Abstract: A data processing method according to an embodiment acquires data including a plurality of records, divides the data based on external condition identification information such as a user ID to generate data sets Di for respective external conditions, divides each of the data sets Di based on label information indicating whether the record corresponds to a positive label indicating that a predetermined event has occurred or a negative label indicating that the predetermined event has not occurred to generate two data sets Di+ and Di? for the respective label information, generates difference data for a combination of a record included in one data set of the two data sets and a record included in the other data set, combines the generated difference data to generate integrated data Dnew, performs statistical analysis using Dnew, and outputs a result of performing the statistical analysis.

Abstract: Provided is an optical modulator manufacturing method capable of determining the quality of an optical modulator having MMI waveguides and realizing improvement in yield during manufacturing. Here, in waveguide fabrication processes, hard mask material deposition, soft mask material application, exposure, and hard mask fabrication are executed, and then in hard mask width length measurement, the hard mask width for fabricating the MMI waveguide is measured at one or more locations. In hard mask width quality determination based on machine learning results, the quality of optical characteristics of the chip is predicted and determined in advance, based on sample data created in advance by analyzing a relationship between the hard mask width and optical characteristics of the optical modulator, depending on whether the hard mask width is present in a permissible range of the sample data. Depending on the result of the above-mentioned determination the mask fabrication is redone.

Abstract: A learning apparatus according to an embodiment includes: input means for inputting training data for learning a classifier and a causal graph representing causal relationships between variables included in the training data; and learning means for learning the classifier by solving a constrained optimization problem in which a mean of causal effects between predetermined variables is within a predetermined range and a variance of the causal effects is equal to or smaller than a predetermined value, using the training data and the causal graph input by the input means.

Abstract: A time-series-data feature extraction device includes: a data processing unit that processes a received unevenly spaced time-series-data group into an evenly spaced time-series-data group including omissions and an omission information group indicating presence or absence of omissions, based on a received input time-series data length and a received minimum observation interval; a model learning unit that learns a weight vector of each layer of a model with a difference between an element not missing in a matrix of the evenly spaced time-series-data group including omissions and an element of an output result of an output layer of the model being taken as an error, and stores the weight vector as a model parameter in a storage unit, the difference being; and a feature extraction unit that receives time-series data of a feature extraction target, calculates a value of the intermediate layer of the model with use of the model parameter stored in the storage unit by inputting the received time-series data of the

Abstract: Assuming that a total number of N selection subjects n (n=1, . . . , N) each select any one of a total number of M selection targets m (m=1, . . . , M), parameter estimation means: inputs acceptable limits ?n,m of each selection subject n and calculates patience ?n and a preference vector ?n, the acceptable limits ?n,m indicating limits of congestion degrees of respective selection subjects m that are acceptable to the selection subject n, the patience ?n indicating the largest value of the acceptable limits ?n,m of the selection subject n with respect to the selection targets m, the preference vector ?n indicating preference of the selection subject n when selecting the selection targets m; and estimates parameters of a model for obtaining acceptable limits ?i,m of each of a total number of I(>N) selection subjects i (i=1, . . . , I) by using the calculated patience ?n and the calculated preference vector ?n.

Abstract: A technique for estimating a causal relation, which can solve problems of a conventional technique and which does not require preliminary setting of a regression model is disclosed. An embodiment of the present invention relates to a causal relation learning device including: a feature value calculation unit that receives a correct label of three or more classification labels related to a causal relation of time-series data and time-series data corresponding to the correct label and calculates a feature value of the time-series data; and a classifier learning unit that learns a classifier using a set of the feature value and the correct label so that an output of the classifier with respect to the feature value is a largest value of an output value of the correct label.

Abstract: Statistical analysis taking into account potential features that affect an occurrence of a predetermined event is enabled. A data processing method according to an embodiment acquires data including a plurality of records, divides the data based on external condition identification information such as a user ID to generate data sets Di for respective external conditions, divides each of the data sets Di based on label information indicating whether the record corresponds to a positive label indicating that a predetermined event has occurred or a negative label indicating that the predetermined event has not occurred to generate two data sets Di+ and Di? for the respective label information, generates difference data for a combination of a record included in one data set of the two data sets and a record included in the other data set, combines the generated difference data to generate integrated data Dnew, performs statistical analysis using Dnew, and outputs a result of performing the statistical analysis.

Abstract: A time-series-data feature extraction device includes: a data processing unit that processes a received unevenly spaced time-series-data group into an evenly spaced time-series-data group including omissions and an omission information group indicating presence or absence of omissions, based on a received input time-series data length and a received minimum observation interval; a model learning unit that learns a weight vector of each layer of a model with a difference between an element not missing in a matrix of the evenly spaced time-series-data group including omissions and an element of an output result of an output layer of the model being taken as an error, and stores the weight vector as a model parameter in a storage unit, the difference being; and a feature extraction unit that receives time-series data of a feature extraction target, calculates a value of the intermediate layer of the model with use of the model parameter stored in the storage unit by inputting the received time-series data of the