Abstract: An artificial intelligence system configured to detect anomalies in transaction data sets. The system includes a processor and a computer readable medium operably coupled thereto, the computer readable medium comprising a plurality of instructions stored in association therewith that are accessible to, and executable by, the processor, to perform modeling operations which include receiving a first data set for training a first machine learning model to detect anomalies in the transaction data sets using a machine learning technique, accessing at least one micro-model trained using at least one second data set separate from the first data set, determining risk scores from the first data set using the at least one micro-model, enriching the first data set with the risk scores, and determining the first machine learning model for the enriched first data set using the machine learning technique.

Abstract: Disclosed is an electronic device. The electronic device includes: a processor, and a memory operatively connected to the processor, the memory stores instructions that, when executed, cause the processor to: select at least one data received through a user input, analyze the selected data, extract additional data based on the analyzed data, learn a personal voice model using the data and the additional data, and provide response data using the personal voice model.

Abstract: There is a need for more effective and efficient text categorization. This need can be addressed by, for example, techniques for semantic text categorization. In one example, a method includes determining an input vector-based representation of an input document; processing the input vector-based representation using a trained supervised machine learning model to generate the categorization based at least in part on the input vector-based representation, wherein: (i) the trained supervised machine learning model has been trained using automatically-generated training data, and (ii) the automatically generated training data is generated by determining an inferred semantic label for each unlabeled training document of one or more unlabeled training documents; and performing one or more categorization-based actions based at least in part on the categorization, and (iii) the labels are described by one or more short documents/short texts.

Abstract: A computer implemented method of performing large-scale machine learning experiments includes expanding on one or more input datasets by systematically generating several data set drift splits. A set of experimental jobs corresponding to the generated data set drift splits are executed to generate experimental results. The experimental results are processed, consolidated, and clustered according to the generated data set drift splits.

Abstract: Disclosed herein are system, method, and computer program product embodiments for detecting erroneous data. In an embodiment, a data monitoring system may store an initial dataset. The data monitoring system may analyze the initial dataset to generate rules associated with the initial dataset. The data monitoring system may receive a new data entry from a client device intended to be associated with the initial dataset. The data monitoring system may compare the new data entry to the previously determined rules to determine if the new data entry complies. If so, the data monitoring system may store the new data entry. If not, the data monitoring system may generate an alert, requesting a confirmation that the noncompliant data entry is correct. If the noncompliant data is confirmed as correct, the data monitoring system may store the new data entry and update the rules associated with the updated dataset.

Abstract: In various embodiments, the disclosed systems and methods may receive documents, analyze the documents, categorize portions of the analyzed documents, and present the images of the documents and at least a portion of the categories. The analysis may include identification of categories and the presentation may include indicia of the portion of the image of the document related to the category. The systems and methods disclosed may allow querying and/or reporting of a plurality of documents to facilitate processing.

Abstract: A document creation system includes a term object creation configured to create term objects in which terms and attributes thereof are integrated, a term object management configured to record and manage the created term objects, a term object selection configured to select a necessary term object from among the term objects managed in the term object management, and a term object placement configured to place the selected term object in a document being edited.

Abstract: Some embodiments of the current disclosure disclose methods and systems for batch active learning using the Shapley values of data points. In some embodiments, Shapley values of a first subset of labeled data are used to measure the contributions of the first subset of data to the performance of neural network. Further, a regression model that correlates the first subset of data to their Shapley values is trained to predict the Shapley values of a second subset of data that are unlabeled. A portion of the second subset of data may then be selected for labeling based on the predicted Shapley values.

Abstract: Various embodiments include a method for monitoring the status of a production order in a factory. The method may include: generating a production IoT model based on a production scheduling system document, the production IoT model comprising a first set process attributes of product processing; generating a product IoT model based on a product design specification document, the product IoT model comprising the first set of process attributes of product processing; associating the production IoT model with the product IoT model; learning data of a production device acquired by a data acquisition automation control system in the factory to obtain a data model representing processing steps of a product; and matching the processing steps against the process attributes of the product IoT model and determining the status of the production order in the factory based on the matching result.

