Estimation device and computer program

Abstract

The present invention provides an estimation device comprising: a personal data acquisition unit that acquires information about the location and activity situation of an individual person and biological information about the biology of said person; an environmental information acquisition unit that acquires environmental information including video and / or audio of the person's surroundings; a correction parameter estimation unit that uses each piece of information acquired by the personal data acquisition unit and the environmental information acquired by the environmental information acquisition unit as a basis for estimating, as a correction parameter, an estimated result of the effect on the biological information after the start or after the end of a state element indicating an activity or situation that affects the biological information; and a psychological state estimation unit that uses the correction parameter estimated by the correction parameter estimation unit to correct the biological information obtained from the person or an estimation parameter to be used for estimation of psychological state, and estimates a psychological state of the person on the basis of the corrected biological information or the corrected estimation parameter.　

Description

Estimation device and computer program

[0001] The present invention relates to an estimation device and a computer program.

[0002] Conventionally, a technology has been proposed for estimating a user's emotions based on the user's biometric information (see, for example, Patent Document 1). For example, the technology described in Patent Document 1 classifies the user U as being in a specific state or a specific situation based on the user U's physical state and location information, and performs a psychological state estimation using an estimator corresponding to the classified specific state or specific situation, as shown in Fig. 17, thereby improving the accuracy of the estimation. For example, in Fig. 17, examples of the specific state or specific situation include desk work, being in a meeting, exercising, and taking a break.

[0003] FIG. 17 shows an example of application of the technology of Patent Document 1, in which a psychological state is estimated according to a classified specific state or specific situation using estimation parameters in which the upper and lower limits of the normal range for each specific state or specific situation are defined and which are calculated based on the user's past data.

[0004] JP 2016-106689 A

[0005] As described above, the technology described in Patent Document 1 classifies a user into a single specific state or situation based on the user's biometric information. However, a specific state or situation is not actually a single state or situation, but rather a combination of multiple locations, postures / movements, actions, health conditions, etc., as shown in FIG. 18 . For example, even if the state is "in a meeting," the user's actions will generally be a combination of states such as "speaking," "seated (sitting position)," "standing," and "moving."

[0006] If all combinations of "situations" such as "speaking during a meeting" and "sitting and listening during a meeting" were considered, the number of possible situations would be extremely large, and calculating and maintaining estimation parameters for all of these situations would be costly. Therefore, estimation is performed by dividing the state into a certain number of variations, but this poses the problem of reduced accuracy in estimating psychological states.

[0007] In view of the above circumstances, an object of the present invention is to provide a technique that can improve the accuracy of estimating a user's psychological state.

[0008] One aspect of the present invention is an estimation device comprising: a personal data acquisition unit that acquires information regarding the location and activity status of each person and biometric information regarding the living body; an environmental information acquisition unit that acquires environmental information including at least one of video and audio around the person; a correction parameter estimation unit that estimates, as a correction parameter, an estimated result of an influence on the biometric information after the start of a state element indicating an action or situation that affects the biometric information or after the end of a state element, based on the information acquired by the personal data acquisition unit and the environmental information acquired by the environmental information acquisition unit; and a psychological state estimation unit that uses the correction parameters estimated by the correction parameter estimation unit to correct the biometric information obtained from the person or estimation parameters used to estimate the psychological state, and estimates the psychological state of the person based on the corrected biometric information or the corrected estimation parameters.

[0009] One aspect of the present invention is a computer program for causing a computer to execute the following steps: a personal data acquisition step for acquiring information regarding the location and activity status of each person and biometric information regarding the living body; an environmental information acquisition step for acquiring environmental information including at least one of video and audio around the person; a correction parameter estimation step for estimating, as a correction parameter, an estimated result of an influence on the biometric information after the start of a state element indicating an action or situation that affects the biometric information or after the end of a state element, based on each piece of information acquired in the personal data acquisition step and the environmental information acquired in the environmental information acquisition step; and a psychological state estimation step for correcting the biometric information obtained from the person or an estimation parameter used to estimate the psychological state, using the correction parameter estimated in the correction parameter estimation step, and estimating the psychological state of the person based on the corrected biometric information or the corrected estimation parameter.

[0010] According to the present invention, it is possible to improve the accuracy of estimating the psychological state of a user.

[0011] FIG. 1 is a diagram (part 1) for explaining the basic concept of the present invention. FIG. 2 is a diagram (part 2) for explaining the basic concept of the present invention. FIG. 1 is a diagram for explaining a method for estimating the impact after the start of a certain state element in the present invention. FIG. 2 is a diagram for explaining a method for estimating the impact after the end of a certain state element in the present invention. FIG. 3 is a diagram for explaining processing when multiple state elements newly start or multiple state elements end in the present invention. FIG. 4 is a diagram for explaining a method for correcting biometric data or correcting estimated parameters in the present invention. FIG. 5 is a diagram showing an example of the configuration of an estimation system in an embodiment. FIG. 6 is a diagram showing an example of the configuration of an estimation device in an embodiment. FIG. 7 is a diagram for explaining a method for accumulating user states in the present invention. FIG. 8 is a diagram for explaining a method for accumulating correction parameters in the present invention. FIG. 9 is a flowchart showing the flow of data acquisition processing performed by an estimation device in an embodiment. FIG. 10 is a flowchart showing the flow of processing for estimating the impact of a state element on biometric data performed by an estimation device in an embodiment. FIG. 11 is a diagram (part 1) for explaining a problem arising from the configuration in an embodiment and a method for solving the problem. FIG. 12 is a diagram (part 2) for explaining a problem arising from the configuration in an embodiment and a method for solving the problem. 1A and 1B are diagrams for explaining a conventional method for estimating a psychological state and a problem of the conventional method.

[0012] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0013] (Overview) Before describing the specific device configuration of the present invention, an overview of the processing of the present invention will be described in detail. First, as a premise, in the present invention, actions or situations that affect biometric data obtained from a user are defined as "status elements." The estimation device in the present invention calculates the impact of each status element on the biometric data for both "after start" and "after end" based on past data.

[0014] "After start" refers to a certain period of time starting immediately after a certain state element starts. "After end" refers to a certain period of time starting immediately after a certain state element ends. The estimation device corrects the estimation parameters or biological data used when estimating the psychological state using the influence (estimated value) of the calculated state element. This makes it possible to estimate the psychological state in a way that more accurately reflects the influence of various states, even when they overlap.

[0015] A specific example will be described below with reference to Figures 1 and 2. Figure 1 is a diagram (part 1) for explaining the basic concept of the present invention. In Figure 1, the basic concept will be explained taking into account the effect on biometric data after the start of status element x.

