JP6965525B2 - Emotion estimation server device, emotion estimation method, presentation device and emotion estimation system - Google Patents

Description

The present invention relates to an emotion estimation server device, an emotion estimation method, a presentation device, and an emotion estimation system.

Conventionally, there is a technique of externally measuring the behavior or physiological reaction of a person using various sensors and estimating the internal psychological state (for example, emotion) of the person from the obtained data. Such emotion estimation techniques are used to improve the performance of human-machine interaction of machines that provide interpersonal services, such as service robots, or in physically distant environments such as telework. It is also used to help understand the situation of the other person in.

As an example, Patent Document 1 proposes an emotion estimation device that estimates the emotion of a person using a plurality of measurement index data such as a camera, a microphone, and a biological signal.

There are many existing techniques for estimating the emotion of a "single" person (hereinafter, "individual emotion"), such as the technique described in Patent Document 1. On the other hand, there are few technologies that estimate the emotional atmosphere of a place created by the collaborative activities of "multiple" people (hereinafter referred to as "team emotion") and support user sharing of the estimated information.

As an example, in Patent Document 2, it is determined whether or not the biometric information received from the biometric sensors of a plurality of individuals is the biometric information whose features have been extracted, and the feature information of the biometric information determined to have been feature extracted is used. A group emotion estimation device that estimates group emotion information based on this has been proposed.

Japanese Unexamined Patent Publication No. 2012-59107 Japanese Unexamined Patent Publication No. 2016-7363

In a modern society where all mechanisms are highly complicated, there is a limit to the results that can be produced by the work of one person alone, and naturally there are many collaborative work by multiple teams. Even in the above-mentioned emotion estimation technology, there will be an increasing need for a method of estimating not only individual emotions but also team emotions cultivated by the collaborative activities of multiple people.

Therefore, it is required to provide a technique capable of estimating team emotions with higher accuracy.

In order to solve the above problem, according to a certain viewpoint of the present invention, the first sensor data corresponding to the communication behavior between a plurality of users or the communication history data including the user-identifiable information regarding the communication behavior Based on the social relationship information management unit that generates information indicating the social relationship, and each combination of the plurality of target users selected by a part of the plurality of users based on the information indicating the social relationship. A team emotion estimation unit that calculates the corresponding weighting coefficient and estimates the team emotion based on the second sensor data of each of the plurality of target users or the information based on the second sensor data and the weighting coefficient. An emotion estimation server is provided.

The emotion estimation server includes an individual emotion estimation unit that estimates individual emotions for each target user based on the second sensor data of each of the plurality of target users, and the team emotion estimation unit is the second sensor. Information indicating the personal feeling may be obtained as information based on the data.

The team emotion estimation unit may obtain feature data of the second sensor data of each of the plurality of target users as information based on the second sensor data.

The team emotion estimation unit calculates a weighting coefficient based on the information indicating the social relationship, and based on the second sensor data or the information based on the second sensor data and the weighting coefficient, the team Emotions may be estimated.

The team emotion estimation unit weights the second sensor data or information based on the second sensor data by the weighting coefficient, and the average value, median value, or mode of the weighted data. The value may be calculated as the team sentiment.

The social relationship information management unit generates social network information based on the first sensor data or the communication history data, and generates information indicating the social relationship based on the social network information. May be good.

The social network information includes a degree indicating the number of other users having a relationship with each of the plurality of users, and the social relationship information management unit generates information indicating the social relationship based on the order. You may.

The social network information includes the strength of the relationship between the two of the plurality of users, and the social relationship information management unit may generate information indicating the social relationship based on the strength.

The social network information includes a cluster coefficient of a social network representing the relationship between the plurality of users, and the social relationship information management unit generates information indicating the social relationship based on the cluster coefficient. May be good.

