Data acquisition device

The data acquisition device addresses the challenge of acquiring balanced learning data by using a psychological state estimation model and determination unit to assess and label user data, enhancing the accuracy of psychological state estimation models.

WO2024237337A1PCT designated stage expired Publication Date: 2024-11-21NTT DOCOMO INC
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2024/018353
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-18
Filing Date
2024-05-17
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing technologies face challenges in acquiring balanced learning data for estimating psychological states, particularly struggling to obtain data from individuals in poor psychological states, leading to biased models.

Method used

A data acquisition device that includes a psychological state estimation model, information acquisition, estimation, determination, label creation, and data storage units, which acquires and labels user data based on test answers to ensure necessary learning data is collected efficiently.

Benefits of technology

The device effectively acquires learning data by determining the need for correct labels based on user responses, reducing bias and improving the accuracy of psychological state estimation models.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2024018353_21112024_PF_FP_ABST
    Figure JP2024018353_21112024_PF_FP_ABST
Patent Text Reader

Abstract

This data acquisition device comprises: an information acquisition unit that acquires psychological reflection information of a user; an estimation unit that estimates the psychological state of the user by using a psychological-state estimation model for which the acquired psychological reflection information of the user serves as input information; a determination unit that determines whether acquisition of a correct-answer label corresponding to the acquired psychological reflection information of the user is necessary; and a label creation unit that, when it is determined by the determination unit that acquisition of the correct-answer label is necessary, creates a correct-answer label corresponding to the psychological reflection information of the user on the basis of the user's answer to a test presented to the user.
Need to check novelty before this filing date? Find Prior Art

Description

Data Acquisition Equipment

[0001] The present disclosure relates to data acquisition devices.

[0002] Patent Literature 1 discloses a technology for checking and managing the mental state of employees. This technology uses a trained model to determine whether an employee is at risk of mental illness based on the employee's work situation.

[0003] Japanese Patent Application Laid-Open No. 2022-37892

[0004] When constructing a learning model that estimates a user's psychological state from user information using machine learning, learning data (teacher data) corresponding to each psychological state is required. For example, by having an information provider provide biometric information or the like and identifying the information provider's psychological state through a predetermined test or the like, learning data that is a set of user information and psychological state can be obtained. Learning data can be obtained from information providers in each psychological state, but it is difficult to obtain learning data in the same way for each psychological state. In particular, people who are cooperative in providing learning data often have a good psychological state, and it is difficult to obtain learning data that corresponds to a poor psychological state.

[0005] The present disclosure has been made in view of the above, and aims to provide a data acquisition device that can acquire necessary learning data.

[0006] A data acquisition device according to the present disclosure includes a psychological state estimation model, an information acquisition unit, an estimation unit, a determination unit, a label creation unit, and a data storage unit. The psychological state estimation model is trained using psychological reflection information and a correct label indicating a psychological state corresponding to the psychological reflection information as training data. The information acquisition unit acquires a user's psychological reflection information. The estimation unit estimates the user's psychological state using the psychological state estimation model with the acquired user's psychological reflection information as input information. The determination unit determines whether or not it is necessary to acquire a correct label corresponding to the acquired user's psychological reflection information. When the determination unit determines that it is necessary to acquire a correct label, the label creation unit creates a correct label corresponding to the user's psychological reflection information based on the user's answer to a test presented to the user. The data storage unit stores the user's psychological reflection information and the user's correct label as a set of data.

[0007] In the data acquisition device, the psychological state estimation model estimates the user's psychological state using the user's psychological reflection information as input data. The data acquisition device acquires training data by administering a predetermined test to the user and creating a correct answer label corresponding to the user's psychological reflection information based on the obtained test answers. Such a data acquisition device has a determination unit that determines whether or not it is necessary to acquire a correct answer label corresponding to the acquired user's psychological reflection information, so that it can acquire training data as needed.

[0008] According to the present disclosure, it is possible to provide a data acquisition device that can acquire necessary learning data.

[0009] FIG. 1 is a diagram illustrating a configuration of a data acquisition device according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating a configuration of an example determination unit. FIG. 3 is a flowchart illustrating an example of operation of the determination unit. FIG. 4 is a flowchart illustrating another example of operation of the determination unit. FIG. 5 is a flowchart illustrating yet another example of operation of the determination unit. FIG. 6 is a flowchart illustrating yet another example of operation of the determination unit. FIG. 7 is a flowchart illustrating yet another example of operation of the determination unit. FIG. 8 is a flowchart illustrating yet another example of operation of the determination unit. FIG. 9 is a flowchart illustrating an example of operation in the data acquisition device. FIG. 10 is a diagram illustrating a hardware configuration of an example of a data acquisition device.