[0016] As shown in (A) of Figure 1, it is assumed that several transitions of biometric data (three types in Figure 1) when various states of user U overlap are acquired. The upper diagram of (A) of Figure 1 shows a case where a state element x is newly added when states a, b, and c overlap, the middle diagram of (A) of Figure 1 shows a case where a state element x is newly added when states d and e overlap, and the bottom diagram of (A) of Figure 1 shows a case where a state element x is newly added when states a, g, and h overlap. Note that states a, b, c, d, e, g, h, and x shown here are state elements that indicate specific states or specific situations (e.g., at one's desk, in a conference room, walking, on the phone, speaking, etc.) described in Figure 18.

[0017] The state transitions from each state shown in FIG. 1A when a state element x is newly added are state {a, b, c} → {a, b, c, x}, state {d, e} → {d, e, x}, and state {a, g, h} → {a, g, h, x}, respectively. The estimation device estimates (calculates) the impact of the transition to state element x when state element x is added based on past data. That is, when the state before the transition is an arbitrary state in which state element x is not performed, the estimation device estimates the impact on the biometric data after the transition to state element x after the addition of state element x based on past data. This point will be explained using FIG. 1B.

[0018] Figure 1B shows the estimated results of the impact of status element x on biometric data. As shown in Figure 1B, when status element x has not started, there is no impact on the biometric data, so the impact on the biometric data is 0, and when status element x has started, there is some impact on the biometric data, so the impact on the biometric data has a value other than 0. For simplicity of explanation, Figure 1B shows a case where the impact on the biometric data when status element x has started increases to a certain value and then continues to maintain that value, but this varies depending on the status element. For example, depending on the status element, the impact on the biometric data may continue to increase, or the impact on the biometric data may continue to decrease, or it may be possible for the impact to increase and decrease repeatedly.

[0019] The estimation device reflects the estimation result of the influence of the status element x shown in FIG. 1B on the biometric data shown in FIG. 1A. Specifically, the estimation device corrects the biometric data using the estimation result of the influence of the status element x shown in FIG. 1B on the biometric data. In this manner, the estimation result of the influence of a certain status element on the biometric data is a correction parameter used to correct the biometric data or an estimation parameter. The estimation device may also correct the estimation parameter using the estimation result of the influence of the status element x shown in FIG. 1B on the biometric data. In the example shown in FIG. 1C, since the numerical value increases after the start of the status element x due to the influence of the status element x as shown in FIG. 1B, the estimation device corrects the biometric data after the start of the status element x in the negative direction based on the estimation result. The estimation device then estimates the psychological state of the user U using the corrected biometric data. The method of estimating the psychological state in this case is the same as conventional methods.

[0020] Furthermore, when the estimation device corrects the estimated parameters, the estimation device corrects the estimated parameters after the start of state element x. For example, as shown in FIG. 1B , because the numerical values ​​increase after the start of state element x due to the influence of state element x, the estimation device corrects the upper and lower limit values ​​of the estimated parameters in the positive direction based on the estimation results for the portion corresponding to after the start of state element x. As a result, the upper and lower limit values ​​become higher than those of the estimated parameters when state element x has not started. The estimation device estimates the psychological state of user U based on the acquired biometric data and the corrected estimated parameters.

[0021] Next, a basic concept based on the impact on biometric data after the end of status element x will be described with reference to FIG. 2. FIG. 2 is a diagram (part 2) for explaining the basic concept of the present invention. As shown in FIG. 2A, assume that several transitions of biometric data (three types in FIG. 1) when various states of user U overlap are acquired. The upper diagram in FIG. 2A shows a case where status element x ends when states a, b, c, and x overlap, the middle diagram in FIG. 1A shows a case where status element x ends when states d, e, and x overlap, and the lower diagram in FIG. 1A shows a case where status element x ends when states a, g, h, and x overlap.

[0022] The state transitions from each state shown in Figure 2A when state element x ends are state {a, b, c, x} → {a, b, c}, state {d, e, x} → {d, e}, and state {a, g, h, x} → {a, g, h}, respectively. The estimation device estimates (calculates) the impact of the transition when state element x ends based on past data. That is, when the state before the transition was an arbitrary state in which state element x was performed, the estimation device estimates the impact on the biometric data after the transition after state element x ends based on past data. This point will be explained using Figure 2B.

[0023] Figure 2(B) shows the estimated results of the impact on biometric data due to the end of status element x. As shown in Figure 2(B), when status element x has started, that state is maintained and the impact on the biometric data is 0, and when status element x has ended, some impact on the biometric data occurs and the impact on the biometric data has a value other than 0. For the sake of simplicity, Figure 2(B) shows a case where the impact on the biometric data when status element x has ended drops to a certain value and then continues to maintain that value, but this differs depending on the status element.

[0024] The estimation device reflects the estimated result of the impact of the end of the status element x shown in FIG. 2B on the biometric data shown in FIG. 2A. Specifically, the estimation device corrects the biometric data using the estimated result of the impact of the end of the status element x shown in FIG. 2B on the biometric data. The estimation device may also correct estimated parameters using the estimated result of the impact of the end of the status element x shown in FIG. 2B on the biometric data. As shown in FIG. 2B, the numerical value decreases after the end of the status element x due to the impact of the end of the status element x. Therefore, in the example shown in FIG. 2C, the estimation device corrects the biometric data shown in FIG. 2A (e.g., the top biometric data in FIG. 2A) after the end of the status element x in the positive direction based on the estimated result of the impact of the end of the status element x shown in FIG. 2B on the biometric data. The estimation device then estimates the psychological state of the user U using the corrected biometric data. The psychological state estimation method in this case is the same as conventional methods.

[0025] Furthermore, when the estimation device corrects the estimated parameters, the estimation device corrects the estimated parameters after the end of state element x. For example, since the numerical value decreases after the end of state element x due to the influence of the end of state element x as shown in FIG. 2B, the estimation device corrects the upper and lower limit values ​​of the part of the estimated parameters corresponding to the end of state element x in the negative direction based on the estimated result of the influence on the biometric data of the end of state element x as shown in FIG. 2B. As a result, the upper and lower limit values ​​become lower than those of the estimated parameters when state element x has started. The estimation device estimates the psychological state of user U based on the acquired biometric data and the corrected estimated parameters.

[0026] The above is the basic concept of the present invention. Next, a method for estimating the effect of the above-mentioned certain condition element on biometric data will be described.

[0027] (Method of estimating impact on biometric data) The estimation device estimates the impact after the start or end of a status element based on the accumulated biometric data of each status element for each user U. A method of estimating the impact after the start or end of a certain status element will be described with reference to Figures 3 to 5. Figures 3 and 4 are diagrams for explaining the method of estimating the impact after the start of a certain status element in the present invention.

