Abstract: Methods and systems are provided for sequencing of input prompts for data structure completion. The method includes: providing a data structure including object classes with typed attributes and rules defining constraints to be met by a data instance of a data structure to meet a regulation body and building a data instance of the data structure by prompting user input via a user interface. The method includes: selecting a working object instance from the top of a stack of object instances; gathering applicable rules for the working object instance with missing required facts in the form of object attributes, including any related rules where a new object class is required; and sequencing input prompts in a user interface of a complex form to obtain the required facts in the form of missing object attributes.

Abstract: Systems and methods for building a response for a machine learning-based dialogue agent includes implementing machine learning classifiers that predict slot segments of the utterance data based on an input of the utterance data; predict a slot classification label for each of the slot segments of the utterance data; computing a semantic vector value for each of the slot segments of the utterance data; assessing the semantic vector value of the slot segments of the utterance data against a multi-dimensional vector space of structured categories of dialogue, wherein the assessment includes: for each of a distinct structured categories of dialogue computing a similarity metric value; selecting one structured category of dialogue from the distinct structured categories of dialogue based on the computed similarity metric value for each of distinct structured categories; and producing a response to the utterance data.

Abstract: An optimization device is applied to a production line in which multiple production devices respectively execute a production process. The optimization device includes a cycle time setting section configured to set a cycle time for each production device, a target time setting section configured to set a target time based on a cycle time and a time required for periodic operation in each of multiple production devices except for the component mounter, and a line design section configured to calculate the unit number of the component mounters constituting the production line based on the target time.

Abstract: A non-transitory computer-readable medium stores a program designed to be used for a patient attempting to quit smoking. The program causes a computer to execute: receiving report information input by the patient and indicative of the presence or absence of smoking by the patient; receiving a biomarker concentration measurement value measured from a biological sample of the patient by a biomarker concentration meter for measuring a biomarker concentration indicative of a smoking state; based on the received biomarker concentration measurement value and a biomarker concentration reference value, determining consistency between the biomarker concentration measurement value and the report information; based on a result of the consistency determination, generating smoking cessation therapy information for smoking cessation therapy to be performed for the patient; and transmitting the generated smoking cessation therapy information to a patient-side device.

Abstract: Computer systems and methods are provided for increasing a rate of simulation for quantum computing devices. A quantum computing device includes a plurality of gates, each of which is coupled to one or more modes. In the provided computer systems and methods, a device definition and state information for the quantum computing device are received. The state information includes a plurality of input patterns, each of which indicates a number of input bosons that correspond to a respective mode of the quantum computing device, and an amplitude that corresponds to each input pattern. The device definition includes a plurality of sets of gate values that indicate modification by a respective gate of an input pattern probability. A first group of input patterns is generated for a first gate. The first group of input patterns includes a subset of the plurality of input patterns that meet grouping criteria.

Abstract: Embodiments include a system for reconciling input datasets with a model ontology, the system comprising a processor coupled to a memory, to execute: a model ontology storage process, storing a model ontology graph defining a concept hierarchy, the concept hierarchy comprising nodes at various levels. The processor and memory further to execute: a user interface process, receiving, via a user interface, user input assignments of nodes from upper levels of the concept hierarchy to input datasets with particular properties from among the lower level nodes. The processor and memory further to execute an input dataset reconciliation process using the concept hierarchy and a record of historical user input assignments to determine which higher level nodes to assign to a new input dataset, based on the lower level nodes represented in the new input dataset.

Abstract: An apparatus comprising a processor and memory including computer program code, the memory and computer program code configured to, with the processor, enable the apparatus at least to: identify, for use in route navigation, a complete set of lane traversals for a road intersection from two or more different complete sets of lane traversals for the road intersection based on respective safety-traffic flow scores assigned according to one or more predefined safety-traffic flow criteria, wherein each lane traversal defines a path of travel from an inbound lane of the road intersection to an outbound lane of the road intersection, and wherein each different complete set of lane traversals comprises a different combination of lane traversals for the road intersection.

Abstract: Disclosed herein is method and system for processing multimodal user queries. The method comprises determining availability of one or more responses to each of one or more sub-queries, wherein the one or more sub-queries are formed by splitting the multimodal user queries. The method detects requirement of an expert to provide the one or more responses upon determining at least one of unavailability of the one or more responses by the response generation system or based predefined conditions. Thereafter, a summarized content is generated by summarizing context of the one or more sub-queries and historical conversation data associated with the one or more sub-queries. Based on the summarized content, the one or more sub-queries are reformulated. Finally, the one or more responses received, from the expert, for the reformulated one or more sub-queries are collated provided as the one or more responses for the multimodal user queries.