Abstract: A method for time series analysis of time-oriented usage data pertaining to computing resources of a computing system. A method embodiment commences upon collecting time series datasets, individual ones of the time series datasets comprising time-oriented usage data of a respective individual computing resource. A plurality of prediction models are trained using portions of time-oriented data. The trained models are evaluated to determine quantitative measures pertaining to predictive accuracy. One of the trained models is selected and then applied over another time series dataset of the individual resource to generate a plurality of individual resource usage predictions. The individual resource usage predictions are used to calculate seasonally-adjusted resource usage demand amounts over a future time period. The resource usage demand amounts are compared to availability of the resource to form a runway that refers to a future time period when the resource is predicted to be demanded to its capacity.

Abstract: A learning device is configured to perform learning of a decision tree. The learning device includes a branch score calculator, and a scaling unit. The branch score calculator is configured to calculate a branch score used for determining a branch condition for a node of the decision tree based on a cumulative sum of gradient information corresponding to each value of a feature amount of learning data. The scaling unit is configured to perform scaling on a value related to the cumulative sum used for calculating the branch score by the branch score calculator to fall within a numerical range with which the branch score is capable of being calculated.

Abstract: The present invention relates to a method, a server and a computer-readable medium for generating an interactive content, and more particularly, to a method, a server and a computer-readable medium for generating an interactive content, in which node generation information for each of a plurality of nodes included in the interactive content is received from a user terminal of a user who wants to generate the interactive content, and a plurality of nodes including an interactive video in which a specific interactive object utters text information included in the received node generation information, and node connection information for connection with one or more other nodes corresponding to a next node according to condition information included in the node generation information are generated, so that the interactive content including the nodes is generated.

Abstract: A schema associated with a type of document is received. A set of documents are queried for documents that match the schema, wherein the query is performed while bypassing user intervention. A subset of documents of the set of documents is identified that match the schema. The subset of documents that match the schema are transmitted to a document controller.

Abstract: In an example, an apparatus comprises a plurality of processing unit cores, a plurality of cache memory modules associated with the plurality of processing unit cores, and a machine learning model communicatively coupled to the plurality of processing unit cores, wherein the plurality of cache memory modules share cache coherency data with the machine learning model. Other embodiments are also disclosed and claimed.

Abstract: This application discloses a method, an apparatus and an electronic device for error correction of numerical contents in a text, and relates to a technology field of artificial intelligence such as natural language processing and deep learning. The implementation method is: obtaining a target text to be processed; determining original numerical contents included in the target text; determining target types corresponding to the original numerical contents; and performing error correction on each original numerical content according to an error correction manner corresponding to each target type. Therefore, the error correction of numerical contents is realized according to types of the numerical contents, which is not only limited to the error correction of the numerical format, but also the logical error correction of the numerical content, so as to improve the numerical error correction capability and thereby improving the recall rate of detection and correction of wrong values.

Abstract: Methods, systems, and computer program products for learning models for entity resolution using active learning are provided herein. A computer-implemented method includes determining a set of data items related to a task associated with structured knowledge base creation, and outputting the set of data items to a user for labeling. Such a method also includes generating, based on a user-labeled version of the set of data items, a candidate model for executing the task, and one or more generalized versions of the candidate model. Additionally, such a method can also include generating a final model based on one or more iterations of analysis of the candidate model and analysis of the one or more generalized versions of the candidate model, and performing the task by executing the final model on one or more datasets.

Abstract: The techniques discussed herein may comprise an autonomous vehicle guidance system that generates a path for controlling an autonomous vehicle based at least in part on a static object map and/or one or more dynamic object maps. The guidance system may identify a path based at least in part on determining set of nodes and a cost map associated with the static and/or dynamic object, among other costs, pruning the set of nodes, and creating further nodes from the remaining nodes until a computational or other limit is reached. The path output by the techniques may be associated with a cheapest node of the sets of nodes that were generated.