[0028] For example, the estimation device may use a first method that uses a simple average and a second method that uses a combination of prediction and weighting as methods for estimating the impact after the onset of a certain status element. The first method that uses a simple average is described with reference to FIG. 3 . The estimation device calculates an average using multiple pieces of time-series biometric data of user U before and after the onset of status element x (see the upper diagram in FIG. 3 ). It is assumed here that the time-series biometric data of user U before and after the onset of status element x has been stored in advance in a storage unit.

[0029] As shown in the lower diagram of Figure 3, the estimation device calculates the difference between the average L1 for a certain period before the start of condition element x and the average L2 after the start of condition element x (an average obtained by simple averaging) as the impact on the biological data after the start of condition element x.

[0030] Next, a second method using a combination of prediction and weighting will be described with reference to Figure 4. Consider a case where there is a lack of data from a sufficient time after the start of state element x. In this case, the estimation device estimates a curve for when state element x continues by using curve approximation or the like on the data from the start to the end of state element x (for example, the upper left diagram in Figure 4). However, the reliability of the data in the estimated portion decreases over time.

[0031] The estimation device weights multiple approximation results based on reliability when combining them. Data when condition element x continues for a sufficient period of time is considered to have a reliability of 100%. Each approximation curve is weighted by reliability to estimate the impact on the biometric data after the start of condition element x. As in the case of simple averaging, the estimation device calculates the difference between the average for a certain period before the start of condition element x and the approximation curve (+ reliability weighting) after the start of condition element x as the impact on the biometric data after the start of condition element x.

[0032] Next, a method for estimating the impact after the end of a certain state element will be described. FIG. 5 is a diagram for explaining a method for estimating the impact after the end of a certain state element in the present invention. For example, the estimation device can estimate the impact after the end of a certain state element using a first method that uses a simple average, a second method that uses a combination of prediction and weighting, and a third method that is based on a correction width after the start. Note that the first and second methods are similar to the method for estimating the impact after the start of a certain state element.

[0033] First, the first method using a simple average will be described. The estimation device calculates the average using multiple pieces of time-series biometric data of user U while he is performing state element x (in FIG. 5, "during state x") and after the end of state element x (see the upper left diagram in FIG. 5). Here, it is assumed that the time-series biometric data of user U while he is performing state element x and after the end of state element x has been stored in a storage unit in advance.

[0034] The estimation device calculates the difference between the average L3 for a certain period during the execution of the state element x and the average L4 after the end of the state element x (an average obtained by simple averaging) as the impact on the biological data after the end of the state element x.

[0035] Next, we will explain the second method, which uses a combination of prediction and weighting. Consider a case where there is a lack of data from a time period after the end of state element x. In this case, the estimation device estimates fluctuations that would occur if a state continued without the occurrence or end of a new state element after the end of state element x. However, the reliability of the data in the estimation portion decreases over time.

[0036] The estimation device weights multiple approximation results based on reliability when combining them. Data when the state in which status element x has ended continues for a sufficient period of time is considered to have a reliability of 100%. Each approximation curve is weighted by reliability to estimate the impact on the biometric data after the end of status element x. As in the case of simple averaging, the estimation device calculates the difference between the average over a certain period during the execution of status element x and the approximation curve (+ reliability weighting) after the end of status element x as the impact on the biometric data after the end of status element x.

[0037] Next, a third method for performing post-termination correction according to the correction width after start, such as when status element x ends in a short time, will be described using the right diagram of Figure 5. If status element x ends in a short time, the state may end before the influence of status element x is fully apparent (before it reaches its peak in the right diagram of Figure 5). If the influence after the end of status element x is applied as is, the correction may be too large, resulting in an overcorrection of the biometric data value or estimated parameter. In such a case, the estimation device may terminate the post-termination correction within the range of influence of status element x (up to the point where the correction after the start of status element x is canceled out), as shown in Method 1 in the right diagram of Figure 5, or may use a method such as multiplying the post-termination correction after status element x by a coefficient according to the influence of status element x, as shown in Method 2 in the right diagram of Figure 5.

[0038] The above description has been given of a case where a single state element (e.g., state element x) is newly started in any state element being performed by the user U, or a case where a single state element (e.g., state element x) ends from any state element being performed by the user U. However, when actually correcting biometric data or estimated parameters, there are also cases where multiple state elements are newly started or multiple state elements end.

[0039] Therefore, the processing when multiple state elements start anew or when multiple state elements end will be described with reference to Fig. 6. Fig. 6 is a diagram for explaining the processing when multiple state elements start anew or when multiple state elements end in the present invention. Fig. 6 shows a case where two new state elements, state element a and state element b, start in addition to an arbitrary state element being performed by user U.

[0040] In the situation shown in Figure 6, the estimation device estimates the impact on the biometric data when status element a is started and the impact on the biometric data when status element b is started, as shown in the left diagram of Figure 6. Then, the estimation device superimposes the impact on the biometric data when the estimated status element a is started and the impact on the biometric data when status element b is started, as a combination of the estimated multiple status elements. The estimation result obtained is shown in the right diagram of Figure 6. The estimation device corrects the biometric data or estimation parameters using the superimposed estimation result. Note that Figure 6 has been described using an example where multiple status elements are newly started, but even when multiple status elements end, it is sufficient to estimate and superimpose the impact on the biometric data after the end of each status element.

[0041] (Correction Method) Next, correction of biometric data or correction of estimated parameters by the estimation device will be described with reference to FIG. 7. FIG. 7 is a diagram for explaining a method for correcting biometric data or correcting estimated parameters in the present invention. As shown in FIG. 7A, it is assumed that an estimation result indicating the influence on biometric data after the end of state element x is obtained. The estimation device may use the obtained estimation result to correct the biometric data as shown in FIG. 7B, or may correct the estimated parameters as shown in FIG. 7C.

[0042] When correcting biometric data as shown in Figure 7 (B), all individual biometric data and analytical data are corrected according to the situation, which has the disadvantage of a large processing load (amount of calculation). However, it has the advantage of being easily applicable even in cases where the relationship between the values ​​of biometric data or analytical data and the parameters used for judgment is complex, such as in machine learning.

[0043] When correcting estimated parameters as shown in Figure 7 (C), since the judgment parameters, which are usually far fewer than the number of biological data, are corrected, there is an advantage in that the processing load (calculation amount) is small, but there is a disadvantage in that it cannot be easily applied to machine learning, etc., in which it is difficult to adjust the judgment by correcting parameters.

[0044] The above is an overview of the processing of the present invention. Next, a specific configuration for realizing the above processing will be described.