Abstract: Methods and systems are disclosed for determining gaps in a user's resume regarding skills or titles for a job, and recommending educational courses to the user based on those gaps. The user's resume is received and analyzed for the user's current set of skills. Various searching methods are used to identify a set of job requisitions based on a user's input query for a job. The user's skills are compared with a set of required skills associated with the set of job requisitions. Educational courses are identified for the skills that are required for the set of job requisitions but are not present in the user's set of skills. The educational courses presented to the user via the search engine. The user's resume can also be ranked relative to other resumes for the skills or titles required for a job.

Abstract: Provided are voiceprint recognition method, device, terminal apparatus and storage medium.

Abstract: A learning management device comprising an inference engine that is input with image data and acquires output for performing guidance display to a user using an inference model that has been specifically learned, and a processor that confirms if there is a specified relationship between a trial input image and trial guidance display for this trial input image under conditions that have been assumed at the time of use, and that determines whether or not it is necessary to relearn the inference model that has been learned based on the result of this confirmation.

Abstract: A system for activity localization in videos is described, comprising a visual concept detection module, which produces a plurality of first visual concept vectors each representing a probability of containing visual concepts for one of a plurality of sampled frames sampled from an input video; wherein each of the plurality of first visual concept vectors dot-product with a second visual concept vector extracted from a given query sentence, resulting a visual-semantic correlation score; a semantic activity proposal generation module, which generates semantic activity proposals by temporally grouping frames with a high visual-semantic correlation score; and a proposal evaluation and refinement module, which takes the semantic activity proposals, the visual concept vectors and the query sentence as input, and outputs alignment scores and refined boundaries for the proposals. The disclosure also relates to methods thereof.

Abstract: An apparatus, termed a destination validation device (DVD), comprises a sensor configured to sense physical presence of a parcel and to obtain metadata from said parcel; at least one data processor connected with at least one memory storing software instructions that when executed causes the DVD to at least determine from reading the obtained metadata whether a signature is required for delivery of the parcel, and prepare a destination delivery acknowledgement indicating whether said signature is required; and a transmitter configured to communicate the prepared destination delivery acknowledgement external of the DVD. In this manner a delivery service can verify that the parcel is at the physical delivery location. Further details provide for also obtaining the (digital) signature when the metadata indicates it is required for delivery of the parcel. Other embodiments include a method and a computer program product comprised of software instructions on a computer-readable medium.

Abstract: A report cost estimation module is used with cloud storage systems that store and process client data. Reports on the cloud stored client data take a significant amount of CPU, memory, storage and networking resources. A data analysis model dynamically estimates pricing for client data reports according to resource consumption by identifying a cluster group for the report and using the designated regression model for the cluster group. The data analysis can store the estimated and actual report costs and improve the estimated report costs using machine learning algorithms and crowed sourcing techniques. The report price is accurately provided to the client before running the report and thus allowing the customer to carefully manage a client report budget.

Abstract: The performance of a neural network (NN) and/or deep neural network (DNN) can limited by the number of operations being performed as well as management of data among the various memory components of the NN/DNN. Using virtualized hardware iterators, data for processing by the NN/DNN can be traversed and configured to optimize the number of operations as well as memory utilization to enhance the overall performance of a NN/DNN. Operatively, an iterator controller can generate instructions for execution by the NN/DNN representative of one more desired iterator operation types and to perform one or more iterator operations. Data can be iterated according to a selected iterator operation and communicated to one or more neuron processors of the NN/DD for processing and output to a destination memory. The iterator operations can be applied to various volumes of data (e.g., blobs) in parallel or multiple slices of the same volume.

Abstract: Systems and methods for controlling an operation of devices for an occupant. A processor to iteratively train a personalized thermal comfort model (PTCM) during an initialization period. Receive a sequence of unlabeled real-time data. A transmitter requests the occupant to label an instance of unlabeled data, when there is a disagreement between the labels of stored historical labeled data (LD) similar to received unlabeled data and a predicted label on the new unlabeled data that exceeds a threshold. The processor, in response to receiving the labeled data, trains the PTCM using different weights of the personalized LD than to the historical LD. Retrains PTCM using the historical database and the updated personalized database. A controller controls the set of devices based on the retrained PTCM.

