Abstract: An information processing method includes acquiring a first prediction result by inputting evaluation data to a first model; determining an anomaly in the first prediction result based on the first prediction result and reference information; acquiring a second model based on the determination result; acquiring a second prediction result by inputting the evaluation data to the second model; determining an anomaly in the second prediction result based on the second prediction result and the reference information; acquiring a third model based on the determination result; acquiring a third prediction result by inputting the evaluation data to the third model; determining an anomaly in the third prediction result based on the third prediction result and the reference information; and if the anomaly in the third prediction result is recognized as being identical to the anomaly in the first prediction result, outputting information about a training limit of the first model.

Abstract: An information processing method includes the following executed by a computer: acquiring a first image and object data of an object appearing in the first image, extracting a portion of the first image that corresponds to a difference between the object data and an object detection result obtained by inputting the first image to a trained model, the trained model receiving an image as input to output an object detection result, acquiring a second image that includes a portion corresponding to the same object data as object data corresponding to the extracted portion of the first image, reflecting an image based on the extracted portion of the first image in the portion of the acquired second image that corresponds to the same object data, and generating training data for the trained model.

Abstract: An information processing method is executed by a computer and includes acquiring a first recognition result that is output as a result of inputting sensing data to a first recognition model trained through machine learning, acquiring reference data for the sensing data, determining a difference in class of a recognition target between the first recognition result and the reference data, generating an additional class for the first recognition model when the difference satisfies a predetermined condition, and outputting the sensing data or processed data obtained by processing the sensing data as training data for the additional class.

Abstract: An information processing method includes acquiring a first prediction result by inputting evaluation data to a first model; determining an anomaly in the first prediction result based on the first prediction result and reference information; acquiring a second model based on the determination result; acquiring a second prediction result by inputting the evaluation data to the second model; determining an anomaly in the second prediction result based on the second prediction result and the reference information; acquiring a third model based on the determination result; acquiring a third prediction result by inputting the evaluation data to the third model; determining an anomaly in the third prediction result based on the third prediction result and the reference information; and if the anomaly in the third prediction result is recognized as being identical to the anomaly in the first prediction result, outputting information about a training limit of the first model.

Abstract: A learning data creation method acquires items of still image data or moving image data as event data and non-event data, presents first data that is still image data or moving image data that is of a predetermined time before an event and included in the event data and still image data or moving image data, presents non-event data as second data, receives a judgment result as to whether the first data and second data are similar, (a) stores the event data as positive data, (b) stores the non-event data as positive data if the first data and second data are similar, and (c) stores the non-event data as negative data if the first data and second data are not similar.

Abstract: An orientation identification method for identifying an orientation of a device installed by being mounted on a moving body includes: obtaining a certain amount of acceleration in three mutually orthogonal directions detected by an acceleration sensor included in the device; and identifying the orientation of the device expressed in a coordinate system from acceleration data indicating the certain amount of acceleration obtained in the obtaining, according to movement characteristics indicated by statistics of acceleration during movement of the moving body expressed in the coordinate system, the coordinate system including a gravitational acceleration direction as an axis.

Abstract: In an information collection method of the present disclosure, a processor of an information collection system acquires event sensor data and identification information from an on-vehicle terminal or a passenger terminal, the event sensor data being output, in response to an event trigger, from an on-vehicle sensor mounted on the on-vehicle terminal or a passenger sensor mounted on the passenger terminal, the identification information identifying an output source of the event sensor data, identifies the passenger terminal associated with the identification information when the identification information indicates the on-vehicle terminal, identifies the on-vehicle terminal associated with the identification information when the identification information indicates the passenger terminal, requests sensor data from the identified passenger terminal or on-vehicle terminal, acquires the sensor data from a request destination, and stores the event sensor data and the sensor data in association with each other in a

Abstract: A contribution level determination method includes: obtaining, from a first storage device, records of annotation works for each of one or more units of target data for which the annotation works carried out by each of workers that use crowdsourcing have been completed; and calculating work contribution levels for each of worker IDs representing a different one of the workers, in accordance with a predetermined weighting rule stored in a second storage device, by referring to the records. In the calculating of the contribution levels, for each of the one or more units of target data, the work contribution level is calculated using a heavier weight for the worker ID representing a particular worker who carried out the annotation work initially than for the work contribution levels for the worker IDs representing the workers who carried out annotation works after the particular worker, in the order of the annotation works.

Abstract: In an information collection method of the present disclosure, a processor of an information collection system acquires event sensor data and identification information from an on-vehicle terminal or a passenger terminal, the event sensor data being output, in response to an event trigger, from an on-vehicle sensor mounted on the on-vehicle terminal or a passenger sensor mounted on the passenger terminal, the identification information identifying an output source of the event sensor data, identifies the passenger terminal associated with the identification information when the identification information indicates the on-vehicle terminal, identifies the on-vehicle terminal associated with the identification information when the identification information indicates the passenger terminal, requests sensor data from the identified passenger terminal or on-vehicle terminal, acquires the sensor data from a request destination, and stores the event sensor data and the sensor data in association with each other in a

Abstract: A learning data creation method acquires items of still image data or moving image data as event data and non-event data, presents first data that is still image data or moving image data that is of a predetermined time before an event and included in the event data and still image data or moving image data, presents non-event data as second data, receives a judgment result as to whether the first data and second data are similar, (a) stores the event data as positive data, (b) stores the non-event data as positive data if the first data and second data are similar, and (c) stores the non-event data as negative data if the first data and second data are not similar.

Abstract: An orientation identification method for identifying an orientation of a device installed by being mounted on a moving body includes: obtaining a certain amount of acceleration in three mutually orthogonal directions detected by an acceleration sensor included in the device; and identifying the orientation of the device expressed in a coordinate system from acceleration data indicating the certain amount of acceleration obtained in the obtaining, according to movement characteristics indicated by statistics of acceleration during movement of the moving body expressed in the coordinate system, the coordinate system including a gravitational acceleration direction as an axis.

Abstract: A method for predicting reactivity to parathyroid hormone by determining the genotype with respect to polymorphism in the parathyroid hormone receptor gene. The results obtained by the analysis of genetic polymophism are useful for predicting the tendency for a hemodialyzed patient to develop severe secondary hyperparathyroidism, or for predicting the tendency for a human individual to develop severe primary hyperparathyroidism. Also, this method is useful for predicting the therapeutic effects by predicting reactivity to parathyroid hormone in a patient who is being treated for regressive osteoporosis by administration of parathyroid hormone.