[0045] 8 is a diagram illustrating an example of the configuration of an estimation system 100 according to an embodiment. The estimation system 100 is a system that estimates a user's psychological state. The estimation system 100 includes a sensor terminal group 10, an estimation device 20, and a notification device 30.

[0046] The sensor terminal group 10 is a sensor that has the function of acquiring biometric data and status information of users U1 and U2. The sensor terminal group 10 includes a personal terminal 11 and an environmental information terminal 12. Users U1 and U2 shown in FIG. 8 are people whose psychological states are to be estimated, and user U3 is a person (e.g., a boss) who is to be notified of the estimation results of the psychological states of users U1 and U2. Note that FIG. 8 shows a case where there are two people, users U1 and U2, whose psychological states are to be estimated, but the number of people whose psychological states are to be estimated is not particularly limited.

[0047] The personal terminal 11 is a terminal device held by each user whose psychological state is to be estimated. For example, the personal terminal 11 may be a terminal device worn on the wrist (e.g., a smart watch) or a terminal device worn on a finger (e.g., a smart ring). The personal terminal 11 is equipped with sensors capable of acquiring biometric data of users U1 and U2, such as a heart rate sensor, a pulse wave sensor, and an electrodermal activity sensor. The personal terminal 11 is also equipped with sensors capable of acquiring position information and state elements of users U1 and U2, such as a positioning sensor and an acceleration / angular velocity sensor. As described above, the state elements indicate specific states or specific situations of users U1 and U2. In other words, the state elements are information indicating the activity status of users U1 and U2.

[0048] The biological data is a value measured using the above-mentioned various sensors. For example, when a reflective optical pulse wave sensor is used, the biological data is a value indicating the level of reflected light from an LED at a specific time point. When an electrodermal activity sensor for detecting sweating is used, the biological data is a voltage value.

[0049] The environmental information terminal 12 is a camera that captures the facial expressions, full-body appearance, movements, etc. of the users U1 and U2. The environmental information terminal 12 may also include a microphone (office microphone) that records the speech of the users U1 and U2 and other sounds that are generated. In addition to the camera and microphone, the environmental information terminal 12 may also be equipped with an environmental information terminal (not shown) that can measure temperature, humidity, illuminance, etc.

[0050] The estimation device 20 estimates the psychological states of users U1 and U2 by combining the biometric data, location information, and state elements acquired by the personal terminal 11 with the video data, audio data, and the like acquired by the environmental information terminal 12. The estimation device 20 is, for example, an information processing device such as a general-purpose computer.

[0051] The notification device 30 executes a notification process for notifying the estimation result obtained by the estimation device 20. The notification device 30 is, for example, an information processing device such as a general-purpose computer.

[0052] The personal terminal 11 transmits biometric data, location information, and the like to the estimation device 20 via a communication network (not shown). For example, the personal terminal 11 communicates with a wireless LAN router connected to the communication network via a wireless connection using Wi-Fi (registered trademark). This allows the personal terminal 11 to communicate with the estimation device 20 connected to the connected communication network.

[0053] The communication connection between the personal terminal 11 and the estimating device 20 is not limited to the above-described configuration in which they are connected using Wi-Fi (registered trademark). For example, the personal terminal 11 and the estimating device 20 may be connected by communication based on a short-range wireless communication standard such as Bluetooth (registered trademark). Alternatively, the personal terminal 11 and the estimating device 20 may be connected by communication based on a mobile phone communication standard such as LTE (Long Term Evolution) (registered trademark). Alternatively, the personal terminal 11 and the estimating device 20 may be connected by a combination of communication using multiple communication methods such as those described above.

[0054] The personal terminal 11 and the estimation device 20 may be configured to be communicatively connected via a mobile information terminal (not shown), such as a smartphone, carried by the users U1 and U2. In this case, for example, the personal terminal 11 and the smartphone may be connected via Bluetooth (registered trademark), and the smartphone and the estimation device 20 may be connected via Wi-Fi (registered trademark) or LTE (registered trademark).

[0055] [Configuration of Estimation Device] The configuration of the estimation device 20 will be described in more detail below.

[0056] 9 is a diagram showing an example of the configuration of the estimation device 20 according to an embodiment. The estimation device 20 includes a personal data acquisition unit 201, a biometric information analysis unit 202, a biometric information storage unit 203, an environmental information acquisition unit 204, a user situation analysis unit 205, a user situation storage unit 206, a correction parameter estimation unit 207, a correction parameter storage unit 208, a psychological state estimation unit 209, and a psychological state storage unit 210.

[0057] The personal data acquisition unit 201 acquires various data from the personal terminal 11 and outputs it to the biometric information analysis unit 202 and the user status analysis unit 205. The personal data acquisition unit 201 is a communication interface for connecting to and communicating with the personal terminal 11 via a communication network. For example, the personal data acquisition unit 201 acquires, from the personal terminal 11, information on the position and activity status of each user U (for example, position information and status elements of user U from a positioning sensor, acceleration / angular velocity sensor, etc.), and biometric data.

[0058] The biometric information analysis unit 202 acquires biometric values ​​that can be used to estimate psychological states, along with time and a user identifier, from the personal data acquisition unit 201. For example, the biometric values ​​include heart rate sensor data, pulse wave sensor data, and electrodermal activity sensor data. For example, the heart rate sensor data is data such as an instantaneous current value or voltage value. For example, the pulse wave sensor data is data such as an instantaneous measurement value of reflected light intensity. For example, the electrodermal activity sensor data is data such as an instantaneous current value or voltage value.

[0059] Note that if sensor data usable for estimating the psychological state can be obtained from a device other than the personal terminal 11 (for example, a device that is not worn by the user, such as a non-contact pulse wave sensor), the biological information analysis unit 202 may acquire the sensor data from such a device. For example, if data obtained by the environmental information terminal 12 can be used for estimating the psychological state, the biological information analysis unit 202 may also acquire sensor data from the environmental information acquisition unit 204, which will be described later.

[0060] The biological information analysis unit 202 analyzes the acquired biological measurements and calculates analysis data that can be used to analyze the psychological state. For example, the analysis data is data related to heart rate, pulse, or electrodermal activity. For example, the data related to heart rate is the heart rate, heartbeat interval, and fluctuations in the heartbeat interval obtained as a result of analyzing the electrocardiogram data. For example, the data related to pulse is the pulse rate, pulse wave interval, and fluctuations in the pulse wave interval. For example, the data related to electrodermal activity is information indicating the degree of increase or decrease in sweat level, such as the difference between the measured value and the average for a given period of time in the past (n hours), the amount of fluctuation in the measured value for a given period of time most recently (n seconds), or the rate of fluctuation in the measured value for a given period of time most recently (n seconds).

