Abstract: Methods of detecting and/or identifying an artificial target within an image are provided. These methods comprise: applying to a region of the image a primary classification algorithm for performing a feature extraction of the image region, the primary classification algorithm being based on a spectral profile defined by one or more spectral signatures with one or more features in at least part of the infrared spectrum; obtaining a relation between the extracted features of the image region and the spectral profile; verifying whether a level of confidence of the obtained relation between the extracted features and the spectral profile is higher than a first predetermined confirmation level; and, in case of positive (or true) result of said verification, determining that the image region corresponds to artificial target to be detected, thereby obtaining a confirmed artificial target. Systems and computer programs are also provided that are suitable for performing said methods.

Abstract: A system and method for utilizing automatically generating related subject matter areas to create footer links for a subject matter page of a tutor matching website is disclosed and claimed. In particular, a document embedding algorithm such as word2vec, doc2vec or GloVe can be trained using on a particular subject using pertinent material such as, for example, textbooks, learned papers, and transcripts of lectures. Once trained, the document embedding algorithm can be used to generate a list of related subjects that can be used to automatically build footer links for a particular subject matter page. The related footer links can improve the positioning of the tutor matching website with common Internet search engines. The same document embedding algorithm can be used to identify previously answered questions that may assist a user posing a question to the tutor matching website.

Abstract: Example methods and systems are provided for network diagnosis. One example method may comprise: detecting an egress packet and determining whether each of multiple network issues is detected for the egress packet or a datapath between a first virtualized computing instance and a second virtualized computing instance. The method may also comprise: generating network diagnosis code information specifying whether each of the multiple network issues is detected or not detected; generating an encapsulated packet by encapsulating the egress packet with an outer header that specifies the network diagnosis code information; and sending the encapsulated packet towards the second virtualized computing instance to cause a second computer system to perform one or more remediation actions based on the network diagnosis code information.

Abstract: A method is used in issuing alerts for storage volumes using machine learning. A machine learning system analyzes Input/Output (I/O) data of a storage volume in a data storage system. The machine learning system is trained with sample I/O data patterns associated with the storage volume. Based on the I/O data, the machine learning system identifies atypical behavior associated with I/O data patterns of the I/O data. The method then issues an alert.

Abstract: An artificial neural network-based method for selecting a surface type of an object is suitable for selecting a plurality of objects. The artificial neural network-based method for selecting a surface type of an object includes performing surface type identification on a plurality of object images by using a plurality of predictive models to obtain a prediction defect rate of each of the predictive models, wherein the object images correspond to surface types of a part of the objects, and cascading the predictive models according to the respective prediction defect rates of the predictive models into an artificial neural network so as to select the remaining objects.

Abstract: The value for equipment to be replaced can be maximized by determining a threshold cutoff value for a failure prediction indicator and a window size for obtaining the threshold cutoff value for a piece of oil or gas drilling or recovery equipment; applying the threshold cutoff value and the window size to an equipment failure prediction model; and deriving a recall value and an average hour-loss value from the equipment failure prediction model. Predictive maintenance for the piece of oil or gas drilling or recovery equipment may be performed based on the recall value and the average hour-loss value to perform predictive maintenance for a piece of equipment in an oil or gas recovery operation.

Abstract: A system and method are disclosed for bottom-up modeling and prediction of asset anomalies. In one embodiment, the system receives first sensor data from a first sensor, the first sensor associated with a first asset; smooths the first sensor data; determines a first stage within sensor data, the sensor data including the first sensor data; determines a first set of anomalies within the first stage; generates a first asset state space associated with the first asset; trains a base model describing a first group of one or more assets, the first group of one or more assets including the first asset; and trains, using the base model, a final model particular to the first asset.

Abstract: Techniques for tuning a machine learning algorithm using automatically determined optimal hyperparameters are described. An exemplary method includes receiving a request to determine a search space for at least one hyperparameter of a machine learning algorithm; determining, according to the request, optimal hyperparameter values from the search space for at least the one hyperparameter of the machine learning algorithm based on an evaluation of hyperparameters from the same machine learning algorithm on different datasets; and tuning the machine learning algorithm using the determined optimal hyperparameter values for the at least one hyperparameter of the machine learning algorithm to generate a machine learning model.