Abstract: A data processing system receives a plurality of electronic documents in image format. For each signature segment, the system determines associated surrounding text using an optical character recognition processor. The system accesses first data stored in a signature knowledge base. The system determines first similarity scores based on the associated surrounding text and the first data using a statistical measure technique. The system selects an optimum signature segment based on a distance metric between each signature segment and each associated surrounding text. For each stamp segment, the system extracts text from the stamp segment using an optical character recognition processor. The system accesses second data stored in a stamp knowledge base. The system determines second similarity scores based on the extracted text and the second data using the statistical measure technique. The system selects an optimum stamp segment based on the second similarity scores.

Abstract: The present disclosure relates to systems and methods of overlaying a plurality of isolated collections to generate an overlaid isolated collection. In an example, a first and second isolated collection having at least one common resource may be overlaid. A first and second inference ruleset may be extracted from the first and second isolated collection, respectively. Based on the first and second inference ruleset, one or more suggestions may be generated relating to conflicting inference rules. A suggestion may comprise selecting a subset of the conflicting inference rules (e.g., none, some, or all of the rules) for inclusion in the overlay isolated collection. Another suggestion may comprise generating a new inference rule based on the conflicting inference rules. An indication relating to the suggestions may be received. The indication may be used to generate a third isolated collection and a third inference ruleset.

Abstract: An approach for generating a trained neural network is provided. In an embodiment, a neural network, which can have an input layer, an output layer, and a hidden layer, is created. An initial training of the neural network is performed using a set of labeled data. The boosted neural network resulting from the initial training is applied to unlabeled data to determine whether any of the unlabeled data qualifies as additional labeled data. If it is determined that any of the unlabeled data qualifies as additional labeled data, the boosted neural network is retrained using the additional labeled data. Otherwise, if it is determined that none of the unlabeled data qualifies as additional labeled data, the neural network is updated to change a number of predictor nodes in the neural network.

Abstract: A method includes monitoring performance of a generated model while the generated model is being used for classification on live data, the monitoring including determining a first performance value of the generated model at a first point in time and determining a second performance value of the generated model at a second point in time; rendering, within a graphical user interface, a plot including a first axis and a second axis, the first axis including a characterization of a first performance metric and the second axis including a characterization of a second performance metric; and rendering, within the graphical user interface and the plot, a first graphical object at a first location characterizing the first performance value and a second graphical object at a second location characterizing the second performance value. Related apparatus, systems, techniques and articles are also described.

Abstract: An electronic device is provided. The electronic device includes a memory configured to store a computer executable instructions; and a processor configured to execute the executable instructions to: determine a text corresponding to a received command, provide response information on the command based on a first artificial intelligence model classifying the text as a text corresponding to one of a plurality of pre-stored texts, and provide error information on the command based on the first artificial intelligence model classifying the text as an error, wherein the first artificial intelligence model is configured to classify the text as the error based on the text corresponding to the command being a similar text having one of an entity and an intent different from at least one of the plurality of pre-stored texts.

Abstract: Various embodiments are generally directed to techniques for intuitive machine learning (ML) development and optimization, such as for application in a content services platform (CSP), for instance. Many embodiments include a ML model developer and a ML model evaluator to provide a graphical user interface that guides ML layman in developing, evaluating, implementing, managing, and/or optimizing ML models. Some embodiments are particularly directed to a common interface that provides a step-by-step user experience to develop and implement ML techniques. For example, embodiments may include computing a health score for various aspects of developing and/or optimizing ML models, and using the health score, and the factors contributing thereto, to guide production of a valuable ML model. These and other embodiments are described and claimed.

Abstract: A system and a method for assisting with troubleshooting a complex system is disclosed in which the troubleshooting procedure can be modeled by a Markov decision process. Combining the fault tree technique with a Markov decision process, in order to determine in an optimal manner the sequence of troubleshooting actions will quickly address the consequences of a failure and ensure maintainability of the complex system.

Abstract: Systems and methods are disclosed for monitoring features of a computing device of a distributed computing system using a self-monitoring module. The self-monitoring module can include multiple feature-specific monitoring modules and one or more parent nodes for the feature-specific monitoring modules. A feature-specific monitoring module can identify or detect a fault status change, such as a fault condition or fault resolution, for one or more features. Based on the identified fault conditions or fault resolutions, the feature-specific monitoring module can determine an internal status and communicate an updated status to a parent node.