[0010] Hereinafter, an embodiment of a data acquisition device according to the present disclosure will be described in detail with reference to the drawings. In the description of the drawings, the same elements are given the same reference numerals, and duplicated explanations will be omitted.

[0011] Fig. 1 is a diagram showing the configuration of a data acquisition system 1 including an example data acquisition device 10. The example data acquisition system 1 includes a terminal device 5 and the data acquisition device 10. For example, the data acquisition device 10 is configured to be able to communicate with the terminal device 5 via a communication network N. Note that although Fig. 1 shows only one terminal device 5, the data acquisition device 10 may be able to be connected to multiple terminal devices 5 simultaneously.

[0012] The terminal device 5 is, for example, a communication device operated by a user. The terminal device 5 receives input, transmits the received information to the data acquisition device 10, and displays information received from the data acquisition device 10. The terminal device 5 may be, for example, a mobile terminal such as a highly functional mobile phone (smartphone), a mobile phone, or a personal digital assistant (PDA). Note that the devices constituting the terminal device 5 are not limited to these.

[0013] The terminal device 5 can transmit the psychological reflection information acquired from the user to the data acquisition device 10 in association with the user's identification information stored in the terminal device 5, etc. The psychological reflection information includes various data reflecting the user's psychological state. The psychological reflection information may include the user's biometric information. The biometric information is information indicating the state of a living organism and may be acquired from the living organism depending on the state at any given time. The biometric information may be acquired by the terminal device 5. For example, the terminal device 5 may include a sensor for acquiring biometric information, may be connected to an external sensor, or may acquire data acquired by an external sensor.

[0014] An example of the biometric information may be at least one of audio information, video information, and other detected information. The audio information may be sound data including a voice uttered by the user. In an example, the audio information may be acquired by a microphone provided in the terminal device 5. The video information may be video data including a video of the user. The video information may include information such as the user's facial expression, eye movements, and gestures. The video information may be acquired by a camera provided in the terminal device 5. The other detected information as biometric information may be, for example, detected data related to the user's heart rate, pulse, blood flow, body temperature, sweating, breathing, brain waves, user movements, posture, and the like.

[0015] The data acquisition device 10 has an aspect as an estimation device that estimates the psychological state of a user who uses the terminal device 5. The data acquisition device 10 also has an aspect as an acquisition device that acquires training data that can be used for a machine learning model used in the estimation device. As shown in Fig. 1 , the exemplary data acquisition device 10 includes, as its functional units, an information acquisition unit 11, an estimation unit 12, a determination unit 13, a label creation unit 15, and a data storage unit 16.

[0016] The information acquisition unit 11 acquires user's psychological reflection information. As described above, the user's psychological reflection information may be transmitted to the data acquisition device 10 from the terminal device 5 used by the user. The information acquisition unit 11 may acquire the user's psychological reflection information in association with the user's identification information. In one example, the information acquisition unit 11 may acquire the user's identification information, psychological reflection information, and reception date and time (time information) in association with each other. Furthermore, the information acquisition unit 11 may acquire data indicating the user's attributes in association with the user's identification information. The attributes may be characteristics of each user for classifying individuals without identifying them. The attributes may be, for example, gender, age, household composition, lifestyle habits, temperament, attendance information, etc. The attribute data may be stored in advance on the terminal device 5.

[0017] The estimation unit 12 estimates the psychological state of the user using the psychological reflection information of the user acquired by the information acquisition unit 11 as input information. As shown in FIG. 1 , an example of the estimation unit 12 includes a first estimation engine 12a having a psychological state estimation model 12b. The psychological state estimation model 12b may be a machine learning model trained using the psychological reflection information and a ground truth label indicating the psychological state corresponding to the psychological reflection information as training data. The machine learning framework may be, for example, any of regression, binary classification, anomaly detection, change detection, multi-class classification, etc. In this embodiment, an example will be described in which the framework of the psychological state estimation model 12b is regression.

[0018] An example of training data used for training the psychological state estimation model 12b includes a data set consisting of data indicating psychological reflection information and correct labels that quantify the psychological state of the provider of the psychological reflection information. When the psychological reflection information is audio data, video data, or the like, the psychological reflection information may be acquired by the terminal device 5. The correct labels are acquired based on the answers to a predetermined test given to the provider at the same time as the acquisition of the psychological reflection information or promptly after the acquisition of the psychological reflection information.

[0019] This test may be composed of, for example, multiple questions for quantifying the degree of psychological stress response. That is, in this test, a measurement value that quantifies the psychological state of the subject is determined. One example of a question is composed of a question addressed to the subject and multiple answer options for the question. After reading the question, the subject can select their own answer from the options. For example, a score may be assigned to each option, and the measurement value may be determined based on the score of the option selected as the answer. The correct answer label that constitutes the learning data may be a numerical value obtained from the results of the above-mentioned test. In the learning model of this embodiment, a test in which the measurement value increases as stress increases is used. Set data composed of psychological reflection information and correct answer data may be associated with the attributes of the provider and stored in the data storage unit 16.