[0061] The biometric information analysis unit 202 stores the acquired biometric measurement values ​​and the calculated analysis data in the biometric information storage unit 203. Hereinafter, data including both the biometric measurement values ​​and the analysis data will be collectively referred to as "biometric data." Note that the biometric information analysis unit 202 may calculate new analysis data using past biometric measurement values ​​and analysis data stored in the biometric information storage unit 203.

[0062] The biometric information storage unit 203 stores the biometric data acquired by the biometric information analysis unit 202. The biometric information storage unit 203 provides the biometric data in response to a request from a correction parameter estimation unit 207 (described later) for calculating correction parameters for estimating a psychological state. The biometric information storage unit 203 also provides biometric data at a specified time in response to a request from a psychological state estimation unit 209 (described later) for estimating the psychological state at each time. The biometric information storage unit 203 also provides biometric measurement values ​​and analysis data in response to a request from the biometric information analysis unit 202.

[0063] The environmental information acquisition unit 204 acquires video data and audio data from the environmental information terminal 12 and outputs the data to the user situation analysis unit 205. The environmental information acquisition unit 204 may also acquire various sensor data, such as a temperature sensor or a humidity sensor, from other environmental information terminals (not shown) and output the data to the user situation analysis unit 205. The environmental information acquisition unit 204 is a communication interface for establishing a communication connection with the environmental information terminal 12 via a communication network.

[0064] The user status analysis unit 205 analyzes various data obtained from the personal terminal 11, the environmental information terminal 12, etc., and generates state element history data showing the history of state elements of users U1 and U2, and environmental information history data showing the history of environmental information.

[0065] The user situation analysis unit 205 acquires data relating to the user environment from the personal data acquisition unit 201 and the environment information acquisition unit 204, together with the time and the user identifier.

[0066] For example, the data that the user situation analysis unit 205 acquires from the personal data acquisition unit 201 is data obtained by the personal terminal 11, such as location data and activity information.

[0067] For example, the location data is location information obtained by a satellite positioning system such as a GPS (Global Positioning System). Furthermore, for example, the activity information is data indicating the user's movement speed, the user's exercise type, and the exercise load level, calculated from the sensor values ​​of a positioning sensor and an acceleration / angular velocity sensor. The user's exercise type here refers to data indicating the user's exercise state, such as standing, walking, or running, and data indicating the user's posture, such as sitting or standing. Furthermore, for example, the activity information also includes data indicating various state elements (such as desk work, work other than desk work, meetings, and travel) obtained by integrating all of the speech estimated from a microphone and the movement, posture, and speech state estimated from the status of the acceleration / angular velocity sensor.

[0068] For example, the data acquired by the user situation analysis unit 205 from the environmental information acquisition unit 204 is data obtained by the environmental information terminal 12 or the like, such as video data, audio data, and environmental data.

[0069] For example, the video data is data showing video of users U1 and U2 in a specific location (e.g., a location where a camera is installed, such as an office, a break room, or a conference room). Furthermore, for example, the audio data is data showing audio accompanying the video, or data showing audio recorded separately from the video, such as conversations in the office. Furthermore, for example, the environmental data is information showing the environmental conditions of the location where users U1 and U2 are located (e.g., a conference room), such as temperature, humidity, illuminance, and noise level.

[0070] The user status analysis unit 205 analyzes users U1 and U2 contained in the video data and audio data. For example, when a person appears in the video, the user status analysis unit 205 converts the coordinates of the person in the video into relative position information within the conference room (or building) or global position information based on the camera's installation position and direction and the camera's angle of view. Note that information indicating the camera's installation position and direction and the camera's angle of view is pre-stored, for example, in a storage medium (not shown) included in the estimation device 20. The user status analysis unit 205 identifies the person (the user ID) in the video by comparing the converted position information with position information obtained from the personal terminal 11.

[0071] Alternatively, the user status analysis unit 205 identifies (the user ID of) a person in the video by, for example, comparing a facial image of the person in the video with a facial image of a previously prepared employee, etc. In this case, data associating the facial image of the employee, etc. with the user ID is stored in advance in, for example, a storage medium (not shown) included in the estimation device 20.

[0072] Alternatively, the user status analysis unit 205 may identify the position of the speaker based on, for example, audio information contained in multiple microphones. The user status analysis unit 205 then compares the identified position information with position information obtained from the personal terminal 11 to identify the person (the user ID) in the video.

[0073] Alternatively, the user situation analysis unit 205 may identify (the user ID of) a person in the video by, for example, comparing the voiceprint of the person in the video with a previously prepared voiceprint of an employee, etc. In this case, data associating the voiceprint of the employee, etc. with the user ID is stored in advance in, for example, a storage medium (not shown) included in the estimation device 20.

[0074] Alternatively, the user status analysis unit 205 identifies (the user ID of) a person in the video based on data obtained from an entry / exit management system (not shown) that can identify the user ID of a person when the person enters or leaves a conference room. That is, the user status analysis unit 205 estimates that a person identified as being in the conference room is a person who appears in the video or a person whose speech is recorded in the audio.

[0075] Alternatively, the user status analysis unit 205 acquires, for example, location information of an area in a conference room where video images can be captured by a camera and audio can be captured by a microphone. When the user status analysis unit 205 confirms from the video that a person has entered the area, it compares the location information of the area with the location information obtained from the personal terminal 11 to identify the person (the user ID) in the video.

[0076] Alternatively, the user status analysis unit 205 acquires conference information including information on participants in a conference, etc., from another system (not shown), such as an in-house conference system, for example. The user status analysis unit 205 identifies the participants of the meeting included in the conference information, thereby identifying the people (user IDs) in the video.

[0077] The user status analysis unit 205 estimates the user's movements and actions based on various data acquired by the personal terminal 11 and the environmental information terminal 12. The user status analysis unit 205 integrates the estimated user movements and actions and classifies them into defined state elements. The user movements and actions estimated by the user status analysis unit 205 are examples of state elements.

[0078] The user situation analysis unit 205 analyzes data indicating the position, speed, and acceleration acquired by, for example, an acceleration / angular velocity sensor and a positioning sensor provided in the personal terminal 11, and estimates the user's posture and movement state.

[0079] The user status analysis unit 205 analyzes the user's posture and behavior from video data acquired by the environmental information terminal 12, for example, and associates the user being analyzed with a user ID based on location information obtained from the personal terminal 11, etc.