Abstract: A method and system assists users to generate training sets of labeled data items for machine learning processes. The method and system receives data sets of unlabeled data items from users. The method and system presents data items from the data set for labeling by the users. The method and system analyzes data items that have been labeled by the user and selects future data items to be presented to the user based on analysis of the labeled data items.

Abstract: Embodiments of the present disclosure disclose a text translation method, a text translation apparatus, a device and a storage medium. The method includes: obtaining a source language text; and translating the source language text with a modified translation model to obtain a target language text corresponding to the source language text, the modified translation model being obtained by modifying an original translation model based on a text evaluation result of one or more translated texts for training, the translated text for training being an output result after translating through the original translation model, and the text evaluation result for evaluating a contextual semantic relation in the translated text for training.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for efficiently processing of rule-based computing workflows. In some implementations, a system includes a database, a web server module to provide resources for a rule management user interface having interactive controls to create and edit rules, and an application server module providing services to carry out functions specified through an application programming interface (API). The application server module is configured to (i) receive a request to apply a set of rules to a set of records, and (ii) perform, in response to receiving the request, a set of operations that includes: obtaining the set of rules from the database. The system dynamically translates the rules into executable or interpretable code configured to apply the rules, and invokes the executable or interpretable code to apply the rules to the records in the set of records.

Abstract: The disclosed computer-implemented method for verifying decentralized federated data using influence evaluation may include (i) calculate an influence score for each of a group of data instances, (ii) rank the data instances based on the influence scores, (iii) determine an anomaly score for each of the ranked data instances, (iv) select the ranked data instances with the highest anomaly scores as containing potentially malicious data, and (v) perform a security action that protects against the potentially malicious data. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A mechanism is provided to implement a text classifier training augmentation mechanism for incorporating unlabeled data into the generation of a text classifier. For each term of a plurality of terms in each document of a plurality of documents in a set of unlabeled data, a term frequency value is determined. The term is normalized by dividing the term frequency value by a total number of terms in the document. An inverse document frequency (idf) value is determined for each term based on the term frequency value. A subset of terms is filtered from the plurality of terms based the determined idf values. The idf values for the remaining terms are transformed into feature weights. Terms from a set of labeled data are re-weighted based on the feature weights determined from the set of unlabeled data. The text classifier is then generated using the re-weighted labeled data.

Abstract: A method and system for graphical enumeration. The method includes creating an ordered set of vertices for a graph such that each vertex is associated with a corresponding index, and wherein each vertex in the ordered set of vertices includes information. A plurality of keys is created for defining the information. A plurality of lists of vertices is created, each of which is associated with a corresponding key such that vertices in a corresponding list include information associated with the corresponding key. For a first list of vertices, a least valued index is determined from a group of associated vertices based on vertices in the first list and vertices pointed to by the vertices in the first list. Also, all associated vertices are pointed to a root vertex associated with the least valued index.

Abstract: Described are methods, systems and computer readable media for GUI control elements and associated processing methods.

Abstract: An autonomous vehicle system includes one or more output devices, one or more input sources, and a controller. In some embodiments, a verified inference engine is used by the controller to generate outputs for the output devices from inputs received from the input sources. The inference engine may be verified to be mathematically correct with an automated theorem proving tool. The automated theorem proving tool may verify that the inference engine meets the design requirements of standards such as DO-178C Level A and/or EAL-7. The controller is configured to validate the inputs received from the input sources, store the validated inputs in a fact base, generate outputs from the validated inputs, validate the one or more outputs, and provide the one or more validated outputs to the output devices.

Abstract: Disclosed in some examples, are methods, systems, and machine readable mediums which provide an event processing system and indicator engine that provide indicators to one or more of the channel systems that pertain to a status of an issue experienced by a customer. The indicators are created in part based upon updates to the issue entered into a customer issue system and when displayed in a channel system, the indicators provide an indication that there is an update on the particular issue affecting the customer.

Abstract: A computer-implemented method for generating a customized tax preparation experience for a user of tax preparation software includes a computing device executing a data capture utility, the data capture utility connecting to one or more remote data sources and capturing and transferring at least some tax data contained therein to a data store associated with the tax preparation software. The computing device executes a topic analysis engine configured to identify highly relevant tax topics from the data transferred to the data store. A user interface manager automatically generates interview questions or statements on said highly relevant tax topics. The user confirms whether other tax topics apply to the user. The computing device executes a tax calculation engine of the tax preparation software configured to compute a tax liability or refund amount.

