Abstract: A new scalable approach to conflict-free deployment of changes across networks. The conflict rules or constraints may be modeled using policies and algorithms to determine an optimized schedule for change deployment.

Abstract: Gradient Boosting Decision Tree (GBDT) successively stacks many decision trees which at each step try to fix the residual errors from the previous steps. The final score produced by the GBDT is the sum of the individual scores obtained by the decision trees for an input vector. Overfitting in GBDT can be reduced by removing the input values that have the least impact on the output from the training data. One way to determine which input variable has the lowest predictive value is to determine the input variable that is used for the first time in the latest decision tree in the GBDT. This method of identifying the low-predictive features to be removed does not require that earlier trees be regenerated to generate the new GBDT. Since the removed feature was already not used in the earlier trees, those trees already ignore the removed feature.

Abstract: In a computer system, an orchestration platform includes extensible components that interact with external systems and technology. The platform is secured by means of architectural features, encryption, and access control.

Abstract: A stream of tuples is processed by a stream application. The stream application includes a plurality of processing elements that operate on one or more compute nodes, each processing element includes one or more stream operators. One or more databases that are capable of communicating with the stream application are monitored during the processing of the stream of tuples. A potential performance condition of a first database of the one or more databases is detected based on the monitoring. An output adjustment is performed, in response to the potential performance condition.

Abstract: One example method includes deploying an application in a distributed computing environment. Telemetry data is collected that corresponds with the deployment of an application. The telemetry data is received by a machine learning model that was trained with test telemetry data to determine whether the deploying is successful or failed. A successful inference results in continued deployment and a fail inference results in a rollback of the application.

Abstract: A memory stores telemetry data for an information handling system. A processor analyzes applications being executed within the information handling system, and determines one or more of the applications that collect and send the telemetry data. The processor also retrieves a first list of different sets of data to be tracked for each of the applications. Each of the sets of data to be tracked is associated with a different one of the applications. The processor creates a second list of data being logged in the information handling system. Based on the first and second lists, the processor creates a comprehensive list of data tracked by the applications.

Abstract: A system for auto-determining solutions for dynamitic issues comprises a processor associated with a server. The processor detects an application issue associated with an application running at a network node in a distributed network. The processor receives a set of data objects associated with the application issue. The processor classifies the set of the data objects of the application issue into one or more issue patterns using a machine learning model. The machine learning model is trained based on a plurality of sets of data objects and issue patterns associated with corresponding previous application issues. The processor processes the one or more issue patterns and application information through a neural network to determine a series of executable operations for solving the application issue. The processor deploys the series of the executable operations to solve the application issue occurring at the network node to prevent a failure operation of the application.

Abstract: An unauthorized activity detection method in an onboard network system. The detection method includes determining whether or not a message sent out onto the network is an attack message, saving information relating to the attack message in at least one memory in a case where the message is an attack message, identifying a communication pattern from information relating to the attack message, and determining whether or not the message matches a communication pattern. The determination of whether an attack message and determination of whether matching a communication pattern are executed on each of a plurality of messages received from the network. In the determining of whether an attack message executed on a message received after executing of determining of whether matching a communication pattern, results of the determination of whether an attack message that has already be executed are used.

Abstract: Aggregation of noisy datasets into a master firmographic database. In an embodiment, firmographic records are received from a plurality of sources, and normalized into a common schema. One or more firmographic records may be cleaned by replacing a value of one or more fields in those firmographic record(s) with a value of those field(s) in another firmographic record. The firmographic records may then be clustered, and each of the clusters may be collapsed into a single conflated firmographic record based on a voting process. A master identifier may be generated for each conflated firmographic record, and the conflated firmographic records may be merged into a master firmographic database that is indexed by master identifiers.