[0080] The user status analysis unit 205 estimates and classifies the user's status elements by integrating, for example, information obtained from the positioning sensor, acceleration / angular velocity sensor, etc. of the personal terminal 11, information obtained from environmental information terminals such as the environmental information terminal 12, and information obtained from the conference system and scheduler. The user status analysis unit 205 then generates status element history data for each user. FIG. 10 shows an example of the status element history data. As shown in FIG. 10, the status element history data is data in which, for example, a user identifier (user ID), a status element, an activity start time, and an activity end time are associated with each other.

[0081] FIG. 10 is a diagram for explaining a method for storing user states (state element history data) in the present invention. Consider the situation shown in FIG. 10 as an example of the occurrence of state elements of user U. As shown in the upper diagram of FIG. 10, user U entered conference room 1 before the conference, and during the conference was in a sitting → standing → sitting state state. He left conference room 1 just under two minutes after the conference ended. The user status analysis unit 205 stores this information as user state in a format in which the start time and end time are indicated by dividing it into state elements as shown in the lower diagram of FIG. 10.

[0082] The user status analysis unit 205 aggregates and analyzes the environmental information output from the environmental information acquisition unit 204. The user status analysis unit 205 aggregates data such as temperature, humidity, and illuminance (in a conference room, etc.) that may affect biometric information, and generates history data by associating the data with time information. Note that the user status analysis unit 205 converts the data as needed to numerical values ​​and data classifications that are easy to use in subsequent analysis processing. For example, the user status analysis unit 205 can convert data indicating temperature and humidity into more easily usable data such as an "uncomfort index."

[0083] The user status storage unit 206 stores the state element history data of users U1 and U2 generated by the user status analysis unit 205. In response to a request from the correction parameter estimation unit 207, the user status storage unit 206 provides information indicating a time period corresponding to a specified state element of a specified user. In addition, in response to a request from the psychological state estimation unit 209, the user status storage unit 206 provides information indicating a state element of a specified user during a specified time period.

[0084] The user situation storage unit 206 may store environmental information aggregated by the user situation analysis unit 205. For example, the user situation storage unit 206 stores data such as temperature, humidity, and illuminance in chronological order for each of the users U1 and U2. This allows the user situation storage unit 206 to provide environmental information for each of the users U1 and U2 at each time.

[0085] The correction parameter estimation unit 207 estimates, as a correction parameter, an estimated result of an influence on the biometric data after the start or end of a status element based on the biometric data stored in the biometric information storage unit 203 and the status element history data stored in the user situation storage unit 206. In this way, the correction parameter estimation unit 207 is a functional unit that estimates an estimated result of an influence on the biometric data by a certain status element described in Figures 1 to 6.

[0086] The correction parameter estimation unit 207 acquires past biometric data and analysis data for each status element. The correction parameter estimation unit 207 acquires information indicating users and time periods in each status element from the user status storage unit 206. The correction parameter estimation unit 207 also acquires biometric data and analysis data for each user and time period from the biometric information storage unit 203, and compiles the acquired biometric data and analysis data for each status element to generate a history data group.

[0087] The correction parameter estimation unit 207 acquires past biometric data for each status element from the biometric information storage unit 203. The correction parameter estimation unit 207 acquires information indicating the user and time period in each status element from the user status storage unit 206. The correction parameter estimation unit 207 also acquires biometric data for each user and time period from the biometric information storage unit 203, and compiles the acquired biometric data for each status element to generate a history data group.

[0088] The correction parameter estimation unit 207 first acquires a list of status elements. The list of status elements is a list of information indicating status elements such as "desk work" and "movement (walking)." Next, the correction parameter estimation unit 207 acquires information in which the status elements in the list of status elements are associated with the user ID and time period from the user status storage unit 206. Next, the correction parameter estimation unit 207 acquires information in which the status elements in the list of status elements are associated with various types of biological information (values ​​such as pulse rate, pulse wave peak interval, and skin electrical resistance) from the biological information storage unit 203.

[0089] The history data group may be generated by acquiring all relevant data, but in practice, the amount of data would be enormous. Therefore, the history data group may be generated by acquiring only a certain number of recent data or only data from a certain period of time. The biometric values ​​acquired here are not data for each user, but data for each condition element for all users.

[0090] The correction parameter estimation unit 207 estimates a correction parameter for each state element based on past biological information.

[0091] The correction parameter storage unit 208 stores the correction parameters for each state element estimated by the correction parameter estimation unit 207. When estimating a psychological state, the correction parameter storage unit 208 provides the correction parameters for the corresponding state element in response to a request from the psychological state estimation unit 209.

[0092] The psychological state estimation unit 209 corrects the biometric data obtained from the user whose psychological state is to be estimated or the estimation parameters used to estimate the psychological state, using the correction parameters for each state element stored in the correction parameter storage unit 208, and estimates the psychological state of the user whose psychological state is to be estimated based on the corrected biometric data or the corrected estimation parameters.

[0093] The psychological state estimation unit 209 estimates the psychological state by applying an existing estimation algorithm. Note that the psychological state estimation parameters for each state element used in estimating the psychological state may be stored in advance in the psychological state estimation unit 209 or may be acquired from another device.

[0094] The psychological state storage unit 210 stores data indicating the psychological state estimated by the psychological state estimation unit 209 together with data indicating a user ID and a time (period). The psychological state storage unit 210 stores data in which, for example, a user ID, a start time (the time when the estimated psychological state was reached), an end time (the time when the estimated psychological state was no longer reached), and information indicating the estimated psychological state are associated with each other.

[0095] (Method of Storing Correction Parameters) Next, a method of storing correction parameters will be described with reference to Fig. 11. Fig. 11 is a diagram for explaining a method of storing correction parameters in the present invention. Methods of storing correction parameters include a method of storing the correction parameters as a mathematical formula, and a method of dividing the correction parameters into a certain range and storing the divided values ​​in association with the correction values.

[0096] Assume that an estimation result indicating the influence on biological data x after the start of state element a is obtained, as shown in the upper left diagram of Fig. 11. The correction parameter estimation unit 207 stores the parameters of the formula for the correction value in a table, as shown in the upper right diagram of Fig. 11. Alternatively, the correction parameter estimation unit 207 divides the change due to the state into stages, and stores the period and correction value of each stage in a table, as shown in the lower diagram of Fig. 11.

[0097] (Operation Example) The following describes the flow of various processes performed by the estimation device 20. The various processes performed by the estimation device 20 include a data acquisition process for acquiring various data, a correction parameter estimation process for estimating correction parameters, and a psychological state estimation process for estimating a psychological state.

[0098] 12 is a flowchart showing the flow of data acquisition processing performed by the estimation device 20 according to the embodiment. Note that the operation of the estimation device 20 shown in the flowchart of FIG. 12 is an example, and the order of some of the operation steps may be changed.

