Abstract: In an embodiment of this invention, in a learning phase, a state estimation device acquires activity state data and biometric data at that time from user terminals of a plurality of users, generates a regression formula representing the relationship between the biometric data and the activity state data using a regression analysis method on the basis of these pieces of measurement data, and calculates a difference between the coefficients of the regression formula of all users and each user to generate a coefficient correction regression formula representing a relationship between the difference of the coefficient and an average value of the biometric data.

Abstract: According to one embodiment of the present invention, food item information on a plurality of food items suitable for snacking classified into a plurality of clusters according to similarity in feature quantities for the shape, taste, and texture is stored. Calories less than the difference between the sum of the calorie consumption by the user during a first period before user-specified desired snacking time and expected calorie consumption during a second period thereafter and the sum of calorie intake during the first period and expected calorie intake during the second period by the user are obtained as ingestible snacking calories, and a food item having calories less than the ingestible snacking calories and belonging to a cluster other than a cluster selected for the previous snacking is selected from the food item information and presented to the user.

Abstract: A monetary risk prediction apparatus according to an embodiment includes: a predictive model recording unit that records a predictive model to predict time-series data showing a future asset amount of a user; and prediction means for receiving evaluation data that is time-series data including an asset amount and a numeric value showing a health condition of the user, inputting the evaluation data to the predictive model recorded on the predictive model recording unit, and outputting the time-series data showing the future asset amount of the user predicted by the predictive model according to the input.

Abstract: A data processing device that makes effective use of a data group containing missing data is provided. A series of learning data containing missing data is acquired, and a representative value of data and a validity ratio representing a proportion of valid data being present are calculated from the series of learning data according to a predefined unit of aggregation. Then, learning of an estimation model is performed so as to minimize an error which is based on a difference between an output resulting from inputting the representative value and the validity ratio to the estimation model, and the representative value. Also, a series of estimation data containing missing data is acquired, and a representative value of data and a validity ratio representing a proportion of valid data being present are calculated from the series of estimation data according to a predefined unit of aggregation.

Abstract: A data processing apparatus according to a first aspect of the present invention includes: a first generation section that generates first input data in which first data related to a first phenomenon and second data related to a second phenomenon that is relevant to the first phenomenon are combined with first auxiliary data that is based on a missing data status in at least one of the first data and the second data; and a learning section that learns a model parameter of a prediction model, based on an error according to the first auxiliary data between output data outputted from the prediction model when the first input data is inputted into the prediction model, and each of the first data and the second data.

Abstract: An object of an aspect of the present invention is to enable estimation of more effective intervention content in order to make a person's health state approximate to an ideal health state, and in a learning phase, measurement values and target values of a health state for a plurality of days in the past are sequentially input to a learning machine configured by a multilayer neural network, and the learning machine is caused to perform learning such that a target achievement expectation value obtained by using a success rate that allows the user's health state to approximate to an ideal health state, continuity that allows the health state approximate to the ideal health state to be maintained, and a target value of the health state to be subsequently recommended and the target achievement expectation value thereof that reflect a temporal change in the health state and a history of interventions until a present time are output.

Abstract: According to one embodiment of the present invention, sets of medical record data that have a common development order of diseases to be focused on and different times until the diseases develop are selected from medical record data, a feature indicating a health state of a user is extracted from each piece of medical record data constituting the set for each of the sets of the medical record data, the extracted feature is set as training data, a risk score for a co-occurrence or an occurrence of a complication of each of the diseases is calculated based on examination data of a first-year examination and a time until each of the diseases occur, and the risk score is set as correct answer data. At this time, the development risk score is calculated such that a user having a short elapsed time until development has a larger value than a user having a long elapsed time until development.

Abstract: In an embodiment of this invention, in a learning phase, a state estimation device acquires activity state data and biometric data at that time from user terminals of a plurality of users, generates a regression formula representing the relationship between the biometric data and the activity state data using a regression analysis method on the basis of these pieces of measurement data, and calculates a difference between the coefficients of the regression formula of all users and each user to generate a coefficient correction regression formula representing a relationship between the difference of the coefficient and an average value of the biometric data.