Abstract: Systems and methods are disclosed for monitoring features of a computing device of a distributed computing system using a self-monitoring module. The self-monitoring module can include multiple feature-specific monitoring modules and one or more parent nodes for the feature-specific monitoring modules. A feature-specific monitoring module can identify or detect a fault status change, such as a fault condition or fault resolution, for one or more features. Based on the identified fault conditions or fault resolutions, the feature-specific monitoring module can determine an internal status and communicate an updated status to a parent node.

Abstract: Molecular structure generation with substructure representations is provided. The computer-implemented method, comprises: selecting, by a device operatively coupled to a processor, a vertex in an existing graph of a molecular structure; replacing, by the device, the selected vertex with a sub-structure representation of a molecule, the sub-structure representation including two or more vertices; connecting, by the device, a new vertex to the sub-structure representation in the existing graph to generate a new graph; labeling, by the device, each of a plurality of vertices of the new graph by a canonical labeling method; and determining, by the device, whether a vertex that has a minimum label is the new vertex.

Abstract: A knowledge graph is divided into a plurality of sub-graphs, each sub-graph comprising a plurality of vertices and a plurality of edges. The knowledge graph is represented as a summary graph comprising for each of the sub-graphs a summary-graph vertex. A local sub-graph is generated as a copy of one of the sub-graphs together with a copy of a surrounding graph to the one of the sub-graphs. The content of the local sub-graph is modified. The local sub-graph is reintegrated, upon a reintegration trigger event, back into the knowledge graph, wherein a structure of the surrounding graph is used as a reintegration aid, by overlaying the structure and the knowledge graph, thereby identifying identical vertices of the surrounding structure and the knowledge graph as anchor points from where changes in the local sub-graph are reintegrated into the knowledge graph.

Abstract: An example system includes a processor to receive a plurality of object aspects of an object to be evaluated using a process, a structure of the process, a plurality of extracted facts from documents, a tree related to the plurality of object aspects and the structure, and a thesis for each leaf in the tree. The processor is also to relate the extracted facts to the theses in the tree. The processor is to generate a score for each leaf corresponding to a fact in the tree. The processor is to generate a thesis score and a thesis summary for each thesis based on the scores and the summaries of related facts for each thesis. The processor is to further generate a final score for the object based on the thesis scores.

Abstract: A knowledge base can include a dictionary associated with classes of a model, e.g., an ontology. A text segment that is not found in the dictionary can be received. Feature(s) can be determined for the text segment and, based partly on providing the feature(s) to a classifier, a set of values can be determined. The distribution can include values respectively corresponding to the classes. One of the values can be greater than a predetermined threshold. That value can correspond to a class. An indication identifying the class can be presented via a user interface having functionality to provide input that the text segment is associated with the class, is not associated with the class, or is associated with another class. Based at least partly on adding a new class to the ontology, a precedence table indicating priorities between motifs defining relationships between classes of the ontology can be updated.

Abstract: A system includes an interface and one or more processors. The interface receives a dataset comprising a plurality of variable length input records, each input record comprising a plurality of fields. The one or more processors compare the input record to a plurality of predetermined record types to determine whether the input record matches one or more of the predetermined record types. Upon a determination that the input record matches one or more of the predetermined record types, the one or more processors determine one or more rules applicable to the input record. The one or more rules are determined based on the predetermined record types that match the input record. The one or more processors apply the one or more rules applicable the input record. The one or more rules determine the quality of the input record based on a structure of one or more of the fields of the input record and/or a value of one or more of the fields of the input record.

Abstract: A computer-implemented method for managing computer security of client computing machines may include (i) monitoring a set of client computing devices, (ii) receiving security data on sets of security-related events from each client computing device in the set of client computing devices, (iii) clustering the sets of security-related events by calculating a dissimilarity value, for each set of security-related events, that indicates a uniqueness of the set of security-related events in relation to other sets of security-related events using a dissimilarity function and adjusting the dissimilarity function based on a homogeneity of clusters of sets of security-related events, (iv) determining, based on clustering the sets of security-related events by the dissimilarity value, that a set of security-related events comprises an anomaly, and (v) performing a security action in response to determining that the set of security-related events comprises the anomaly.