[0020] The first estimation engine 12a of the estimation unit 12 inputs the user's psychological reflection information into the trained psychological state estimation model 12b, thereby acquiring a numerical value (measured value) indicating the user's psychological state output from the psychological state estimation model 12b. The first estimation engine 12a classifies the user's psychological state into a first category and a second category. That is, the first estimation engine 12a classifies the acquired numerical value using a first threshold, thereby classifying the user's psychological state into a bad (first category) and a good (second category).

[0021] The estimation unit 12 estimates that the user's psychological state is bad when the numerical value output from the psychological state estimation model 12b is equal to or greater than a predetermined threshold. Furthermore, the estimation unit 12 estimates that the user's psychological state is good when the numerical value output from the psychological state estimation model 12b is less than a predetermined threshold. The estimation unit 12 may transmit the estimation result to the user's terminal device 5. The transmitted estimation result may be only a classification (good, bad) indicating the psychological state, or may include an estimated measurement value. The estimation unit 12 may store the estimated measurement value and classification in association with the user's identification information. That is, the estimation unit 12 may store previously estimated psychological states for each user.

[0022] The determination unit 13 determines whether or not it is necessary to obtain a correct label corresponding to the psychology reflecting information obtained by the information obtaining unit 11. For example, in order to contribute to improving the accuracy of the psychological state estimation model, the determination unit 13 determines whether or not it is necessary to obtain a correct label based on the content of the psychology reflecting information, the state of the psychology reflecting information, the user's usage status, the state of the learning data being used, etc. Details of the determination unit 13 will be described later.

[0023] When the determination unit 13 determines that it is necessary to obtain a correct label, the label creation unit 15 creates a correct label corresponding to the obtained psychological reflection information of the user. The correct label is created based on the user's answer to a predetermined test presented to the user. When the determination unit 13 determines that it is necessary to obtain a correct label, the test is presented on the display of the user's terminal device 5. In one example, the label creation unit 15 may transmit the test to the terminal device 5 of the user who needs to obtain a correct label. Note that this test may have the same content as the test used when generating the training data used to train the psychological state estimation model.

[0024] The data storage unit 16 stores the user's psychological reflection information acquired by the information acquisition unit 11 and the correct label created by the label creation unit 15 as a set of data. This data may be stored in the data storage unit 16 in association with the user's attributes. The psychological state estimation model 12b used by the estimation unit 12 can be updated with the set data newly stored in the data storage unit 16 as additional learning data.

[0025] Next, a detailed description will be given of the determination unit 13. An example of the determination unit 13 may be any one of a first determination unit 13A, a second determination unit 13B, a third determination unit 13C, a fourth determination unit 13D, a fifth determination unit 13E, and a sixth determination unit 13F illustrated below.

[0026] Fig. 2 is a block diagram showing the functions of the first determination unit 13A. Fig. 3 is a flowchart showing the operation of the first determination unit 13A. The first determination unit 13A has a second estimation engine 13Aa having a psychological state estimation model 13Ab. The psychological state estimation model 13Ab included in the second estimation engine 13Aa may be the same learning model as the psychological state estimation model 12b included in the estimation unit 12. In the first determination unit 13A, the second estimation engine 13Aa uses the psychological state estimation model 13Ab to obtain a second psychological state estimation result (step S11). The second psychological state estimation result has a higher rate of being determined to be unwell compared to the psychological state estimated by the estimation unit 12.

[0027] For example, the second estimation engine 13Aa may classify the user's psychological state into a third category and a fourth category. The third category may be broader than the first category, which indicates that the user's psychological state is poor, and may include the first category. The fourth category may be narrower than the second category, which indicates that the user's psychological state is good, and may be a category obtained by subtracting the overlap between the second category and the third category from the second category. The second estimation engine 13Aa classifies the user's psychological state into the third category and the fourth category by dividing the measurement values ​​estimated by the psychological state estimation model 13Ab using a second threshold different from the first threshold. In this case, psychology-reflecting information indicating measurement values ​​near the threshold separating the first category and the second category is classified into the third category. That is, users whose psychological state is poor according to the first estimation engine 12a and users whose measurement values ​​are high among users whose psychological state is good according to the first estimation engine 12a are classified into the third category. The first determination unit 13A determines whether or not it is necessary to obtain a correct label corresponding to the user's psychology-reflecting information based on the estimation result of the second estimation engine 13Aa (step S12). When the second estimation engine 13Aa estimates that the psychology-reflecting information belongs to the third category, the first determination unit 13A determines that it is necessary to obtain a correct label.

