Abstract: The anomaly detection device according to an embodiment has a calculator and a determiner. The calculator is configured to calculate a degree of anomaly according to a predictive value that is predicted through machine learning using data acquired from a target device and a measurement value that is actually measured for the target device. The determiner is configured to determine whether a change of the degree of anomaly indicates an anomaly of the target device according to a degree of a change of the degree of anomaly calculated by the calculator within a predetermined time range.

Abstract: A causal analysis device according to an embodiment includes an extractor, a counter, a calculator, and a generator. The extractor performs extraction from combination data in which an incident and one or a plurality of factors causing the incident are associated with each other in accordance with occurrence of combination data. The counter counts the number of occurrences of the incident or the number of occurrences of a combination of the plurality of factors on the basis of a combination of the incident and the factors included in the extracted combination data. The calculator calculates an index value that corresponds to the combination of the incident and the factors and changes in accordance with the number of occurrences of the combination of the incident and the factors on the basis of a result of the counting of the counter.

Abstract: An anomaly detection device according to the embodiment includes a prediction unit and an anomaly score calculation unit. The prediction unit performs a process to obtain, at each time step of the time series data of m dimensions, distribution parameters required to express a continuous probability distribution representing a distribution state of predicted values that can be obtained at a time step t of the time series data of m dimensions. The anomaly score calculation unit performs a process to calculate, using distribution parameters obtained by the prediction unit, an anomaly score corresponding to an evaluation value representing evaluation of a magnitude of anomaly in an actual measurement value at the time step t of time series data of m dimensions.

Abstract: According to an embodiment, a training apparatus is a training apparatus training an inference model performing predetermined inference based on data, the training apparatus changes, according to a predefined distance for a selected piece of data selected based on predefined scores given to a plurality of pieces of data in a predefined classification space, the predefined score of a piece of data other than the selected piece of data, gives labels in the predefined classification space to pieces of data including the piece of data the predefined score of which is changed, builds training data from the pieces of data to which the labels are given, and updates the inference model using the training data.

Abstract: According to an embodiment, a job execution control device controls execution of a job for receiving input data, applying processing thereto, and outputting output data. The device includes a job executor, a controller, a recorder, and a generator. The job executor executes a job and record a job execution history including the input and output data. The controller displays, on a display, a job screen for presenting information on the executed job to a user and for receiving a user operation for re-executing the executed job. The recorder records a job operation history regarding the user operation through the job screen. The generator generates a warning condition for input or output data of a job based on the histories. The controller appends to the job screen warning information associated with an executed job the input or output data of which satisfies the warning condition, among executed jobs.

Abstract: An information processing device according to an embodiment includes a determination unit and a first training unit. The determination unit determines whether an unlabeled data point whose class label is unknown is a non-targeted data point that is not targeted for pattern recognition. The first training unit trains a first classifier for use in the pattern recognition through semi-supervised learning using a first training dataset including unlabeled data determined not to be the non-targeted data and not including unlabeled data determined to be the non-targeted data.

Abstract: A speech synthesis device according to an embodiment includes a speech synthesizing unit, a speaker parameter storing unit, an availability determining unit, and a speaker parameter control unit. Based on a speaker parameter value representing a set of values of parameters related to the speaker individuality, the speech synthesizing unit is capable of controlling the speaker individuality of synthesized speech. The speaker parameter storing unit is used to store already-registered speaker parameter values. Based on the result of comparing an input speaker parameter value with each already-registered speaker parameter value, the availability determining unit determines the availability of the input speaker parameter value. The speaker parameter control unit prohibits or restricts the use of the input speaker parameter value that is determined to be unavailable by the availability determining unit.

Abstract: An abnormality detection device according to an embodiment includes a detector, a remover, and a learner. The detector detects first abnormal data in detection target data which is an abnormality detection target by inputting the detection target data to a first autoencoder which performed learning based on first learning target data which is a learning target. The remover removes data associated with the first abnormal data from the first learning target data to generate second learning target data by inputting the first learning target data to a second autoencoder which performed learning based on the first abnormal data detected by the detector. The learner causes the first autoencoder to perform learning based on the second learning target data generated by the remover.

Abstract: An information processing apparatus according to one embodiment includes first and second processing circuitry. The first processing circuitry acquires traveling data including a plurality of features of a mobile apparatus based on a travelling state of the mobile apparatus. The second processing circuitry calculates a predicted value of consumed energy of the mobile apparatus based on a first classification rule and the traveling data, the first classification rule associating a plurality of first conditions with a plurality of first prediction models for the consumed energy used on satisfaction of the first conditions being associated with at least one of the plurality of features. The second processing circuitry determines whether update of the first classification rule is required based on a difference between the predicted value and an actual value. The second processing circuitry generates a second classification rule based on the traveling data if the update is required.

Abstract: A state analysis device includes a data classification unit, an analysis unit, and a determination unit. The data classification unit is configured to classify input data for each of a plurality of verification items associated with each of a plurality of control items. The analysis unit is configured to execute analysis associated with each combination of a plurality of control items and a plurality of verification items on the basis of data classified by the data classification unit. The determination unit is configured to comprehensively determine analysis results in units of combinations of the control items and the verification items on the basis of analysis results analyzed by the analysis unit.