Abstract: A demand prediction method using a computer includes calculating an error between a prediction value of an amount of a past demand and a measurement value of the amount of the past demand, the prediction value of the amount of the past demand being calculated by inputting a measurement value of a past explanatory variable to a prediction model that is constructed in accordance with a measurement value of an amount of a demand at a predetermined location and an explanatory variable that serves as an external factor that affects an increase or a decrease in the amount of the demand at the predetermined location, determining whether the calculated error is an abnormal value, acquiring a new explanatory variable if the error is determined to be the abnormal value, updating the prediction model in accordance with the acquired new explanatory variable, and newly predicting an amount of a future demand using the updated prediction model.

Abstract: The serving of advertisements with (e.g., on) video documents may be improved in a number of ways. For example, a system may (a) accept information defining at least one ad spot associated with at least one instance of an video document, (b) accept offers to have advertisements served in the ad spot(s), and (c) arbitrate among competing advertisements, using at least the offers, to determine at least one advertisement to be served in that ad spot(s). As another example, a system may (a) accept relevance information for an advertisement, (b) determine at least one video document using the accepted relevance information, (c) present information about the video document(s) to an advertiser associated with the advertisement, and (d) accept, from the advertiser, an offer to have its advertisement served with at least one of the video document(s) accepted.

Abstract: A flash memory system may include a flash memory and a circuit for decoding a result of a read operation on the flash memory using a first codeword. The circuit may be configured to generate an estimated codeword based on a result of hard decoding the first codeword and a result of hard decoding a second codeword. The circuit may be further configured to generate soft information based on the hard decoding result of the first codeword and the estimated codeword. The circuit may be further configured to decode the result of the read operation on the flash memory using the soft information.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to estimate an audience population.

Abstract: Systems and methods for unsupervised training and evaluation of anomaly detection models are described. In some embodiments, an unsupervised process comprises generating an approximation of a data distribution for a training dataset including varying values for a metric of a computing resource. The process further determines, based on the size of the training dataset, a first quantile probability and a second quantile probability that represent an interval for covering a prescribed proportion of values for the metric within a prescribed confidence level. The process further trains a lower limit of the anomaly detection model using a first quantile that represents the first quantile probability in the approximation of the data distribution and an upper limit using a second quantile that represents the second quantile probability in the approximation. The trained upper and lower limits may be used to monitor input data for anomalous behavior and, if detected, trigger responsive action(s).

Abstract: Facilitation of an explanation about an object to be analyzed is realized with high accuracy and with efficiency. A data analysis apparatus is disclosed which uses a first neural network configured with an input layer, an output layer, and two or more intermediate layers provided between the input layer and the output layer. Each performs a calculation by giving data from a layer of a previous stage and a first learning parameter to a first activation function and outputs a calculation result to a layer of a subsequent stage. The data analysis apparatus includes a conversion section; a reallocation section; and an importance calculation section.

Abstract: A differential pressure monitoring system includes a differential pressure monitoring device and at least one client device. The differential pressure monitoring device includes a water pressure sensor that detects a water pressure at an inlet of the dry pipe valve, a valve air pressure sensor that detects an air pressure at an outlet of the dry pipe valve, and a control circuit that computes a ratio of the water pressure and the air pressure, predicts whether a valve tripping event is expected to occur based on the computed ratio, and in response to predicting that the valve tripping event is expected to occur, provides a prediction that the valve tripping event is expected to occur for remedial action. The system includes at least one client device that receives the prediction from the control circuit and presents display data regarding the prediction.

Abstract: Technology described in this document can be embodied in a method that includes generating, using one or more processing devices of a vehicle operating in an environment, a discretized representation of the environment, including a plurality of cells. The cells are occupied by particles representing at least one of: an object, or a free space in the environment. Sensor data indicative of a state of at least one particle is received, and an update to a time-varying particle density function associated with a location of the particle in the dynamic occupancy grid is calculated from the sensor data and one or more models. The method further includes generating a prediction of occupancy of at least one cell in the discretized representation based on the updated particle density function, and operating the vehicle, using a controller circuit of the vehicle, based at least in part on the prediction.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to calculate long-term effects of marketing campaigns.