[0028] FIG. 4 is a flowchart showing the operation of the second determination unit 13B. The second determination unit 13B determines whether or not it is necessary to obtain a correct label corresponding to the acquired user's psychology-reflecting information based on the user's psychological state history. First, the second determination unit 13B obtains the user's psychological state history stored by the estimation unit 12 (step S21). For example, the second determination unit 13B may obtain the user's psychological state history obtained by the estimation unit 12 over a certain period of time in the past. Next, the second determination unit 13B performs statistical processing on the user's past psychological states based on the obtained psychological state history (step S22). Then, the second determination unit 13B determines whether or not it is necessary to obtain a correct label corresponding to the user's psychology-reflecting information based on the result of the aggregation processing (step S23).

[0029] For example, the second determination unit 13B may acquire the percentage of times the user's psychological state is estimated to be poor based on the acquired psychological state history. In this case, when the percentage of times the user's psychological state is estimated to be poor is equal to or greater than a predetermined threshold, it may determine that acquisition of a correct label is necessary. Note that the second determination unit 13B may further perform a determination by the first determination unit 13A when the percentage of times the user's psychological state is estimated to be poor is equal to or greater than the predetermined threshold. In other words, the second determination unit 13B may determine that acquisition of a correct label is necessary when the percentage of times the user's psychological state is poor in the user's past estimation results is equal to or greater than a predetermined percentage and the current estimation result by the first determination unit 13A is in the third category.

[0030] 5 is a flowchart showing the operation of the third determination unit 13C. The third determination unit 13C determines whether or not it is necessary to obtain a correct label based on the distribution of data already stored in the data storage unit 16. For example, if the proportion of a specific category that constitutes the multiple categories in the data already stored in the data storage unit 16 is equal to or less than a predetermined value, and the user's psychological state estimated using the psychological state estimation model with the user's psychological reflection information as input information corresponds to the specific category, the third determination unit 13C determines that it is necessary to obtain a correct label corresponding to the user's psychological reflection information.

[0031] In one example, the third determination unit 13C first acquires the learning data stored in the data storage unit 16 (step S31). Next, the third determination unit 13C statistically processes the learning data (step S32) and, based on the results of the aggregation process, determines whether or not it is necessary to acquire a correct label corresponding to the user's psychological reflection information (step S33). For example, the third determination unit 13C calculates the proportion of defective data based on the classification (good, bad) associated with the acquired learning data. Then, when the proportion of data indicating defective data is equal to or less than a predetermined threshold, the third determination unit 13C may determine that it is necessary to acquire a correct label. Note that, when the proportion of defective data is equal to or less than the predetermined threshold, the third determination unit 13C further performs a determination by the first determination unit 13A or the second determination unit 13B and determines whether or not it is necessary to acquire a correct label based on the result.

[0032] FIG. 6 is a flowchart illustrating the operation of the fourth determination unit 13D. The fourth determination unit 13D evaluates the quality of the user's biometric information and determines whether or not it is necessary to acquire a correct label corresponding to the user's psychological information based on the evaluation result. First, the fourth determination unit 13D acquires the biometric information constituting the user's psychological information (step S41). Next, the fourth determination unit 13D determines the quality of the biometric information (step S42). The quality of the biometric information may be an index of whether the data constituting the biometric information can contribute to improving the learning accuracy of the psychological state estimation model. For example, the quality of the biometric information may be the signal-to-noise ratio of audio data, video data, etc. The fourth determination unit 13D determines whether or not it is necessary to acquire a correct label based on the determination result of the quality of the biometric information (step S43). For example, the fourth determination unit 13D determines that it is necessary to acquire a correct label when the quality of the biometric information exceeds a predetermined standard, and determines that it is not necessary to acquire a correct label when the quality of the biometric information falls below the predetermined standard.

[0033] FIG. 7 is a flowchart illustrating the operation of the fifth determination unit 13E. When a user's psychology-reflecting information is acquired, the fifth determination unit 13E determines that a correct label corresponding to the user's psychology-reflecting information needs to be acquired if a predetermined time has elapsed since the user's previous answer to a test for obtaining a correct label. For example, when the information acquisition unit 11 acquires the user's psychology-reflecting information, the fifth determination unit 13E stores the user's identification information and the acquisition date (step S51). The fifth determination unit 13E also stores, in association with the acquisition date, whether or not the user has answered the test for obtaining a correct label. The user's identification information and the acquisition date may be associated with the psychology-reflecting information and stored as a user history in the fifth determination unit 13E. When the user's psychology-reflecting information is acquired, the fifth determination unit 13E refers to the user history and acquires the answer date of the most recent test for the user for whom the psychology-reflecting information was acquired (step S52). Then, the fifth determination unit 13E calculates the number of days that have passed since the most recent acquisition date until the current acquisition date of the user's psychology reflecting information (step S53), and determines whether or not it is necessary to acquire a correct label based on this number of days that have passed (step S54). For example, the fifth determination unit 13E determines that it is necessary to acquire a correct label if the number of days that have passed is equal to or greater than a predetermined period, and determines that it is not necessary to acquire a correct label if the number of days that have passed is less than the predetermined period.