Abstract: The disclosure below describes a knowledge pattern machine that goes beyond and is distinct from a traditional search engine as simple information aggregator. Rather than acting as a search engine of the data itself, the knowledge pattern machine use variously layers of artificial intelligence to discover correlations within the queries and historical data, and to derive and recognize data patterns based on user queries for predictively generating new knowledge items or reports that are of interest to the user. Previous patterns and knowledge items or reports are accumulated and incorporated in identification of new data patterns and new predictive knowledge items or reports in response to future user queries, thus providing a stateful machine. The predictive knowledge items are updated in real-time without user interference as the underlying data sources evolve overtime. The data patterns and knowledge items are organized hierarchically and may be shared among different users at various levels.

Abstract: A method, system, and computer-usable medium are disclosed for, comprising: examining pairs of samples in each leaf node of a plurality of decision trees and for each pair of samples, determining a neighbor frequency of each pair of samples as a frequency of how often such pair of samples are filtered into a particular leaf node of the plurality of decision trees and for each leaf node of the plurality of decision trees, determining a neighbor frequency of the leaf node as an average of the neighbor frequencies of the samples of the leaf node.

Abstract: A bot traffic detection system detects scripted network traffic. The bot traffic detection system may use a one-sided unsupervised machine learning technique to estimate distributions for human, non-scripted traffic (clean distributions). The clean distributions may be dynamically updated based on the latest traffic patterns. To estimate the clean distributions the bot traffic detection system may identify, for a certain subset of network traffic, feature values of the certain subset of network traffic that do not include bot traffic (clean buckets). Using clean traffic may provide more robust and stable behavior that can be tracked over time. Using the clean distributions, the bot traffic detection system may generate a rules table that indicates a likelihood that network traffic with a given combination of feature values is scripted network traffic. The bot traffic detection system may apply the rules table in real time to identify scripted network traffic.

Abstract: The present disclosure relates to a concept for machine-learning-based prediction of a destination for a user. Based on historic search data associated with the user, at least one candidate destination is determined based on the user and a given context. A plurality of embedding vectors are determined from an embedding matrix, wherein the embedding vectors are associated with the at least one candidate destination, the user, and the given context. The embedding matrix comprising embedding vectors for different components of the historic search data. The plurality of embedding vectors are fed into one or more first neural network layers to generate a semantic embedding for the candidate destination. The semantic embedding is into one or more second neural network layers to generate a probability score for the candidate destination.

Abstract: A dialog change service builds a directed graph of nodes, each described by code for a conversational agent stored in source control, the code comprising one or more dialog-based features. The dialog change service, responsive to receiving a request for a dialog change to a particular node of the conversational agent, evaluates the directed graph to identify an additional selection of nodes that require a similar changes as the dialog change based on textual and structural content. The dialog change service updates the code for the particular node and the additional selection of nodes to propagate the dialog change in the conversational agent stored in source control.

Abstract: A prediction manager for providing responsiveness predictions for deployments includes persistent storage and a predictor. The persistent storage stores training data and conditioned training data. The predictor is programmed to obtain training data based on: a configuration of at least one deployment of the deployments, and a measured responsiveness of the at least one deployment, perform a peak extraction analysis on the measured responsiveness to obtain conditioned training data, obtain a prediction model using: the training data, and a first untrained prediction model, obtain a confidence prediction model using: the conditioned training data, and a second untrained prediction model, obtain a combined prediction using: the prediction model, and the confidence prediction model, and perform, based on the combined prediction, an action set to prevent a responsiveness failure.

Abstract: Embodiments of the present disclosure pertain to verifying medical diagnoses. In one embodiment, a first classification unit is configured to process manually generated medical input data. The first classifier generates one or more first probabilities corresponding to one or more medical diagnoses. A second classification unit is configured to process machine generated medical input data associated with the manually generated medical data. The second classifier generates one or more second probabilities corresponding to the one or more medical diagnoses. A verification unit is configured to verify the manually generated input data by combining the first and second probabilities.

Abstract: The present invention provides a system for activating personal assistance services. The system includes an audio data collector adapted to collect a sample of speech, a processing module, and a service activator couple to an output device. The processing module further includes an audio feature extractor that extracts a plurality of acoustic features from the sample of speech, and a classification unit that classifies a status of a user from the plurality of acoustic features. The Service activator activates a personal assistance service according to the status of the user classified by the classification unit.