Abstract: A method is used in issuing alerts for storage volumes using machine learning. A machine learning system analyzes Input/Output (I/O) data of a storage volume in a data storage system. The machine learning system is trained with sample I/O data patterns associated with the storage volume. Based on the I/O data, the machine learning system identifies atypical behavior associated with I/O data patterns of the I/O data. The method then issues an alert.

Abstract: Methods, systems and computer program products for determining an accurate flow rate based on limited observations are provided. Aspects include receiving a limited sample of flow rate data associated with a test site of a test river. Aspects also include determining topographical characteristics of the test site. Aspects also include determining a plurality of similar tested sites and generating a test site rating curve based on a combination of the limited sample of flow rate data and sufficient flow rate data associated with one or more of the plurality of similar tested sites. Aspects also include obtaining a stage reading associated with the test site from a sensor disposed at the test site. Aspects also include determining a flow rate of the test site based on the stage reading and the test site rating curve.

Abstract: Devices, systems, and methods are provided for classifying detected objects as static or dynamic. A device may determine first light detection and ranging (LIDAR) data associated with a convex hull of an object at a first time, and determine second LIDAR data associated with the convex hull at a second time after the first time. The device may generate, based on the first LIDAR data and the second LIDAR data, a vector including values of features associated with the first convex hull and the second convex hull. The device may determine, based on the vector, a probability that the object is static. The device may operate a machine based on the probability that the object is static.

Abstract: A method for detecting a fall by a user wearing a fall detector, including: detecting a trigger event identifying the time location of a possible fall event in user data; extracting motion features from motion data and physiological features from physiological data from within a time window around the identified time location; and determining whether the detected trigger event is a fall by the user by inputting the at least one of the motion features and at least one of the physiological features into a classifier.

Abstract: Processor systems to estimate audience sizes and impression counts for different frequency intervals are disclosed. An example processor system includes a memory management unit (MMU) to assign requests from computing devices indicative of accesses to media to a first block of memory and to assign user-identified impression data corresponding to user-identified impressions to a second block of memory. The processor system including an arithmetic logic unit (ALU) to determine multipliers relating a first probability distribution for the user-identified impressions to a second probability distribution for census impressions. The multipliers based on probability constraints defined by weighted probabilities associated with the second probability distribution, where different ones of the probabilities weighted based on estimated populations for corresponding ones of the different demographics.

Abstract: A substation asset management method can include the steps of: determining whether to compensate a reliability model by element of the substation by comparing reliability of a reference reliability model by substation type with reliability depending on health index by element thereof generated based on state data and real-time monitoring data by element of the substation; compensating the reference reliability model by substation type and generating a unique reliability model by element of the substation as a result of the determination; evaluating system reliability and economic feasibility by maintenance scenario based on a reference system reliability model for each candidate element subject to maintenance among the elements of the substation; and drawing a maintenance scenario of the elements of the substation by using a fitness function of maintaining the substation based on the result drawn through the evaluation of the system reliability and the economic feasibility.

Abstract: Techniques for predicting failure mode specific reliability characteristics of tangible equipment using parametric probability models are disclosed. In some example embodiments, a computer system receives a model training configuration entered via a user interface, trains a failure curve model for a selected failure mode of a selected equipment model based on the model training configuration at a time indicated by training schedule data, generating, and generates analytical data for the selected failure mode of the selected equipment model using the trained failure curve model. The failure mode corresponds to a specific way in which the equipment model is capable of failing. In some example embodiments, the training of the failure curve model comprises determining a shape parameter and a scale parameter for the failure curve model based on a fitting of failure event data to a continuous probability distribution, and storing the parameters for use in generating the analytical data.

Abstract: Provided herein are systems and methods for providing digital guidance in an underlying computer application. In one exemplary implementation, a method includes recording, in a computing device, steps of digital guidance content as the steps are created by a content author. The exemplary method also includes automatically segmenting, in the computing device, the digital guidance content as it is being created such that the digital guidance content is only associated with segments of the underlying computer application where the content is relevant. The exemplary method further includes making the digital guidance content available for playback to an end user on a computing device only when the end user is in a segment of the underlying computer application that is relevant to the digital guidance content.