[0034] FIG. 8 is a flowchart illustrating the operation of the sixth determination unit 13F. The sixth determination unit 13F determines whether or not it is necessary to obtain a correct label based on the distribution of data already stored in the data storage unit 16. For example, if a user's attribute is rare, the sixth determination unit 13F determines that it is necessary to obtain a correct label corresponding to the user's psychological reflection information. First, the sixth determination unit 13F obtains the user's attributes based on the user's identification information (step S61). The attributes may be stored in the data storage unit 16 in association with the identification information. Next, the sixth determination unit 13F calculates the proportion of each attribute in the training data (including additional training data) stored in the data storage unit 16 (step S62). Next, the sixth determination unit 13F obtains the proportion of the user's attributes in the training data based on the calculated proportion of each attribute (step S63). The sixth determination unit 13F determines whether or not it is necessary to obtain a correct label based on the proportion of the user's attributes (step S64). For example, the sixth judgment unit 13F may determine that it is necessary to obtain a correct label when the proportion of the user's attributes is less than a predetermined threshold, and may determine that it is not necessary to obtain a correct label when the proportion of the user's attributes is greater than or equal to a predetermined threshold.

[0035] FIG. 9 is a flowchart illustrating an example of the operation of the data acquisition system 1. As shown in FIG. 9, first, psychological reflection information is output from the terminal device 5 used by the user to the data acquisition device 10 (step S101). Next, the information acquisition unit 11 acquires the psychological reflection information output from the terminal device 5 along with the user's identification information (step S102). The estimation unit 12 inputs the psychological reflection information acquired by the information acquisition unit 11 into the psychological state estimation model 12b to acquire an estimation result of the user's psychological state (step S103). The estimation unit 12 outputs the acquired estimation result to the terminal device 5 (step S104). The terminal device 5 acquires the estimation result output from the estimation unit 12 and displays the estimation result on a screen or the like (step S105). Note that steps S104 and S105 may be omitted.

[0036] The determination unit 13 determines whether or not it is necessary to obtain a correct label for the psychology reflecting information transmitted by the user (steps S106 and S107). The determination unit 13 may be any one or a combination of the above-described determination units 13A to 13F. If it is determined that it is not necessary to obtain a correct label, the processing of the data processing device may be terminated. On the other hand, if it is determined that it is necessary to obtain a correct label, the label creation unit 15 outputs the test to the terminal device 5 (step S108). Note that if information related to the test has been stored in advance in the terminal device 5, the label creation unit 15 may output a command to the terminal device 5 to display the test.

[0037] Next, the terminal device 5 displays the test on the screen (step S109), allowing the user to answer the test. After the user answers the test, the terminal device 5 outputs the answer results to the data acquisition device 10 (step S110). The information acquisition unit 11 acquires the answer results (step S111), and the label creation unit 15 creates a correct label based on the answer results (step S112). The correct label is a value that quantifies the degree of psychological stress response. The correct label is determined based on the test answers. The created correct label is stored in the data storage unit 16 as learning data together with the user's psychological reflection information and user attributes (step S113). The newly stored learning data in the data storage unit 16 can be used to update (additional learning) the psychological state estimation model. In addition, the user's psychological state (good, bad) based on the created correct label may be transmitted to the terminal device 5.

[0038] As described above, the example data acquisition device 10 includes a psychological state estimation model 12b that has been trained using psychological reflection information and a correct label indicating a psychological state corresponding to the psychological reflection information as training data, an information acquisition unit 11 that acquires the user's psychological reflection information, an estimation unit 12 that estimates the user's psychological state using the psychological state estimation model 12b with the acquired user's psychological reflection information as input information, a determination unit 13 that determines whether or not it is necessary to acquire a correct label corresponding to the acquired user's psychological reflection information, a label creation unit 15 that creates a correct label corresponding to the user's psychological reflection information based on the user's answers to a test presented to the user when the determination unit 13 determines that it is necessary to acquire a correct label, and a data storage unit 16 that stores the user's psychological reflection information and the user's correct label as a set of data.

[0039] When constructing a learning model that estimates a user's psychological state using machine learning, learning data corresponding to the psychological state is required. In one example, learning data generated with the cooperation of providers is used to construct the learning model. While learning data can be obtained from providers in each psychological state, it is difficult to obtain learning data in the same manner for each psychological state. In particular, providers who cooperate in generating learning data rarely have a poor psychological state, making it difficult to obtain learning data corresponding to a poor psychological state. In this case, a bias occurs between the number of learning data corresponding to a good state and the number of learning data corresponding to a poor state, which may make it difficult to accurately estimate the psychological state.