Abstract: Various aspects of the subject technology relate to systems, methods, and machine-readable media for relational image querying. A system may receive a search query for content from a client device, where the query specifies one or more objects and one or more spatial relationships between the one or more objects. The system may generate a query vector for the query using a computer-operated neural language model. The system may compare the query vector to an indexed vector for each of the one or more spatial relationships between the one or more objects of an image. The system may determine a listing of relational images from a collection of images based on the comparison. The system may determine a ranking for each image in the listing of relational images, and provide search results responsive to the search query to the client device, which may include a prioritized listing of the relational images based on the determined ranking.

Abstract: A computer-implemented method assesses operation of a financial computing system (FCS). An assessment computer system generates code for a model of the FCS that comprises a model specification for the FCS and a model environment for the FCS. The code for the model uses a type-system based logical programming language that supports typed recursive functions. The assessment computer system generates mathematical axioms that describe the operation of the FCS by compiling the code for the model and assesses the operation of the financial computer system by analyzing the mathematical axioms.

Abstract: A system that includes a first device and a second device. The first device is configured to send correlithm objects having a first bit having a first bit string length and to send a test correlithm object having the first bit string length to the second device. The second device is configured to receive the test correlithm object, to determine a distance between the test correlithm object and a reference correlithm object, and to send the switch command to the first device in response to determining the distance between the test correlithm object and the reference correlithm object exceeds a distance threshold value. The first device is further configured to receive a switch command, and to send correlithm objects having a second bit string length that is greater than the first bit string length to the second device in response to receiving the switch command.

Abstract: A device configured to emulate image mapping in a correlithm object processing system comprises a memory and one or more processors. The memory stores a correlithm object mapping table configured with multiple source image elements and multiple corresponding target correlithm objects. The processors receive an input image element comprising an n-pixel array of binary values and determine n-dimensional distances between the input image element and each of the source image elements. The processors then identify a source image element with the closest determined n-dimensional distance. The processors determine whether a deviation between the input image element and the identified source image element is within a predetermined tolerance. The processors identify a target correlithm object corresponding to the identified source image and determine a perturbation to be applied to the identified target correlithm object based on the determined deviation.

Abstract: A big data analysis system may include a big data repository communicatively coupled to a data accumulation server and a predictive graph processing system. The data accumulation server may be configured to receive information from a plurality of data sources, the information corresponding to user interaction with one or more computing devices associated with an organization via a networked computing system, store the information received from the plurality of sources in the big data repository; and monitor the plurality of data sources to update the data stored in the big data repository. The predictive graph processing system is configured to receive information stored in the big data repository, transform the information received from the big data repository into a predictive graph data set based on a predictive model, and store the predictive graph data set to a visualization data repository.

Abstract: Methods, systems, and computer program products for tracking missing data using provenance traces and data simulation are provided herein. A computer-implemented method includes generating, for each of multiple stages in a data curation sequence, a machine learning model of the data curation sequence, wherein the model is based on historical input records within the data curation sequence, historical output records within the data curation sequence, and provenance data within the data curation sequence; creating a simulated output record based on a detected anomaly corresponding to the data curation sequence; predicting the content of absent input records that precede the simulated output record in the data curation sequence and provenance data corresponding to the simulated output record; and outputting, to a user, in response to a query pertaining to the detected anomaly, the predicted input records and information relating the predicted input records to the detected anomaly.

Abstract: An evaluation platform receives a data set and a description of an outcome, such as predicting results of trends, recognizing patterns, and evaluating options according to specified criteria. The description is evaluated to select candidate evaluators that may be capable of achieving the outcome, and to translate the outcome into a goal for each selected candidate evaluator. The evaluator candidate set is trained using a training data set, and an initial evaluator is selected that exhibits the highest performance to achieve the outcome over the data set. The initial evaluator is applied to achieve the requested outcome over the data set. Optionally, the performance of the initial evaluator may be monitored to detect performance drift. In this event, the evaluator candidate set is reevaluated to identify a substitute evaluator exhibiting higher performance than the initial evaluator, which replaces the initial evaluator in the continued evaluation of the data set.