[0099] First, the personal data acquisition unit 201 and the environmental information acquisition unit 204 acquire various sensor data (step S101). For example, the personal data acquisition unit 201 acquires various data from the personal terminal 11. The personal data acquisition unit 201 outputs the acquired various data to the biometric information analysis unit 202 and the user situation analysis unit 205. The environmental information acquisition unit 204 acquires video data and audio data from the environmental information terminal 12. The environmental information acquisition unit 204 outputs the acquired video data and audio data to the user situation analysis unit 205.

[0100] The biometric information analysis unit 202 acquires biometric values ​​that can be used to estimate the psychological state from the personal data acquisition unit 201, analyzes the acquired biometric values, and calculates analytical data that can be used to analyze the psychological state (step S102). The biometric information analysis unit 202 stores the acquired biometric values, the calculated analytical data, etc. as biometric data in the biometric information storage unit 203 in association with the user ID (step S103).

[0101] The user status analysis unit 205 acquires activity information usable for analyzing the user's activity status (e.g., data indicating the user's movement speed, the user's exercise type, and exercise load level calculated from sensor values ​​of the positioning sensor and acceleration / angular velocity sensor) from the personal data acquisition unit 201. Furthermore, the user status analysis unit 205 analyzes various data obtained from the personal terminal 11, the environmental information terminal 12, etc., and generates status element history data indicating the history of status elements of users U1 and U2 (step S104). The user status analysis unit 205 associates the generated status element history data with the user ID and stores it in the user status storage unit 206 (step S105).

[0102] 13 is a flowchart showing the flow of the process of estimating the effect of a status element on biological data, performed by the estimation device 20 according to the embodiment. Note that the operation of the estimation device 20 shown in the flowchart of FIG. 13 is an example, and the order of some of the operation steps may be changed.

[0103] The correction parameter estimation unit 207 acquires a list of status elements (step S201). Next, the correction parameter estimation unit 207 repeatedly executes the processes from loop L1s to loop L1e. Loop L1s to loop L1e are repeated until the effects of all status elements shown in the acquired list of status elements on the biometric data are estimated.

[0104] The correction parameter estimation unit 207 acquires information indicating the time period of each state element from the user state storage unit 206 (step S202). Note that the correction parameter estimation unit 207 may acquire information indicating the time period of one state element from the user state storage unit 206, estimate the influence of the one state element, and then acquire information indicating the time periods of other state elements.

[0105] Next, the correction parameter estimation unit 207 repeatedly executes the processes from loop L2s to loop L2e. Loops L2s to L2e are repeated until the influence of the acquired status elements on the biometric data is estimated for each time period. The correction parameter estimation unit 207 determines whether the duration is sufficient based on the acquired information indicating the time period of a certain status element (step S203). A sufficient duration means that a certain status element has continued for a predetermined time or more. An insufficient duration means that a certain status element has not continued for a predetermined time or more. If the correction parameter estimation unit 207 determines that the duration is insufficient (step S203-NO), the correction parameter estimation unit 207 acquires information indicating another time period and executes the process of step S203 again.

[0106] If the correction parameter estimation unit 207 determines that the duration is sufficient (step S203—YES), the correction parameter estimation unit 207 repeatedly executes the processes from loop L3s to loop L3e. Loops L3s to L3e are repeated until they are executed for each type of biological data. The types of biological data include, for example, heart rate sensor data, pulse wave sensor data, and electrodermal activity sensor data. The correction parameter estimation unit 207 acquires the biological data for the corresponding time period from the biological information storage unit 203 (step S204). Based on the acquired biological data for the time period, the correction parameter estimation unit 207 calculates the influence of the state element from the fluctuation of the biological data (step S205).

[0107] When the processing from loop L2s to loop L2e is completed, the correction parameter estimation unit 207 calculates correction parameters as state elements (step S206). The correction parameter estimation unit 207 stores the calculated correction parameters in the correction parameter storage unit 208 (step S207). When the processing from loop L1s to loop L1e is completed, the correction parameter estimation unit 207 ends the processing of FIG. 13.

[0108] Fig. 14 is a flowchart showing the flow of the psychological state estimation process performed by the estimation device 20 according to the embodiment. Note that the operation of the estimation device 20 shown in the flowchart of Fig. 14 is an example, and the order of some of the operation steps may be changed.

[0109] The psychological state estimation unit 209 acquires biometric data for a time period to be subjected to psychological state estimation (referred to as a "target time period") from the biometric information storage unit 203 (step S301). The psychological state estimation unit 209 acquires information indicating state elements of the user U for the target time period from the user situation storage unit 206 (step S302). The psychological state estimation unit 209 repeatedly executes the processes from loop L4s to loop L4e. Loops L4s to L4e are repeated until they have been executed for all types of biometric data.

[0110] The psychological state estimation unit 209 acquires correction parameters for state elements that affect the time period from the correction parameter storage unit 208 (step S303). The psychological state estimation unit 209 calculates the total value of correction parameters that affect the time period based on the acquired correction parameters for the state elements (step S304). For example, it is possible that multiple state elements occur depending on the time period. Therefore, the psychological state estimation unit 209 calculates the total value of correction parameters by combining the correction parameters for all state elements that affect the time period.

[0111] The psychological state estimation unit 209 corrects the biometric data obtained from the user using the calculated total value of the correction parameters (step S305). The psychological state estimation unit 209 estimates the user's psychological state based on the corrected biometric data (step S306). Specifically, the psychological state estimation unit 209 estimates the user's psychological state using the corrected biometric data and estimation parameters corresponding to the state elements. The psychological state estimation unit 209 stores the estimation result of the user's psychological state in the psychological state storage unit 210 (step S307).

[0112] In the processing of Figure 14, a configuration has been shown in which the psychological state estimation unit 209 corrects biometric data and estimates the user's psychological state based on the corrected biometric data, but the psychological state estimation unit 209 may correct estimation parameters instead of the biometric data and estimate the user's psychological state based on the corrected estimation parameters and the biometric data.

[0113] The estimation device 20 configured as described above includes a personal data acquisition unit 201 that acquires biometric data related to the user's biometrics, an environmental information acquisition unit 204 that acquires environmental information including at least one of video and audio around the user, a correction parameter estimation unit 207 that estimates, as a correction parameter, an estimation result of the influence on the biometric data after the start of a state element indicating an action or situation that affects the biometric data or after the end of the state element, based on the biometric data and the environmental information, and a psychological state estimation unit 209 that corrects the biometric data obtained from the user or the estimation parameters used to estimate the psychological state, using the correction parameters estimated by the correction parameter estimation unit 207, and estimates the user's psychological state based on the corrected biometric data or the corrected estimation parameters.