Abstract: A computer-implemented method for creating an object for data access is provided. The computer-implemented method includes defining a function-expression in source code using an object-oriented programming language. The defined function-expression is exposed in a graphical user interface of a development tool used to create the object. A selection of at least one field among a list of available fields for data access is received. The defined function-expression is included in the list of available fields. An integrated development environment is also provided.

Abstract: The disclosed computer-implemented method for improving forest-based malware detection within an organization may include (i) receiving, at a backend computing system, organization data from at least one organization computing device within an organization computer network, (ii) adjusting, at the backend computing system, a general use forest model based on the organization data to generate an organization-specific forest model for detecting malicious computer files within the organization computer network, and (iii) sending, from the backend computing system, the organization-specific forest model to the at least one organization computing device. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A method, apparatus and product for automatic detection of anomalies in graphs. The method comprising obtaining training data, the training data comprising a plurality of graphs, each defined by nodes and edges connecting between the nodes, at least some of the nodes are labeled; determining a statistical model of a graph in accordance with the training data, the statistical model takes into account at least one structured and labeled feature of the graph, wherein the structured and labeled feature of the graph is defined based on a connection between a plurality of nodes and based on at least a portion of the labels of the plurality of nodes; obtaining an examined graph; and determining a score of the examined graph indicative of a similarity between the examined graph and the training data, wherein the score is based on a value of the structured and labeled feature in the examined graph.

Abstract: The present disclosure relates to a computer-implemented method for iteratively solving a constraint satisfaction problem. The method may include assigning a value to each of one or more variables associated with the constraint satisfaction problem, each of the one or more variables having a first domain. The method may also include identifying an invalid solution resulting from a first value assigned to a first variable. The method may further include replacing the first value with a second value assigned to the first variable. The method may also include, upon identifying the invalid solution, generating a second domain larger than the first domain.

Abstract: A multidimensional analytical view (MDAV) type can be supported for an in-memory, columnar database, and MDAV calculations can be integrated into the database. Such an MDAV can be combined with other MDAV types as desired. Calculations typically performed at the application layer can be pushed down to the database. A model-based approach allows generation of a calculation plan, and calculations can be performed directly on the database.

Abstract: A service monitoring system receives a selection of key performance indicators (KPIs) that each indicate a different aspect of how a service provided by one or more entities is performing. Each entity of the one or more entities produces machine data or wherein each entity has its operation reflected in machine data not produced by the entity. Each KPI is defined by a different search query that derives one or more values from the machine data pertaining to the one or more entities providing the service, where each of the one or more values is associated with a point in time and representing the aspect of how the service is performing at the associated point in time. For each of the selected KPIs, the service monitoring system derives the one or more values and causes display of a graphical visualization of the derived one or more values for the KPI along a time-based graph lane.

Abstract: A method for performing relational analysis of parsed input is employed to create a visual map of knowledge information. A title, header or subject line for an input item of information is parsed into syntactical components of at least a subject component and any predicate component(s) relationally linked as topic and subtopics. A search of topics and subtopics is carried out for each parsed component. If a match is found, then the parsed component is taken as a chosen topic/subtopic label. If no match is found, then the parsed component is formatted as a new entry in the knowledge map. A translation function for translating topics and subtopics from an original language into one or more target languages is enabled by user request or indicated user preference for display on a generated visual map of knowledge information.

Abstract: In some examples, systems and techniques can determine a respective visit likelihood for each respective destination of a plurality of destinations based at least in part on a respective distance between the respective destination and a geographic location from a location history associated with a user and a comparison between a time associated with the geographic location and a visit likelihood distribution across time. The systems and techniques can then sort at least some of the plurality of destinations. In other examples, systems and techniques can determine whether a user is likely to visit a place during a future instance of a timeslot based at least in part on a location history associated with the user. The systems and techniques can then output information relating to the place prior to the beginning of the future instance of the timeslot.