[0040] In the data acquisition device 10, the psychological state estimation model estimates the user's psychological state using the user's psychological reflection information as input data. The data acquisition device 10 can acquire training data by administering a predetermined test to the user and creating a correct label corresponding to the user's psychological reflection information based on the obtained test answers. This data acquisition device 10 includes a determination unit 13 that determines whether or not it is necessary to acquire a correct label corresponding to the acquired user's psychological reflection information, thereby enabling it to acquire training data as needed. In other words, if the estimated psychological state is poor, it is determined that acquisition of a correct label is necessary, thereby enabling efficient acquisition of training data corresponding to the poor state that is difficult to acquire. Furthermore, if acquisition of a correct label is not necessary, the user's test responses are not required, thereby reducing excessive burden on the user.

[0041] As an example, the psychological state estimation model may be updated using the learning data stored in the data storage unit 16. In this configuration, the learning data including the psychological reflection information determined to be necessary by the determination unit 13 is reflected in the psychological state estimation model, so that the psychological state estimation model can be updated according to the determination criterion of the determination unit 13.

[0042] The determination unit 13, for example, may determine whether or not it is necessary to obtain a correct label based on the distribution of data already stored in the data storage unit 16. In this configuration, for example, training data that is relatively in short supply can be efficiently obtained according to the distribution of the training data stored in the data storage unit 16.

[0043] In one example, the correct labels constituting the training data may be classifiable into a plurality of categories, and the determination unit 13 may determine that it is necessary to acquire a correct label corresponding to the user's psychology reflection information when the proportion of specific sections constituting the plurality of categories in the data stored in the data storage unit 16 is equal to or less than a predetermined value and the user's psychology state estimated using the psychology state estimation model with the user's psychology reflection information as input information corresponds to a specific category. With this configuration, when there is a category for which it is expected in advance that it will be difficult to collect data, training data corresponding to that category can be efficiently acquired.

[0044] In one example, the information acquisition unit 11 may acquire user attributes, the data storage unit 16 may store the attributes along with the data, and the determination unit 13 may calculate the proportion of each attribute in the data stored in the data storage unit 16, and determine that it is necessary to acquire a correct label corresponding to the user's psychological reflection information when the proportion of the user's attributes acquired by the information acquisition unit 11 is equal to or less than a predetermined proportion. This configuration makes it possible to efficiently acquire learning data corresponding to users with relatively few attributes. It is also possible to reduce the burden on users with attributes for which sufficient learning data has been accumulated.

[0045] In one example, the estimation unit 12 may classify the user's psychological state into a first category and a second category, and the determination unit 13 may classify the user's state into a third category that is broader than the first category and encompasses the first category, and a fourth category that is the second category minus the third category, and may determine that it is necessary to obtain a correct label corresponding to the user's psychological reflection information when the user's state is classified into the third category. In this configuration, psychological reflection information that has the potential to be classified into the first category from the psychological reflection information classified into the second category can be efficiently obtained.

[0046] In one example, the estimation unit 12 may store previously estimated psychological states for each user, and the determination unit 13 may determine whether or not it is necessary to acquire a correct label corresponding to the acquired psychological reflection information of the user, based on the history of the user's psychological state stored by the estimation unit 12. For example, if it is found that the rate of poor health is high as a result of referring to the history of the user's psychological state, highly reliable data can be obtained by acquiring psychological reflection information at the time when the user's psychological state was estimated to be poor health.

[0047] In one example, the information acquisition unit 11 may acquire biometric information of the user as the psychological reflection information, and the determination unit 13 may evaluate the quality of the biometric information of the user and determine that it is unnecessary to acquire a correct label corresponding to the psychological reflection information of the user if the quality of the biometric information of the user is below a predetermined standard. In this configuration, low-quality data is not acquired, thereby suppressing a decrease in the estimation accuracy of the psychological state estimation model.

[0048] In one example, the determination unit 13 may determine, when the user's psychology-reflecting information is acquired, that it is necessary to acquire a correct label corresponding to the user's psychology-reflecting information if a predetermined time has elapsed since the user's last response to a test for acquiring a correct label. If a long period of time has elapsed since the user last responded to a test for acquiring a correct label, the reliability of the correct label may not necessarily be considered high. This configuration can prevent a decrease in the reliability of the correct label. Furthermore, because the determination of whether or not the correct label needs to be updated is based on the elapsed time since the last correct label was acquired, the burden of answering questions can be reduced for users for whom a certain period of time has not yet elapsed since the last correct label was acquired.

