor skill data.

Abstract: The present disclosure relates to systems and methods for carrying out predictive analysis where a plurality of data sets may be ingested from a data lake. A data analyzer may tag the ingested data sets, detect redundant occurrence of multiple attributes such as, a row, a column, and a list in the tagged data set. The data analyzer may eliminate the detected redundant multiple attributes. Further, a model selector and evaluator may execute a machine learning (ML) model to conduct predictive analysis on the data set. The execution may be done based on a predefined set of instructions stored in a database. The executed ML model may be validated upon determining that the predictive analysis yields a positive response for the transformed data set.

Abstract: A document processing system processes documents including typewritten and/or handwritten data by converting them to document images for entity extraction. A received document is initially processed to generate a deep document data structured and for classification as one of a structured or an unstructured document. If the document is classified as a structured document, it is processed for entity extraction based on a matching template and image alignment of the document image with the matching template. If the document is classified as an unstructured document, entities are extracted by obtaining nodes and providing the nodes to a self-supervised masked visual language model.

Abstract: The disclosure provides a non-opaque, abstract, unified query language that exposes the query as a first-class citizen of the underlying architecture. The present disclosure thus facilitates the creation of no-code or low-code applications by enabling a level of collaboration between components that may be difficult to achieve if the language employed were opaque to the architecture. The disclosed query language may be considered “SQL-like,” which may allow contributors familiar with structured query language (SQL) to effectively participate in the design of an application. The defined structures of a data objects of the non-opaque query language described herein are not-hidden and inspectable.

Abstract: An automated system and method of converting legacy decision logic to a target format. The legacy files are received by the decision logic translation system, which outputs the business rule content in a standard rule structure, according to the selected target format. The process involves decision logic-based rule extraction. In general, methods or processes for extracting business rules have been difficult to reproduce and do not present clearly the extracted rules regarding the concepts of business rules, their composition and categorization. These drawbacks lead to incomplete extraction of rules and massive manual effort to achieve a complete extraction and verification. In contrast, the proposed system overcomes these drawbacks, and outputs files that can be easily used to migrate the business rules to a new platform.

Abstract: Implementations include providing, by the PKG platform, an initial knowledge graph based on user-specific data associated with a user, and a domain-specific knowledge graph, receiving, by the PKG platform, data representative of at least one answer provided from the user to a respective question, providing, by the PKG platform, an expanded knowledge graph based on the initial knowledge graph, the expanded knowledge graph including one or more nodes and respective edges based on the data, generating, by the PKG platform, a weighted knowledge graph based a groundtruth knowledge graph, and a targeted knowledge graph, the groundtruth knowledge graph including one or more true answers, and the targeted knowledge graph including the at least one answer provided from the user, and generating, by the PKG platform, the hyper-personalized knowledge graph (hpKG) based on the weighted knowledge graph, the hpKG being unique to the user within a domain.

Abstract: A plurality of personalized news feeds are generated from input feeds including digital content items based on a dynamic taxonomy data structure. Entities are extracted from the input feeds and relationship strengths are obtained for the extracted entities and the digital content items. The dynamic taxonomy data structure is updated with the extracted entities and entries for the digital content news items are included at the corresponding branches based on the relationship strengths. Attributes are obtained for the entities and those entities corresponding to the trending topics are identified. Personalized news feeds are generated including the digital content items listed under the entities. Digital content items are added or removed from the digital content feeds based on one or more entity attributes.

Abstract: In some implementations, a device may receive a request to merge a first cloud computing instance with a second cloud computing instance to generate a multi-cloud computing instance. The device may access a first application programming interface to obtain a first configuration of the first cloud computing instance. The device may access a second application programming interface to obtain a second configuration of the second cloud computing instance. The device may generate a target configuration based on the first configuration or the second configuration. The device may instantiate a set of resources with the target configuration for the multi-cloud computing instance. The device may provide output identifying the multi-cloud computing instance.

Abstract: A method and system for cloud based service provider performance evaluation are disclosed. The method may include extracting service record data and ticket status change record data from the service ticket data and aggregating the service record data and the ticket status change record data. The method may include calculating ticket level performance metric data based on the aggregated record data and generating ticket level performance scores based on the ticket level performance metric data. The method may further include generating service level performance scores based on the service level performance metric data and generating service feedback performance scores based on the service feedback metric data. The method may further include merging the ticket level performance scores, the service level performance scores, and the service feedback performance scores to generate performance vectors and evaluating overall performances of the service providers based on the set of performance vectors.

Abstract: Systems and methods for managing flow configurable event-driven microservices are disclosed. A system provides, for each function in sequence of steps corresponding to transaction flow associated with model of event-driven microservices, an address of associated memory space using route name. The system connects each function with another function in memory space and/or different application instances with automatic service discovery. The system transports, for event-driven microservices, an input and/or output comprising message and/or an event comprising object corresponding to payload, in an event envelope, based on a configuration file/a configuration file-like handle. Additionally, the system executes a sequence of steps connecting each function in sequential mode or parallel mode, based on a configuration file/a configuration file-like handle.

Abstract: A temporal-aware or permutation-dependent Graph Neural Network (GNN) is disclosed. The example GNN is implemented by combining temporal-awareness with multi-layer neighborhood aggregation to further provide the GNN with inductive capabilities with respect to generating embeddings of a dynamic graph, all without creating multiple time snapshots of the graph. By using a temporal-aware message pass scheme involving a temporal-aware and permutation-dependent GNN, a set of temporal-aware local neighborhood aggregator functions may be effective trained and used for generating embeddings for unknow nodes and for providing more accurate embeddings for subsequent prediction tasks.

Abstract: Implementations are directed to processing a content object model through a ML model to provide a set of base content feature representations, processing a style object model through the ML model to provide sets of base style feature representations, executing iterations including: generating, by the ML model, sets of stylized feature representations for an initial stylized object model, the initial stylized object model having one or more adjusted parameters relative to a previous iteration, determining a total loss based on the sets of stylized feature representations, the set of base content feature representations, and the sets of base style feature representations, and determining that the total loss is non-optimized, and in response, initiating a next iteration, executing an iteration of the iterative process, the iteration including determining that the total loss is optimized, and in response providing the initial stylized object model as output of the iterative process.