Abstract: A system for the dynamic encoding in a semantic network of both syntactic and semantic information into a common four valued logical notation. The encoding of new information being benign to prior syntactic constructions, tests for N conditionals in time O(C) and allows for the proper quantification of variables at each recursive step. The query/inference engine constructed from such an implementation is able to optimize short term memory for maximizing long term storage in the automaton. In a parallel context this can be viewed as optimizing communication and memory allocation between processes. Disclosed embodiments of a four valued encoding achieve efficiencies in minimizing the run time stack and maximizing heap size especially in a parallel architecture. Disclosed embodiments are useful in simulation and search engine implementations.

Abstract: In some examples, systems and techniques can determine a respective visit likelihood for each respective destination of a plurality of destinations based at least in part on a respective distance between the respective destination and a geographic location from a location history associated with a user and a comparison between a time associated with the geographic location and a visit likelihood distribution across time. The systems and techniques can then sort at least some of the plurality of destinations. In other examples, systems and techniques can determine whether a user is likely to visit a place during a future instance of a timeslot based at least in part on a location history associated with the user. The systems and techniques can then output information relating to the place prior to the beginning of the future instance of the timeslot.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for estimating demographic compositions of television audiences from audience similarities. In an aspect, pairwise cosine similarities of segment of viewers of rated and unrated television programs are determined, demographic information of rated programs is attributed to the unrated programs from the cosine similarity measures and demographic information for the rated programs.

Abstract: A quantified belief propagation (QBP) algorithm receives as input an existentially quantified boolean formula (QBF) of existentially quantified boolean variables, universally quantified variables, and boolean operators. A tripartite graph is constructed, and includes (i) there-exists nodes that correspond to and represent the existentially quantified variables, (ii) for-all nodes that correspond to and represent the universally quantified variables, and (iii) sub-formula nodes that correspond to and represent sub-formulas of the QBF. A set of boolean values of the existentially quantified variables is found by (i) passing a first message from an arbitrary sub-formula node to an arbitrary for-all node, and (ii) in response, passing a second message from the arbitrary for-all node to the arbitrary sub-formula node.

Abstract: A system and method for secure data management is presented. The method comprises receiving a query, performing the query and obtaining answers to it, creating certain formulas representing the answers; and determining whether there is a minimum number of distinct models of these formulas conjoined with the relevant anonymity predicates along with previous answers and general background knowledge, and when that number of models does not exist, suppressing the answers. In one aspect, the method further comprises creating formulas encoding the information conveyed by the answer; and combining the formulas encoding the information with prior information.

Abstract: An interpretation engine for an interoperable device is disclosed. The interpretation engine is configured to receive data and to determine a function to be executed in response to the received data and in accordance with a knowledge base that comprises one or more relationships that are structured in triplets. Each triplet comprises a subject, a predicate and an object. The interpretation engine is configured to recognize only a predetermined, fixed plurality of predicate values.

Abstract: An approach for building an ontology is provided. Based on a grammar, extracted complex triples are syntactically transformed to identify core terms. The syntactically transformed complex triples are semantically transformed into simplified triples referring to new terms that conceptualize the core adjectives, adverbs and verbs, and assigning the core terms to respective definitions and keys in a reference ontology, thereby retaining the semantics of the complex triples. Based on a meta-schema of the reference ontology, an enrichment transformation of the simplified triples is performed to create simplified and enriched triples by adding relations derived from a correspondence each term in the simplified triples has with the reference ontology and by adding representations of semantics of reference ontology definitions of the terms. The simplified and enriched triples are stored as an ontology representing knowledge in an application providing the free-form text from which the complex triples were extracted.

Abstract: A prediction method that estimates the real-time position of a mobile device based on previously observed data is provided. The present invention can be used in real-time navigation, including providing real-time alerts of an upcoming destination and notifications of emergency events in close geographic proximity. The prediction method utilizes neural networks and/or functions generated using genetic algorithms in estimating the mobile device's real-time position. The prediction method provides reliable Location-Based Services (LBS) in events where traditional positioning technologies become unreliable. It is also seamless, as the user remains unaware of any interruption in accessing the positioning technology.