[0049] Although one embodiment has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment. For example, the determination unit 13 may be a combination of the above-mentioned determination units 13. For example, the determination unit 13 may be a combination of any of the first determination unit 13A, the second determination unit 13B, and the third determination unit 13C and at least one of the fourth determination unit 13D, the fifth determination unit 13E, and the sixth determination unit 13F. In this case, if all of the combined determination units 13 determine that acquisition of a correct model is necessary, the determination unit 13 may determine that acquisition of a correct model is necessary.

[0050] Although an example has been given in which the user's psychological state is classified into two categories by the estimation unit or the like, the user's psychological state may be classified into more categories.

[0051] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of hardware and / or software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are connected directly or indirectly (e.g., via wire, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining the single device or multiple devices with software.

[0052] Functions include, but are not limited to, judgment, determination, assessment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0053] For example, the data acquisition device 10 according to an embodiment of the present disclosure may function as a computer that performs the psychological state estimation process and data acquisition process of the present disclosure. Fig. 10 is a diagram illustrating an example of the hardware configuration of a data acquisition device according to an embodiment of the present disclosure. The above-described data acquisition device 10 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

[0054] In the following description, the term "apparatus" can be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the information acquisition unit 11, the estimation unit 12, the determination unit 13, the label creation unit 15, and the data storage unit 16 may be configured to include one or more of the devices shown in the figure, or may be configured to exclude some of the devices.

[0055] Each function in the data acquisition device 10 is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.

[0056] The processor 1001 controls the entire computer by running, for example, an operating system. The processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, the above-mentioned estimation unit 12, determination unit 13, label creation unit 15, etc. may be realized by the processor 1001.

[0057] The processor 1001 also reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these programs. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, the estimation unit 12, the determination unit 13, the label creation unit 15, etc. may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and similar implementations may be used for other functional blocks. While the above-described various processes have been described as being executed by a single processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0058] The memory 1002 is a computer-readable recording medium and may be configured by, for example, at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for implementing a data acquisition method according to an embodiment of the present disclosure.

[0059] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0060] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as, for example, a network device, a network controller, a network card, or a communication module. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD). For example, the above-mentioned information acquisition unit 11 and the like may be realized by the communication device 1004.

[0061] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. Note that the input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0062] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0063] The data acquisition device may also be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0064] The notification of information is not limited to the aspects / embodiments described in the present disclosure and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI) and Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB) and System Information Block (SIB))), other signals, or a combination thereof. Furthermore, the RRC signaling may be referred to as an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.

[0065] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0066] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0067] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0068] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0069] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0070] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0071] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.

[0072] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0073] In addition, terms explained in this disclosure and terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings.

[0074] Furthermore, the information, parameters, etc. described in the present disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information. For example, a radio resource may be indicated by an index.

[0075] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0076] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0077] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0078] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0079] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0080] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0081] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0082] A data acquisition device disclosed herein has the following configuration: [1] A data acquisition device comprising: a psychological state estimation model trained using psychological reflection information and a correct label indicating a psychological state corresponding to the psychological reflection information as training data; an information acquisition unit that acquires the psychological reflection information of a user; an estimation unit that estimates the psychological state of the user using the psychological reflection information of the user acquired by the information acquisition unit as input information and the psychological state estimation model; a determination unit that determines whether or not it is necessary to acquire the correct label corresponding to the psychological reflection information of the user acquired by the information acquisition unit; a label creation unit that creates the correct label corresponding to the psychological reflection information of the user based on the user's answer to a test presented to the user when it is determined by the determination unit that it is necessary to acquire the correct label; and a data storage unit that stores the psychological reflection information of the user and the correct label of the user as a set of data. [2] The data acquisition device described in [1], wherein the psychological state estimation model is updated with the data stored in the data storage unit as the training data. [3] The data acquisition device according to [1] or [2], wherein the determination unit determines whether or not it is necessary to acquire the correct label based on a distribution of the data already stored in the data storage unit. [4] The data acquisition device according to [3], wherein the correct labels constituting the training data can be classified into a plurality of categories, and the determination unit determines that it is necessary to acquire the correct label corresponding to the psychological reflection information of the user when a proportion of a specific category constituting the plurality of categories in the data already stored in the data storage unit is equal to or less than a predetermined value and the psychological state of the user estimated using the psychological reflection information of the user and the psychological state estimation model corresponds to the specific category.[5] The data acquisition device according to [3], wherein the information acquisition unit acquires attributes of the user, the data storage unit stores the attributes together with the data, and the determination unit calculates a ratio for each attribute of the data stored in the data storage unit, and determines that it is necessary to acquire the correct label corresponding to the psychology-reflecting information of the user when the ratio of the attribute of the user acquired by the information acquisition unit is equal to or less than a predetermined ratio. [6] The data acquisition device according to any of [1] to [5], wherein the estimation unit classifies the psychological state of the user into a first category and a second category, and the determination unit classifies the user's state into a third category that is broader than the first category and includes the first category, and a fourth category obtained by subtracting the third category from the second category, and determines that it is necessary to acquire the correct label corresponding to the psychology-reflecting information of the user when the user's state is classified into the third category. [7] The data acquisition device according to any one of [1] to [6], wherein the estimation unit stores the previously estimated psychological states for each user, and the determination unit determines whether or not it is necessary to acquire the correct label corresponding to the acquired psychological reflecting information of the user based on a history of the user's psychological states stored by the estimation unit. [8] The data acquisition device according to any one of [1] to [7], wherein the information acquisition unit acquires biometric information of the user as the psychological reflecting information, and the determination unit evaluates the quality of the biometric information of the user and determines that it is unnecessary to acquire the correct label corresponding to the psychological reflecting information of the user if the quality of the biometric information of the user is below a predetermined standard. [9] The data acquisition device according to any one of [1] to [8], wherein the determination unit determines that it is necessary to acquire the correct label corresponding to the psychological reflecting information of the user when a predetermined time has elapsed since the user's previous answer to the test was acquired.