Further, according to another aspect of the present invention, the social relationship is based on the first sensor data corresponding to the communication behavior between a plurality of users or the communication history data including the information that can identify the user regarding the communication behavior. Based on the information indicating the sex and the information indicating the social relationship, a part of the plurality of users calculates the weighting coefficient corresponding to each combination of the selected plurality of target users, and the plurality of users. Provided is an emotion estimation method including estimating team emotions based on the second sensor data of each of the target users or the information based on the second sensor data and the weighting coefficient.

Further, according to another aspect of the present invention, it is generated based on the first sensor data corresponding to the communication behavior between a plurality of users or the communication history data including the information that can identify the user regarding the communication behavior. A weighting coefficient corresponding to each combination of the plurality of target users selected by some of the plurality of target users calculated based on the information indicating the social relationship, and the second sensor data of each of the plurality of target users or A presentation device is provided that includes a presentation unit that presents information indicating an estimated team sentiment based on the information based on the second sensor data.

The presentation unit may present information indicating the team emotions of each of the plurality of teams in a region closer to each other as the connection between the teams is stronger.

The presenting unit may present information indicating the team emotion and individual emotions of each of the plurality of target users for each emotion category.

Further, according to another aspect of the present invention, the social relationship is based on the first sensor data corresponding to the communication behavior between a plurality of users or the communication history data including the information that can identify the user regarding the communication behavior. A social relationship information management unit that generates information indicating sex, and a weighting coefficient corresponding to each combination of a plurality of target users selected by a part of the plurality of users based on the information indicating the social relationship. Emotion estimation including a team emotion estimation unit that calculates and estimates team emotions based on the second sensor data of each of the plurality of target users or information based on the second sensor data and the weighting coefficient. An emotion estimation system is provided that includes a server and a presentation device including a presentation unit that presents information indicating the team emotion.

As described above, according to the present invention, it is possible to provide a technique capable of estimating team emotions with higher accuracy.

It is explanatory drawing which shows an example of the schematic structure of the emotion estimation system which concerns on this embodiment. It is a block diagram which shows an example of the hardware configuration of the emotion estimation server, PBX, a sensor terminal, and an information presentation terminal which concerns on this embodiment. It is a block diagram which shows an example of the functional structure of PBX which concerns on the same embodiment. It is a block diagram which shows an example of the functional structure of the emotion estimation server which concerns on the same embodiment. It is a block diagram which shows an example of the functional structure of the information presenting terminal which concerns on this embodiment. It is explanatory drawing which shows an example of the abstracted social network. It is explanatory drawing which shows an example of the abstracted social network. It is explanatory drawing which shows an example of the abstracted social network. It is explanatory drawing which shows an example of the information which shows the social relationship based on the communication behavior among users. It is explanatory drawing which shows an example of the information which shows the individual feeling and the information which shows the team feeling in a certain organization. It is explanatory drawing which shows an example of the application screen of an information presenting terminal. It is a flow chart which shows an example of the schematic flow of the information processing of the emotion estimation system which concerns on this embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the present specification and the drawings, components having substantially the same functional configuration are designated by the same reference numerals, so that duplicate description will be omitted.

Further, in the present specification and the drawings, a plurality of components having substantially the same or similar functional configurations are distinguished by adding different alphabets after the same reference numerals. However, if it is not necessary to distinguish each of the plurality of components having substantially the same or similar functional configurations, only the same reference numerals are given.

First, the above-mentioned problems of Patent Document 2 will be described. The above-mentioned Patent Document 2 describes a method of estimating group emotional information from individual emotions, but has two major problems described below.

The first problem is that Patent Document 2 does not explain at all how to define and select the constituent persons of the group for which emotions are estimated. For example, in a huge organization with thousands or tens of thousands of people, the number of people who belong to the same organization but have no contact in daily life and do not know their names or faces. Would be far more than the number of acquaintances. Even if the group's emotions are combined and calculated for people who do not interact with each other, the information indicating the team's emotions will have little meaning (because the persons are not a team).