Abstract: In some examples, systems and techniques can determine a respective visit likelihood for each respective destination of a plurality of destinations based at least in part on a respective distance between the respective destination and a geographic location from a location history associated with a user and a comparison between a time associated with the geographic location and a visit likelihood distribution across time. The systems and techniques can then sort at least some of the plurality of destinations. In other examples, systems and techniques can determine whether a user is likely to visit a place during a future instance of a timeslot based at least in part on a location history associated with the user. The systems and techniques can then output information relating to the place prior to the beginning of the future instance of the timeslot.

Abstract: A prediction method that estimates the real-time position of a mobile device based on previously observed data is provided. The present invention can be used in real-time navigation, including providing real-time alerts of an upcoming destination and notifications of emergency events in close geographic proximity. The prediction method utilizes neural networks and/or functions generated using genetic algorithms in estimating the mobile device's real-time position. The prediction method provides reliable Location-Based Services (LBS) in events where traditional positioning technologies become unreliable. It is also seamless, as the user remains unaware of any interruption in accessing the positioning technology.

Abstract: First order predicate logics are provided, extended with a bilattice based uncertainty handling formalism, as a means of formally encoding pattern grammars, to parse a set of image features, and detect the presence of different patterns of interest implemented on a processor. Information from different sources and uncertainties from detections, are integrated within the bilattice framework. Automated logical rule weight learning in the computer vision domain applies a rule weight optimization method which casts the instantiated inference tree as a knowledge-based neural network, to converge upon a set of rule weights that give optimal performance within the bilattice framework. Applications are in (a) detecting the presence of humans under partial occlusions and (b) detecting large complex man made structures in satellite imagery (c) detection of spatio-temporal human and vehicular activities in video and (c) parsing of Graphical User Interfaces.

Abstract: A method of managing an axiom includes determining if an H in an A?H axiom in existing axioms is a predetermined expression; and if the H is the predetermined expression, collecting, from the existing axioms, a B?I axiom in which the B includes the H, deleting the A?H axiom from the existing axioms, and generating a new B?I axiom from the collected B?I axiom by replacing the H in the B with the A.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for estimating demographic compositions of television audiences from audience similarities. In an aspect, pairwise cosine similarities of segment of viewers of rated and unrated television programs are determined, demographic information of rated programs is attributed to the unrated programs from the cosine similarity measures and demographic information for the rated programs.

Abstract: A system and method for secure data management is presented. The method comprises receiving a query, performing the query and obtaining answers to it, creating certain formulas representing the answers; and determining whether there is a minimum number of distinct models of these formulas conjoined with the relevant anonymity predicates along with previous answers and general background knowledge, and when that number of models does not exist, suppressing the answers. In one aspect, the method further comprises creating formulas encoding the information conveyed by the answer; and combining the formulas encoding the information with prior information.

Abstract: Information cluster systems and methods are presented. In one embodiment, a cluster method comprises: performing an engine process including issuing requests to bring a resource online, offline, and monitor the resources, wherein the engine process is performed by an engine; performing a resource interaction process including interacting with a resource and directing a resource to comply with the request from the engine process, wherein the resource interaction process is performed by a resource interaction agent; performing a predicate logic process including performing predicate logic operations to determine if a predicate logic condition associated with the resource is satisfied and forwarding an indication of the results of the predicate logic operations to the engine process, wherein the predicate logic process is performed by a predicate logic agent that is separate from the engine performing the engine process.

Abstract: Systems, methods, and other embodiments associated with data sources adapted for parallel inference on triples associated with a semantic model are described. One example method includes creating a source table that stores triples for entailment in a manner that is adapted for parallel inference. The source table may be partitioned by triple predicate or may store compact triple identifiers that have been mapped to triple identifiers from the semantic model.

Abstract: A system and method for automatic formal verification of an executable system includes an assertion monitor configured to verify a system against an assertion in the specification. The assertion monitor includes a parser configured to generate a propositional formula representing an assertion in the specification using Boolean propositions, a filter configured to generate a trace of truth assignments for the propositional symbols and a trace verifier configured to verify the assertion using the trace of truth assignments for the propositional symbols and the propositional formula.