[0083] 10...data acquisition device, 11...information acquisition unit, 12...estimation unit, 13...determination unit, 15...label creation unit, 16...data storage unit, 1001...processor, 1002...memory, 1003...storage, 1004...communication device, 1005...input device, 1006...output device, 1007...bus.

Claims

1. A data acquisition device comprising: a psychological state estimation model trained using psychological reflecting information and a correct label indicating a psychological state corresponding to the psychological reflecting information as learning data; an information acquisition unit that acquires the psychological reflecting information of a user; an estimation unit that estimates the psychological state of the user using the psychological reflecting information of the user acquired by the information acquisition unit as input information and utilizes the psychological state estimation model; a judgment unit that judges whether or not it is necessary to acquire the correct label corresponding to the psychological reflecting information of the user acquired by the information acquisition unit; a label creation unit that creates the correct label corresponding to the psychological reflecting information of the user based on the user's answer to a test presented to the user when the judgment unit judges that it is necessary to acquire the correct label; and a data storage unit that stores the psychological reflecting information of the user and the correct label of the user as a set of data.

2. The data acquisition device according to claim 1, wherein the psychological state estimation model is updated using the data stored in the data storage unit as the learning data.

3. The data acquisition device according to claim 1, wherein the determination unit determines whether or not it is necessary to acquire the correct label based on the distribution of the data already stored in the data storage unit.

4. The data acquisition device of claim 3, wherein the correct labels constituting the learning data can be classified into a plurality of categories, and the determination unit determines that it is necessary to acquire the correct label corresponding to the psychological reflection information of the user when the psychological state of the user estimated using the psychological state estimation model with the psychological reflection information of the user as input information corresponds to the particular category when a proportion of a specific category constituting the plurality of categories in the data stored in the data storage unit is equal to or less than a predetermined value.

5. The data acquisition device of claim 3, wherein the information acquisition unit acquires attributes of the user, the data storage unit stores the attributes together with the data, and the determination unit calculates a proportion of each attribute of the data stored in the data storage unit, and when the proportion of the attributes of the user acquired by the information acquisition unit is below a predetermined proportion, determines that it is necessary to acquire the correct label corresponding to the psychological reflection information of the user.

6. The data acquisition device of claim 1, wherein the estimation unit classifies the psychological state of the user into a first category and a second category, and the determination unit classifies the user's state into a third category that is broader than the first category and includes the first category, and a fourth category that subtracts the third category from the second category, and determines that it is necessary to obtain the correct label corresponding to the psychological reflection information of the user when the user's state is classified into the third category.

7. A data acquisition device as described in claim 1, wherein the estimation unit stores the psychological states estimated in the past for each user, and the determination unit determines whether or not it is necessary to acquire the correct label corresponding to the acquired psychological reflection information of the user based on the history of the psychological states of the user stored by the estimation unit.

8. The data acquisition device of claim 1, wherein the information acquisition unit acquires biometric information of the user as the psychological reflection information, and the determination unit evaluates the quality of the biometric information of the user and, if the quality of the biometric information of the user falls below a predetermined standard, determines that it is unnecessary to acquire the correct label corresponding to the psychological reflection information of the user.

9. The data acquisition device of claim 1, wherein the determination unit determines that it is necessary to acquire the correct answer label corresponding to the user's psychological reflection information when the user's psychological reflection information is acquired and a predetermined time has elapsed since the user's previous answer to the test was acquired.

Citation Information

Patent Citations

  • Questionnaire question presentation system, questionnaire question presentation method, and questionnaire question presentation program

    JP2011039883A

  • Information analysis device, information analysis method, information analysis program, and learning model generation method

    JP2020181577A

  • Stress sensitivity evaluation form and program

    WO2013042418A1