Abstract: A present invention embodiment adjusts application of a set of data quality rules for data within one or more reports. A first report field is selected that receives data from a same source data field as at least one second report field within the one or more reports. An active data quality rule associated with the at least one second report field is identified that is inactive for the first report field. An effectiveness of the identified data quality rule is determined for data quality of the first report field. Application of a set of data quality rules is adjusted for the first report field to include the identified data quality rule in response to the effectiveness satisfying a threshold.

Abstract: The example embodiments are directed systems and methods for managing a contextual digital twin. In one example, the method may include receiving a request message configured to trigger an action to a digital twin, the digital twin comprising a virtual representation of an asset, determining a behavior of the digital twin that is linked to the request message based on capabilities of the behavior stored with respect to the digital twin, determining input parameters of the behavior from the request message, and generating an executable script for performing the determined behavior in association with the digital twin based on the input parameters. The context can provide information about operations of the asset as well as insight into how to handle the operations based on similar occurrences to the asset or other related assets.

Abstract: A method, system, and apparatus are provided for resolving conflicts between training data conflicts by retrieving independent training data sets, each comprising a plurality of intents and end-user utterances for use in training one or more classifiers to recognize a corresponding intent from one or more of the end-user utterances, providing a first test end-user utterance associated with a first intent from the first independent training data set to the one or more classifiers to select an output intent generated by the one or more classifiers; identifying a first conflict when the first intent does not match the output intent, and automatically generating, by the system, one or more conflict resolution recommendations for display and selection by an end user to resolve the first conflict.

Abstract: In one embodiment, a node in a network detects an anomaly in the network based on a result of a machine learning-based anomaly detector analyzing network traffic. The node determines a packet capture policy for the anomaly by applying a machine learning-based classifier to the result of the anomaly detector. The node selects a set of packets from the analyzed traffic based on the packet capture policy. The node stores the selected set of packets for the detected anomaly.

Abstract: This disclosure involves generating and outputting a segmentation model using 3D models having user-provided labels and scene graphs. For example, a system uses a neural network learned from the user-provided labels to transform feature vectors, which represent component shapes of the 3D models, into transformed feature vectors identifying points in a feature space. The system identifies component-shape groups from clusters of the points in the feature space. The system determines, from the scene graphs, parent-child relationships for the component-shape groups. The system generates a segmentation hierarchy with nodes corresponding to the component-shape groups and links corresponding to the parent-child relationships. The system trains a point classifier to assign feature points, which are sampled from an input 3D shape, to nodes of the segmentation hierarchy, and thereby segment the input 3D shape into component shapes.

Abstract: Techniques and solutions are described for detecting and correcting errors in software, such as in source code, using machine learning. A machine learning algorithm can be trained using source code, source code corrections, and extracted code features to provide a trained classifier. Test data, such as code to be scanned for errors or code known to have an error, can be submitted to the trained classifier. The classifier can generate a code correction or corrected code. Code corrections can be applied to the source code to provide corrected code.

Abstract: A method for dynamically enforcing a code usage policy including mapping machine code compiled from source code to identify construct interactions in the machine code, determining whether the construct interactions identified by the mapping conform to one or more construct interaction maps in a construct interaction registry associated with a code usage policy, and generating instructions to render one or more error messages within a graphical user interface of a development environment in response to a determination that one or more of the construct interactions in the machine code fails to comply with at least one of the construct interaction maps in the construct interaction registry.

Abstract: A control logic analyzer for controlling a plurality of devices is provided. The control logic analyzer comprises: a control logic decomposer configured to analyze control logics from different sources to identify devices involved in the control logics from the plurality of devices, and decompose the control logics into control instructions to be executed by the identified devices; and a potential conflict searcher configured to search a database for storing decomposed control logics and determine whether there is any potential conflict between the current control logic and the control logics previously stored in the database.