Abstract: The invention provides for a computer-implemented method for solving a linear program (LP), the method comprising the steps of: receiving (100) the linear program; determining (101) a kernel K and determining a kernel matrix G of the kernel K, wherein the kernel matrix is a non-singular submatrix of the original matrix; determining (102) a set of non-basic variable indices and a set of extra constraint indices; computing (103) a primal kernel vector (xK) from the determined kernel; computing (104) a dual kernel vector (yK) from the determined kernel; and evaluating (105) the feasibility of the primal kernel vector and of the dual kernel vector.

Abstract: Systems, methods, and other embodiments associated with data manipulation operation restriction enforcement on graph data are described. A statement specifying a data manipulation operation to modify graph data that will modify one or more triples in the graph data is received. One or more resources that the one or more triples describe is determined. Data constraints associated with the resources are accessed. The access constraints are evaluated on the graph data. The data manipulation operation is selectively restricted based, at least in part on, the evaluation of the access constraints.

Abstract: An apparatus for making probabilistic inferences based on a belief network includes a processing system configured to receive as input one or more parameters of a causal influence model. The belief network has a child node Y and one or more parent nodes Xi (i=1, . . . , n) for the child node Y. The causal influence model describes the influence of the parent nodes Xi on possible states of the child node Y. The processing system is further configured to use a creation function to convert the parameters of the causal influence model into one or more entries of a conditional probability table. The conditional probability table provides a probability distribution for all the possible states of the child node Y, for each combination of possible states of the parent nodes Xi.

Abstract: A website may allow users to rate content items displayed thereon. The user ratings associated with the content items may be weighted according to user rating weight. The rating weight of a particular user may be based, in whole or part, upon whether the user consistently rates content items similarly to one or more “expert” raters. The experts may be selected based on various factors including: professional training, experience, reputation on the website, or the like. If a particular user consistently rates content items similarly to the experts, the user may be given a high rating weight. Users whose ratings consistently diverge from the ratings assigned by the experts may be given a lower rating weight. The weights may be displayed in connection with a user profile, or may be hidden from the user community. The rating weights may be used in calculating an overall rating of a particular content item.

Abstract: Methods and apparatus to characterize and predict network health status are presented. A particular method of characterizing and predicting a health status of a network node includes receiving a plurality of report values associated with the network node, where each report value is one of alert and normal, and where each report value is received at a corresponding time. The method includes calculating a current report weight associated with a current report value received at a current time. When a first previous report value was received at a first previous time before the current time, the current report weight is calculated based at least partially on the current report value and at least partially on a first previous report weight associated with the first previous value. The method includes storing the current report weight at a memory device.

Abstract: Policies can combine the efficiency of rule sets with the flexible expression power of workflow engines, as well as advantages of other programming languages and engines. Consistent modelling of policy rules is provided across different levels, such as network, control, and service levels. In one system, a policy module can orchestrate rule set and workflow engines in order to evaluate different conditions of a policy. In another embodiment, a hybrid language can be used that allows rule set and other variables to be used in a business process container, workflow and other variables to be used in a rule set container, rule set and business process variables to be used in other containers. In this way, a subset of conditions in a policy can be evaluated using multiple languages and/or engines in order to enforce or evaluate a policy.

Abstract: A method and an arrangement in electronic systems. Uncertainties are handled in a special way. The uncertainties can regard a parameter that is to be determined or uncertainties in signals or data that constitute a basis for determination of the sought parameter. There is often also a time-critically within the application. In certain applications, changes can be rapid and the invention can also handle such dynamic processes. Detection and prediction in electronic systems are carried out.

Abstract: Architecture that facilitates validation of a data mapping of data from a data source to a data target. There is included a signature generation component that generates a source signature of all or a portion of the data source and a target signature of all or a corresponding portion of the data target, and a sampling component that obtains a sample of the source data a corresponding sample of the target data. The data signatures and data samples are compared respectively and processed with a processing component to determine the status of the validation process.