Abstract: This patent specification relates to systems and methods that use networks of standardized time series and models in a generic and extensible platform. More particularly, this patent specification relates to standardizing time series and models, using standardized time series and models in a network of time series (NOTS), and creating networks of time series in a NOTS platform. In addition, this patent specification relates to use of nodes that can be arranged in any user defined order, and each node is evaluated to return a time array that may be used to populate a time series. A layering framework may be used to define how each node is evaluated.

Abstract: An artificial intelligence (AI) based system for detecting defects in infrastructure uses an image recognizer and image data. A set of annotated training and validation data is generated to train and validate the image recognizer. The image data is annotated with classification data such as defect type and severity of the defect. Once trained and validated, the image recognizer can analyze inspection images to identify detects therein and generate an output report including the identification and classification of the defect, and remediation recommendations.

Abstract: Seismic data of a subterranean region is received by data processing apparatus. The seismic data includes multiple seismic data points. For each seismic data point, gradients are computed based on the received seismic data and a dip angle is computed based on the gradients for the each seismic data point. The dip angle is smoothed using anisotropic diffusion.

Abstract: The present disclosure provides an apparatus for complementing automotive radar, which comprises a camera for obtaining an image of a field of view, a radar sensor for transmitting a radar signal and detecting a radar signal reflected by an object, and a controller for obtaining first information on a target from the image, setting a monitoring range of the radar sensor according to the obtained first information, obtaining second information on the target based on a radar signal detected in the monitoring range, and detecting malfunction of the radar by determining whether or not the first information matches the second information, and further provides a method thereof. According to the present disclosure, it is possible to quickly detect malfunction of the radar sensor.

Abstract: The present disclosure relates to a signal processing apparatus and system. In calibration, a sensor section is used to acquire a time of day when a stereo camera image capturing a target is acquired and a time of day when a radar image capturing the target is acquired. The target has a radar reflector and a marker. The image acquisition times of day are acquired while at the same time changing a distance between the sensor section and the target so as to find an amount of time-of-day discrepancy, a difference in time of day between the stereo camera image and the radar image at the same distance. At the time of object detection, the radar image acquired first is buffered and then output together with the stereo camera image acquired by the time-of-day discrepancy later, thus combining the stereo camera image and the radar image.

Abstract: A research analytics system and method, includes an Internet-based data storage service maintaining, using an integrated database scheme, standardized research process flows, organizational tasks, and research outcomes for a plurality of research programs. The system maintains definitions of research performance indicators, identifies targets for the research performance indicators that are aligned with goals associated with a research organization, maps performance-related data from various data sources to the integrated database scheme and stores the standardized research process flows, the organizational tasks, and the research outcomes, grouped by the research programs.

Abstract: Examples of techniques for anomaly detection with feedback are described. An instance includes a technique is receiving a plurality of unlabeled data points from an input stream; performing anomaly detection on a point of the unlabeled data points using an anomaly detection engine; pre-processing the unlabeled data point that was subjected to anomaly detection; classifying the pre-processed unlabeled data point; determining the anomaly detection was not proper based on a comparison of a result of the anomaly detection and a result of the classifying of the pre-processed unlabeled data point; and in response to determining the anomaly detection was not proper, providing feedback to the anomaly detection engine to change at least one emphasis used in anomaly detection.

Abstract: Systems and methods for treating a subject are provided. A first dataset comprising timestamped autonomous glucose measurements of the subject over a first time course is obtained. A second dataset, associated with a standing insulin regimen for the subject over the first time course and comprising insulin medicament records, is also obtained. Each record comprises a timestamped injection event including an amount and type of insulin medicament injected into the subject by an insulin pen. The first and second datasets serve to calculate a glycaemic risk measure and an insulin sensitivity factor of the subject during the first time course, which are used to obtain a basal titration schedule and a fasting blood glucose profile model over a subsequent second time course for the subject. The model predicts the fasting blood glucose level of the subject based upon amounts of basal insulin medicament injected into the subject.