Abstract: According to one embodiment, a train operation planning support system displays a timetable of a train in which a first required time and a first buffer time are placed for running time of the train and a second required time and a second buffer time are placed for dwell time of the train. The first required time is time required for a train running between a departure station and an arrival station. The first buffer time is buffer time obtained by subtracting the first required time from running time of the train set as a target value. The second required time is time required for the train stopping at the arrival station. The second buffer time is buffer time obtained by subtracting the second required time from dwell time of the train which is set as a target value.

Abstract: An information presentation device includes a plot unit, a histogram generation unit, and a visual information generation unit. The plot unit plots a plurality of values included in each of a plurality of pieces of multidimensional data on a plurality of axes in a parallel coordinate plot coordinate system. The histogram generation unit divides each of the plurality of axes into a plurality of sections, and generates a plurality of histogram bars each of which corresponds to the number of values plotted in each section in the plurality of sections. When one or more values are selected as selected values, the visual information generation unit generates visual information with which the ratio of the number of selected values to the total number of values contained in any one histogram bar out of the respective histogram bars generated by the histogram generation unit can be identified.

Abstract: An information presentation device according to an embodiment includes a contour extraction unit, an abstraction processing unit, and a contour correction unit. The contour extraction unit extracts a contour of each structural object included in data showing a layout of a plurality of structural objects. The abstraction processing unit abstracts a contour of each structural object extracted by the contour extraction unit and draws the abstracted contour on a plane grid surface in which grid lines in two directions orthogonal to each other are drawn. The contour correction unit corrects, among contour lines to constitute a contour abstracted by the abstraction processing unit, a contour line deviating from both of the grid lines in the two directions so as to match a grid line in at least one direction of the grid lines in the two directions.

Abstract: According to one embodiment, a voice synthesizing device includes a first operation receiving unit, a score transforming unit, and a voice synthesizing unit. The first operation receiving unit configured to receive a first operation specifying voice quality of a desired voice based on one or more upper level expressions indicating the voice quality. The score transforming unit configured to transform, based on a score transformation model that transforms a score of the upper level expression into a score of a lower level expression which is less abstract than the upper level expression, the score of the upper level expression corresponding to the first operation into a score of one or more lower level expressions. The voice synthesizing unit configured to generate a synthetic sound corresponding to a certain text based on the score of the lower level expression.

Abstract: A speaker retrieval device includes a first converting unit, a receiving unit, and a searching unit. The first converting unit converts, using an inverse transform model of a first conversion model for converting score vectors representing the features of voice quality into acoustic models, pre-registered acoustic models into score vectors; and registers the score vectors in a corresponding manner to a speaker identifier in score management information. The receiving unit receives input of a score vector. The searching unit searches the score management information for the speaker identifiers whose score vectors are similar to the received score vector.

Abstract: According to an embodiment, a recognition device includes a detector, a recognizer, and a matcher. The detector is configured to detect a character candidate from an input image. The recognizer is configured to generate recognition candidate from the character candidate. The matcher is configured to match the recognition candidate with a knowledge dictionary and contains modeled character strings to be recognized, and generate a matching result obtained by matching a character string presumed to be included in the input image with the dictionary. Any one of a real character code that represents a character and a virtual character code that specifies a command is assigned to an edge. The matcher gives, when shifting a state of the dictionary in accordance with an edge to which the virtual character code is assigned, a command specified by the virtual character code assigned to the edge to a command processor.

Abstract: A component management device according to an embodiment includes a database, a component searching unit, and a component operating unit. In the database, components are registered. Input-output of each component is defined for type of the component. The component searching unit receives input of the names, or the input examples, or the evaluation data of components; searches for the components registered in the database, according to the input; and presents the search result. The component operating unit sends the input from an application, which uses the searched component, to a component and sends the output of the component to the application.

Abstract: According to one embodiment, a recognition device includes storage and a processor. The storage is configured to store a first recognition model, a first data set, and tags, for each first recognition model. The processor is configured to acquire a second data set, execute recognition processing of the second recognition target data in the second data set by using the first recognition model, extract a significant tag of the tags stored in the storage in association with the first recognition model, based on the recognition processing result and the second correct data in the second data set, and create a second recognition model based on the acquired second data set and the first data set stored in the storage in association with the extracted tag.

Abstract: A device cooperation support system includes a user apparatus configured to receive an operation result output from an external device, to output a control signal for operating the external device, and to perform device cooperation control based on a device cooperation rule, and a device cooperation support apparatus configured to support device cooperation between the user apparatus and the external device. The device cooperation support apparatus receives registration of an external device from a device registrant, creates the device cooperation rule by combining condition clause information with execution clause information created on the basis of device operation definition information of the registered external device, and delivers the device cooperation rule to the selected user apparatus based on association between the selected user apparatus and the device cooperation rule.

Abstract: According to an embodiment, a dictionary updating apparatus includes a candidate extraction unit, a selection control unit, and a word registration unit. The candidate extraction unit extracts, based on a recognition result text obtained by a voice recognition engine performing a voice recognition processing using a word dictionary and a correction result text obtained by correcting at least a part of the recognition result text, candidates of words to be additionally registered in the word dictionary. The selection control unit generates a selection screen on which the extracted candidates are displayed selectably and, at the same time, information indicating the influence on the voice recognition processing at least when the candidates are additionally registered in the word dictionary is displayed, and accepts an operation of selecting the candidates displayed on the selection screen. The word registration unit registers additionally the candidates selected on the selection screen in the word dictionary.