Abstract: Software that selects portions of unlabeled text for labeling, by performing the following operations: (i) receiving a set of unlabeled input text for classification with respect to a particular domain, wherein the domain includes a labeled corpus for which topics of a set of topics correspond to labels from the corpus, and wherein the topics include statistical probability distributions of words in the corpus; (ii) performing topic modeling on the input text to associate portions of the input text with respective classifications, wherein the classifications include statistical probability distributions of topics of the set of topics in the respective portions of the input text; and (iii) applying a machine learning-based selection strategy to the portions of the input text and their respective classifications to identify one or more portions of the input text for labeling.

Abstract: A method and system of inferential-based communications applies two sets of values, whereby the first set of values are determined from inferences of the relevancy of textual elements with respect to text-based content and the second set of values are determined from behavioral-based inferences with respect to topics, so as to select appropriate words to be included in communications that are generated for delivery to a user. The inferences that are with respect to text-based content may be determined by applying analytic methods such as Bayesian or statistical learning-based methods. The values that are determined from behavioral-based inferences with respect to topics may correspond to inferences of interest and/or expertise. Words in the communications may be selected and/or arranged in accordance with syntactical rules and may reference elements of the text-based content and/or behavioral inferences.

Abstract: An algorithm data store may contain information about a pool of available algorithms. An algorithm analysis engine may compare the information about each of the pool of available algorithms with at least one requirement associated with the current algorithm executing in the live environment. Based on the comparison, the algorithm analysis engine may select at least one of the pool of available algorithms as a potential replacement algorithm and transmit an indication of the selected at least one potential replacement algorithm. A deployment platform may include a lifecycle manager that manages execution of the current algorithm in the live environment. The lifecycle manager may also receive the indication of the selected at least one potential replacement algorithm, manage execution of the at least one potential replacement algorithm in a shadow environment, and report performance information associated with the current algorithm and the at least one potential replacement algorithm.

Abstract: A method and device for checking false sharing in deletion of a data block are disclosed. The method includes setting weight bits and a weight reset bit in a mapping pointer (MP) pointing to the data block, wherein the weight reset bit is configured to define whether a weight indicated by the weight bits has been increased; and determining, based on the weight bits and the weight reset bit, whether false sharing occurs. The method determines whether false sharing occurs accurately and with low cost by using a computable formula. All information for such computable formula comes from in-memory MP, with no need to load any extra metadata, for example, with no need to access Block Meta Data. The method also considers a scenario where a weight increase occurs during a weight split and merge with other pointers.

Abstract: A system and method for improving a machine learning-based dialogue system includes: sourcing a corpus of raw machine learning training data from sources of training data based on a plurality of seed training samples, wherein the corpus of raw machine learning training data comprises a plurality of distinct instances of training data; generating a vector representation for each distinct instance of training data; identifying statistical characteristics of the corpus of raw machine learning training data based on a mapping of the vector representation for each distinct instance of training data; identifying anomalous instances of the plurality of distinct instances of training data of the corpus of raw machine learning training data based on the identified statistical characteristics of the corpus; and curating the corpus of raw machine learning training data based on each of the instances of training data identified as anomalous instances.

Abstract: A machine learning method includes installing a plurality of model replicas for training on a plurality of computer learning nodes; receiving training data at a each model replica and updating parameters for the model replica after trailing; sending the parameters to other model replicas with a communication batch size; evaluating received parameters from other model replicas; and dynamically adjusting the communication batch size to balance computation and communication overhead and ensuring convergence even with a mismatch in processing abilities on different computer learning nodes.

Abstract: Systems and method are provided for controlling a vehicle.

Abstract: Methods, apparatuses, and systems for applying models to score insurance data are disclosed. In one aspect, a system comprising a master node, a plurality of nodes in at least one cluster connected to the master node, and one or more computing devices connected to the master node is disclosed, where the master node is configured to distribute, using a HIVE module, an insurance scoring script and a predictive model to each of the plurality of nodes. The master node may call a function of the HIVE module to instruct each of the plurality of nodes to execute the insurance scoring script to generate scored results, wherein the scored results are written into a results table, and wherein the scored results comprise insurance scores for a plurality of customers.

Abstract: Disclosed are examples for adjusting screen brightness based on screen content being presented on a display screen of a mobile device. The described examples may determine a time at which the screen content is to be evaluated. The screen content is categorized based on the evaluation. A category of the screen content may be input into a machine learning algorithm that may be used to determine whether a screen brightness adjustment is appropriate. If a screen brightness adjustment is appropriate, a degree of the screen brightness adjustment may be determined.