Abstract: The invention shows how to use noise-like perturbations to improve the speed and accuracy of Markov Chain Monte Carlo (MCMC) estimates and large-scale optimization, simulated annealing optimization, and quantum annealing for large-scale optimization.

Abstract: A data collection apparatus includes: a collection unit which collects data relating to respective vehicles from vehicular devices installed in the respective vehicles; a reception unit which receives a data collection condition specified by a user; an analysis unit which calculates a probability of occurrence of an event for collection of data specified in the collection condition based on past data collection results; and a providing unit which provides the user with information that is based on the probability of occurrence calculated by the analysis unit.

Abstract: One embodiment can provide a system for estimating useful life of a load-bearing structure. During operation, the system can perform a degradation measurement on the structure to obtain degradation data for a predetermined time interval, apply a constraint convex regression model to the degradation data, estimate a total useful life (TUL) of the structure based on outputs of the constraint convex regression model, and predicting a remaining useful life (RUL) based on the TUL and a current time.

Abstract: Ensemble classification algorithms having subclass resolution are disclosed. An example disclosed apparatus includes a fingerprint generator to generate a fingerprint of class probabilities of each of a plurality of samples, a distribution creator to create a distribution of the samples based on the generated fingerprints, and a distribution applicator to apply the distribution to a population to predict sub-class probabilities of each of the population.

Abstract: A plurality of basic simulations independent of one another are carried out, which determine respective remaining service life predictions for the machine. The remaining service life predictions and characteristic data are fed to a neural network, which outputs weights for the remaining service life predictions. A final prediction is calculated from the remaining service life predictions by weighting the remaining service life predictions relative to one another. A hybrid model is produced, which results from the combination of the basic simulations with the neural network. The remaining service life can be predicted not only for a small number of machines for which a specific simulation model has been manually created. The hybrid model enables condition monitoring for any further types and configurations of machines that merely belong to the same machine class. The basic simulations can therefore also be applied to previously unknown machines.

Abstract: Embodiments of the invention are directed to techniques for detecting anomalous values in data streams using forecasting models. In some embodiments, a computer can receive a value of a data stream comprising a plurality of data values, where the received value corresponds to a time interval and previously received values each correspond to a previous time interval. Models can be selected based on the time interval, where each of the models has a different periodicity. For each of the selected models, the computer may generate a score by generating a prediction value based on the model and generating the score based on the prediction value and the received value. A final score can then be generated based on the scores. Next, a score threshold can be generated. If the final score exceeds the score threshold, the computer may generate a notification that indicates that the data value is an anomaly.

Abstract: An approach is provided that detects when multiple apps are being displayed on a display screen with each of the apps having any number of user interface controls. A set of user interface controls is identified from the user interface controls being displayed. The identified set of user interface controls have an increased display preference. The identified set of user interface controls are enlarged on the display screen making such controls easier to read and select.

Abstract: A log-to-metrics transformation system includes a log-to-metrics application executing on a processor. The log-to-metrics transformation system receives a format associated with machine data, and further receives, via a first graphical control, a first set of metric identifiers corresponding to a first set of metrics associated with the machine data. The log-to-metrics transformation system generates a first set of mappings between the first set of metric identifiers and a first set of field values included in the machine data. The log-to-metrics transformation system stores the first set of mappings and an association with the format of the machine data. The log-to-metrics transformation system, based on the first set of mappings, causes the first set of field values to be extracted from the machine data. Further, a first metric included in the first set of metrics is determined based on at least a portion of the first set of field values.

Abstract: Disclosed herein are computer-implemented methods and systems for forecasting security ratings for an entity. The methods and systems can include generating a plurality of simulated instantiations of a security scenario for the entity, in which the security scenario characterized by a plurality of security events associated with at least one event type. The methods and systems can further include determining a security rating for each instantiation of the plurality of instantiations; and generating a forecast cone based on the determined security ratings for the plurality of instantiations. In some examples, for each event type of the at least one event type, the methods and systems can include determining a rate, duration, and/or temporal placement of the security events associated with the event type over a forecasting period.