[0114] With this configuration, the estimation device 20 estimates, from a user's state, the impact on biometric data caused by a state transition due to the start of another state element or the end of a state element. The estimation device 20 then corrects the biometric data or estimation parameters using the estimation results (correction parameters). This allows the user's psychological state to be estimated taking into account the impact on biometric data caused by a state transition due to the start of another state element or the end of a state element. This improves the accuracy of estimating the user's psychological state.

[0115] (Variation 1) In the above-described embodiment, the configuration was described in which the influence of each "status element" is calculated based on past biometric data. In contrast, it is also conceivable to define a status element such that, for example, a single status element (e.g., status element x) never initiates, but only a combination of multiple status elements (e.g., status elements a, b, and c) always initiates. For example, if "in a meeting" is status element x, "online meeting" is status element a, "conference room 1" is status element b, and "conference room 2" is status element c, then "in a meeting" never initiates alone, but only a combination of "online meeting," "conference room 1," and "conference room 2," etc., may always initiate.

[0116] Even if multiple state elements occur simultaneously, if there is a time difference between the start timing of each state element (for example, if the time difference is equal to or greater than a predetermined threshold), it is possible to estimate the impact of each state element individually, as shown in (A) of Figure 15. For example, if state element y starts after state element x starts, the correction parameter estimation unit 207 can estimate the impact of state element x, and then estimate the impact of state element y. Furthermore, if such "simultaneous starts" only occur occasionally, the respective effects can be calculated by "utilizing data that do not occur simultaneously" (for example, by excluding data that occurs at the same time from the calculation).

[0117] On the other hand, as shown in (B) of Figure 15, if there is no time difference in the start timing of each status element (for example, if the time difference is less than a predetermined threshold), each of the multiple status elements x, a, b, and c is not independent, and the combination of the multiple status elements that occur can be treated as one status element, and the effects of the status elements x, a, b, and c can be estimated based on past data of the biological data.

[0118] 15B , if there is no time difference between the start timings of state element x and state element y, the correction parameter estimation unit 207 may treat state element x and state element y as a single state element and estimate the influence of the single state element x + y. For example, instead of treating "in a meeting," "online meeting," "conference room 1," and "conference room 2" as independent state elements, it is possible to calculate the influence by treating "in a meeting & online meeting," "in a meeting & conference room 1," and "in a meeting & conference room 2" as new state elements.

[0119] (Variation 2) In the above-described embodiment, the influence of each "status element" is calculated based on past biometric data. However, it is also possible that the end of a status element always coincides with the start of another status element. For example, this is the case when the end of a specific status always coincides with the start of another status. As shown in the left diagram of FIG. 16 , this is the case when the end of status element a coincides with the start of status element b or status element c.

[0120] In such a case, the correction parameter estimation unit 207 can combine the ending state element and the starting state element to use as a new single state element, as shown in the middle diagram of Figure 16, and can also independently estimate and use the effect of starting or ending each state element by using an average from or to each state, as shown in the right diagram of Figure 16.

[0121] Part or all of the configuration of the estimation device 20 and the notification device 30 in the above-described embodiments may be implemented by a computer. In this case, a program for implementing this function may be recorded on a computer-readable recording medium, and the program may be loaded into a computer system and executed. Note that the term "computer system" as used herein includes hardware such as an OS and peripheral devices. Furthermore, "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as storage devices such as hard disks built into computer systems. Furthermore, "computer-readable recording medium" may also include devices that dynamically store programs for a short period of time, such as communication lines used when transmitting programs via networks such as the Internet or telephone lines, or devices that store programs for a certain period of time, such as volatile memory within a computer system that serves as a server or client. The program may be designed to implement part of the above-described functions, or may be capable of implementing the above-described functions in combination with a program already stored in the computer system, or may be implemented using a programmable logic device such as an FPGA (Field Programmable Gate Array).

[0122] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention.

[0123] The present invention can be applied to a technique for estimating a user's psychological state.

[0124] REFERENCE SIGNS LIST 10...sensor terminal group, 11...personal terminal, 12...environmental information terminal, 20...estimation device, 30...notification device, 100...estimation system, 201...personal data acquisition unit, 202...biological information analysis unit, 203...biological information storage unit, 204...environmental information acquisition unit, 205...user situation analysis unit, 206...user situation storage unit, 207...correction parameter estimation unit, 208...correction parameter storage unit, 209...mental state estimation unit, 210...mental state storage unit

Claims

1. An estimation device comprising: a personal data acquisition unit that acquires information about the location and activity status of each person and biometric information about the living body; an environmental information acquisition unit that acquires environmental information including at least one of video and audio around the person; a correction parameter estimation unit that estimates, as a correction parameter, an estimated result of the influence on the biometric information after the start of a state element indicating an action or situation that affects the biometric information or after the end of a state element, based on the information acquired by the personal data acquisition unit and the environmental information acquired by the environmental information acquisition unit; and a psychological state estimation unit that corrects the biometric information obtained from the person or an estimation parameter used to estimate the psychological state, using the correction parameter estimated by the correction parameter estimation unit, and estimates the psychological state of the person based on the corrected biometric information or the corrected estimation parameter.

2. The estimation device according to claim 1, wherein the correction parameter estimation unit estimates the correction parameter after the start of the state element using a first method using a simple average or a second method using a combination of prediction and weighting, based on multiple pieces of biometric information for each state element of the person, and estimates the correction parameter after the end of the state element using the first method, the second method, or a third method that takes into account the influence of the state element after it has started.

3. The estimation device according to claim 1 or 2, wherein, when there is a predetermined period difference between the start timing or end timing of the plurality of state elements, the correction parameter estimation unit estimates and superimposes the correction parameters after the start of each state element or after the end of each state element, thereby estimating the correction parameters after the start of each state element or after the end of each state element.

4. A computer program for causing a computer to execute the following steps: a personal data acquisition step for acquiring information regarding the location and activity status of each person and biometric information regarding the living body; an environmental information acquisition step for acquiring environmental information including at least one of video and audio around the person; a correction parameter estimation step for estimating, as a correction parameter, an estimated result of the influence on the biometric information after the start of a state element indicating an action or situation that affects the biometric information or after the end of a state element, based on each piece of information acquired in the personal data acquisition step and the environmental information acquired in the environmental information acquisition step; and a psychological state estimation step for correcting the biometric information obtained from the person or an estimation parameter used to estimate the psychological state, using the correction parameter estimated in the correction parameter estimation step, and estimating the psychological state of the person based on the corrected biometric information or the corrected estimation parameter.

Images