Abstract: The stockpiled number of maintenance components is more appropriately managed. A maintenance component management device (15) includes: an acquisition means (151) for acquiring, from a plurality of maintenance target devices (25) requiring maintenance using a maintenance component, information representing an operating state of each of the plurality of maintenance target devices (25); and a calculation means (152) for extracting information representing a factor of a main cause for performing maintenance using the maintenance component from the information representing the operating state thus acquired, and calculating a number of the maintenance components estimated as becoming necessary in order to maintain the plurality of maintenance target devices (25), based on the information representing the factor thus extracted.

Abstract: System and methods for generating a trajectory of a dynamical system are described herein. An example method can include generating a desired effect on the dynamical system in response to a plurality of system states and effects, where the desired effect is associated with a confidence value. The method can also include predicting a respective corresponding effect on the dynamical system in response to each of a plurality of possible system actions, where each respective corresponding effect is associated with a confidence value. The method can further include comparing each respective corresponding effect on the dynamical system to the desired effect on the dynamical system, and selecting a next system action in the trajectory of the dynamical system based on the comparison.

Abstract: A method of using a virtual machine operator model includes providing a virtual machine operating environment having a current state and including a virtual machine and a virtual operator agent acting within the virtual machine operating environment. The method also includes executing a control action, by the virtual operator agent, relative to the virtual machine based on the virtual machine operator model. The method also includes analyzing a new state of the virtual machine operating environment resulting from execution of the control action, assigning a positive reward or a negative reward to the control action based on the new state, assigning the negative reward to the control action resulting in a high implement jerk movement of the virtual machine, and executing a learning algorithm to update the virtual machine operator model based on the positive reward or the negative reward.

Abstract: A consistent meaning framework (CMF) graph including a plurality of nodes linked by a plurality of edges is maintained in data storage of a data processing system. Multiple nodes among the plurality of nodes are meaning nodes corresponding to different word meanings for a common word spelling of a natural language. Each of the multiple word meanings has a respective one of a plurality of associated constraints. A natural language communication is processed by reference to the CMF graph. The processing includes selecting, for a word in the natural language communication, a selected word meaning from among the multiple word meanings based on which of the plurality of associated constraints is satisfied by the natural language communication. An indication of the selected word meaning is stored in data storage.

Abstract: According to embodiments illustrated herein, a method and a system is provided for screening candidates for job opportunities. The method includes grouping the candidates into batches based on predetermined time duration, a count of the candidates, and a chronology of receiving job applications from the candidates. Each batch comprises a first set of candidates. Thereafter, a sliding window is moved over the batches, to encompass a set of batches at a first time instance. A second set of candidates is identified from the first set of candidates in a batch from the set of batches based on a score assigned to each of the first set of candidates during an interview. Further, a candidate is selected from the second set of candidates obtained from a first batch, encompassed by the sliding window at a second time instance before the first time instance.

Abstract: A method for partitioning a plurality of entities each associated with a plurality of ordered sequences of events received by a computer system, the method including: defining a minimal directed acyclic graph data structure representing the sequences of events to define a plurality of categories of behavior of the entities; defining a threshold degree of similarity as an xmu number, the xmu number having cardinality that is able to vary across a normalized range; defining a relation for each entity including a degree of association of the entity with each of the categories; defining a cluster of entities as a set of entities comprising a first entity; comparing a relation for the first entity with a relation for a second entity to define a xmu Jaccard similarity coefficient for the first and second entities; and responsive to the coefficient meeting the threshold degree of similarity, adding the second entity to the cluster.

Abstract: In certain embodiments, a computer-implemented method includes determining, by a processor, characteristics of a new transaction and processing, by the processor, the characteristics of the new transaction using a risk assessment model. The risk assessment model is generated using an iterative training process that processes, over time, historical transactions to train the risk assessment model based on characteristics of historical transactions and associated risk assessment analysis of the historical transactions. The method includes generating, by the processor, a risk assessment indicator for the new transaction based on processing the characteristics of the new transaction using the risk assessment model. The risk assessment indicator indicates a degree of risk of the new transaction.

Abstract: A method, a processing device, and a computer program product are provided. At least one processing device correlates textual medical information related to the subject with characteristics of an image of a medical condition of the subject to generate a subject signature. The at least one processing device compares the subject signature with multiple reference signatures to determine at least one reference signature corresponding to the subject signature. Each reference signature is associated with a corresponding medical condition and is generated by correlating textual medical information regarding the corresponding medical condition with characteristics of an image of the corresponding medical condition. The at least one processing device identifies the medical condition of the subject based on the medical conditions associated with the determined at least one reference signature. Information is provided regarding the identified medical condition of the subject.