Abstract: This application relates to apparatus and methods for identifying anomalies within a time series. In some examples, a computing device receives sales data identifying a sale of at least one item, and aggregates the received data in a database. The computing device may generate a plurality of time series based on the aggregated sales data. The computing device may extract features from the plurality of time series, and generate an alerting algorithm that is based on clusters of the extracted features. The computing device may apply the alerting algorithm to a time series generated from received sales data to determine whether the time series is an anomaly. Based on the determination, the computing device may generate and transmit anomaly data identifying whether the time series is an anomaly, such as to another computing device.

Abstract: Described is a system for predicting future social activity. The system extracts social activities from spatial-temporal social network data collected in a first time period ranging from hours to days to capture spatial structures of social activities in a graph network representation. A graph matching technique is applied over a set of spatial-temporal social network data collected in a second time period ranging from weeks to months to capture temporal structures of the social activities. A spatial-temporal structure of each social activity is represented as an activity core, where each activity core is defined as active nodes that participate in the social activity with a frequency over a predetermined threshold over the second time period. For each activity core, the system computes statistics of the social activity and uses the statistics to generate a prediction of future behaviors of the social activity.

Abstract: A media curation system configured to perform operations that include, capturing an image at a client device, wherein the image includes a depiction of an object, identifying an object category of the object based on the depiction of the object within the image, accessing media content associated with the object category within a media repository, generating a presentation of the media content, and causing display of the presentation of the media content within the image at the client device.

Abstract: A reading device includes wireless slave devices that communicate via a multihop wireless network, a measuring unit that measures the state of an inspection target to generate measurement data, a measurement data analysis unit, and a failure cause analysis unit that analyzes a failure cause when the analysis of the failure cause analysis unit fails. The reading device transmits the measurement data to the data collecting device by associating the measurement data with failure cause information. The data collecting device includes a wireless master device, a measurement data acquisition unit that acquires measurement data from each reading device, a recording unit that records the measurement data when the measurement data is not associated with failure cause information, and a retry control unit that retries data acquisition according to the type of the failure cause when the measurement data is associated with the failure cause information.

Abstract: A video packaging and origination service can process requests for content segments from requesting user devices. The video packaging and origination service can receive multiple inputs from content sources and determine events in the depictions of the inputs. The video packaging and origination service can generate content streams from the detected events. The content streams can be provided according to subscription requests.

Abstract: Techniques for training and evaluating seasonal forecasting models are disclosed. In some embodiments, a network service generates, in memory, a set of data structures that separate sample values by season type and season space. The set of data structures may include a first set of clusters corresponding to different season types in the first season space and a second set of clusters corresponding to different season types in the second season space. The network service merges two or more clusters the first set and/or second set of clusters. Clusters from the first set are not merged with clusters from the second set. After merging the clusters, the network service determines a trend pattern for each of the remaining clusters in the first and second set of clusters. The network service then generates a forecast for a metric of a computing resource based on the trend patterns for each remaining cluster.

Abstract: In one embodiment, a device obtains temperature and utilization measurements for a set of network interface transceivers in a network. The device computes, for each of the transceivers, a correlation coefficient between its temperature and utilization measurements. The device applies a k-nearest neighbor classifier to the correlation coefficients, to predict a correlation coefficient. The device uses the predicted correlation coefficient to determine an impact of temperature on utilization of a particular network interface transceiver. The device initiates a mitigation action, when the determined impact of temperature on the utilization of the particular network interface transceiver exceeds a predefined threshold.

Abstract: A failure positioning method for positioning leakage defect cell between the gate and the active region of transistor cells arranged in an array. The positioning method includes the steps of: measuring the resistance between a first metal wire connecting the active regions and a second metal wire connecting the gates, and positioning a first region where the defect cell is located by resistance ratio; electrically isolating the active region contact holes and the gate contact holes from each other; shorting the gate contact holes in the first region; and performing active voltage contrast analysis on the plurality of columns of transistor cells in the first region to position the leakage defect in the first region by comparing the voltage contrast images. With the positioning method, the transistor cell having a leakage defect at nA level may be accurately found from a plurality of transistor cells arranged in an array.