Abstract: A computer-implementable method for managing a cognitive graph comprising: receiving data from a plurality of data sources; processing the data from the plurality of data sources, the processing the data from the plurality of data sources identifying a plurality of knowledge elements; and, storing the knowledge elements within the cognitive graph as a collection of knowledge elements, the storing universally representing knowledge obtained from the data, each knowledge element being structurally defined within the cognitive graph.

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

Abstract: A computer-implementable method for managing a cognitive graph comprising: receiving data a data source, the data comprising a query and information relating to an answer to the query; processing the query, the processing the query identifying a plurality of query related knowledge elements; processing the information relating to the answer to the query, the processing the information relating to the answer to the query identifying a plurality of answer related knowledge elements; and, storing the plurality of query related knowledge elements and the plurality of answer related knowledge elements within the cognitive graph as a collection of knowledge elements.

Abstract: According to some embodiments, system and methods are provided, comprising an installed product, including a plurality of components; a computer programmed with a damage metric model for the installed product, the damage metric model for providing an estimate of an extent of damage on one or more components; the computer programmed with a dynamic process control model for providing a dynamic response of the installed product with respect to its one or more operating parameters; the computer further programmed with a true-up model for providing a control action to reduce an uncertainty of the estimate provided by the damage metric model; the computer including a processor and a memory in communication with the processor, the memory storing the damage metric model and the true-up model; the memory storing additional program instructions, the processor operative with the additional program instructions to perform functions as follows: receiving an estimate output of the damage metric model, wherein the output in

Abstract: A storage unit 81 stores information associating each of a plurality of prediction targets with a predictor-related index related to a predictor for predicting the prediction target. Scatter graph generation means 82 generates, based on the information stored in the storage unit 81, a scatter graph in which a symbol representing the prediction target of the predictor is located at a position determined by the predictor-related index in a coordinate space where the predictor-related index is defined as at least one dimension.

Abstract: Techniques for tenant performance isolation in a multiple-tenant database management system are described. These techniques may include providing a reservation of server resources. The server resources reservation may include a reservation of a central processing unit (CPU), a reservation of Input/Ouput throughput, and/or a reservation of buffer pool memory or working memory. The techniques may also include a metering mechanism that determines whether the resource reservation is satisfied. The metering mechanism may be independent of an actual resource allocation mechanism associated with the server resource reservation.

Abstract: Systems, methods, and non-transitory computer readable media are configured to select a first content item and a second content item to display to each user of a plurality of users in pairwise comparison in a survey associated with a content feed of the user. A selection by each user of the plurality of users of one of the first content item and the second content item can be obtained via a user interface. A ranking model configured to assign a score indicative of quality of each content item of a plurality of content items associated with content feeds based at least in part on the selections by the plurality of users can be determined.

Abstract: A method for providing priority notification analysis on a mobile communications device, includes receiving a notification associated with a software application accessible by the mobile communications device and continues with the mobile device transmitting the notification to one or more computing devices that includes one or more modules configured to classify natural language. The method continues with the mobile communications device receiving a priority score associated with the notification that is generated by the one or more computing devices that includes one or more modules configured to classify natural language and in response to determining that the priority score compares favorably to a threshold priority score, rendering the notification for display by the mobile communications device to indicate that the notification meets the threshold priority score.

Abstract: A sequence of events by a single user with at least one computing system are monitored. Each event characterizes user interaction with the at least one computing system and the sequence of events form a plurality of pairwise disjoint log samples. Thereafter, it is determined, using an adjacency graph trained using a plurality of log samples generated by a plurality of users, whether any of the log samples is anomalous. Data can be provided that characterizes the log samples determined to be anomalous. Related apparatus, systems, techniques and articles are also described.

Abstract: A computer implemented method which accesses multiple sets of narrative data, each set of narrative data comprising event records mapped to one of a plurality of story rules to create a plurality of story event sequences based on the event records and the story rules. The method compares a first story event sequence and a second story event sequence. The comparison comprises a method for determining a taxonomical distance between elements of narrative data in one set of narrative data to a corresponding element of narrative data in another set of narrative data; and performing an optimal matching between events in two sets of narrative data. The method outputs a similarity result between the sets of narrative data.