The second problem is that Patent Document 2 does not explain that when calculating the emotions of a group, the estimation process is performed with the "social relationship" between the members of the group as an influential factor. Is. For example, in a large business project in which dozens to hundreds of people are involved at the same time, even if it is a collaborative relationship, there is also a strong social relationship in which colleagues in the same team collaborate every day. If so, there may be a weak social relationship in which the departments to which they belong are different and they have only a few points of contact each month. Naturally, the former person with a strong social relationship has a greater influence on each other in team emotions than the latter case with a weak social relationship, and the influence on the estimation of team emotions. It is considered that setting a large degree contributes to the improvement of estimation accuracy. Patent Document 2 describes a method of using the time and place information from which the characteristic information of the sensor data is extracted for weighting the emotion calculation of the group, but the above-mentioned reference to the social relationship and its estimation processing There is no description to use it.

In order to solve the above two problems of Patent Document 2 with respect to the team emotion estimation technique, whether or not the constituent persons who are the elements of the team emotion of the group have a social relationship with each other, and further, the social relationship. There is a need for a mechanism in which the estimation system uses information on sexual strength to estimate team emotions.

Therefore, the present invention relates to a technique related to an estimation process of "team emotion" based on individual emotions, and weights are weighted between the constituent persons in the selection of the constituent persons of the group which are the elements of the team and the calculation of the team emotions. It realizes an emotion estimation system that can calculate highly accurate team emotions by processing based on the existence and strength of "social relations", and further, the social relations are among the constituent persons. Parameters for estimation calculation based on the data of "social network" information calculated by the "social relationship information management department" of the system from the sensor data of communication behavior that occurs on a daily basis and the communication history data of the communication support function server. It is an object of the present invention to provide a team emotion estimation system characterized in that the team emotion is estimated by being converted into data.

In particular, the present invention is characterized in that it uses information indicating "informal" social relationships based on daily communication behavior.

Subsequently, with reference to FIG. 1, a schematic configuration of a team emotion estimation system (hereinafter, “emotion estimation system”) according to the embodiment of the present invention will be described.

FIG. 1 is an explanatory diagram showing an example of a schematic configuration of an emotion estimation system according to the present embodiment. Referring to FIG. 1, the emotion estimation system includes an emotion estimation server 100, a PBX 300, a plurality of users 900, a plurality of sensor terminals 200, a presenting device (hereinafter, also referred to as an “information presenting terminal”) 400, and a LAN (Local Area Network). ) 50 is included.

The PBX 300 does not have to be newly prepared for this emotion estimation system. As the PBX300, for example, an existing server facility having a communication support function and a scheduler function such as PBX (Private Branch eXchange) on an existing in-house information infrastructure existing on the in-house intra-network (or on the cloud network) is used. You may.

The plurality of users 900 have a social relationship with each other (details will be described later).

FIG. 2 shows the "book" without distinguishing each of the emotion estimation server 100, PBX 300, sensor terminal 200, and information presentation terminal 400 (hereinafter, emotion estimation server 100, PBX 300, sensor terminal 200, and information presentation terminal 400) according to the present embodiment. It is a block diagram which shows an example of the hardware structure of (the device which concerns on embodiment"). It should be noted that it is not necessary for all of the above devices to be equipped with all of the following hardware configurations (for example, the emotion estimation server 100 does not need to be directly equipped with a sensor), and for each device described later. A hardware module capable of realizing a functional configuration may be provided with an appropriate limitation.

Referring to FIG. 2, the apparatus according to the present embodiment includes a bus 801 and a CPU (Central Processing Unit) 803, and a ROM (Read Only).
It includes a Memory) 805, a RAM (Random Access Memory) 807, a storage device 809, a communication interface 811 and a sensor 813, an input device 815, a display device 817, and a speaker 819.

The CPU 803 executes various processes in the apparatus according to the present embodiment. Further, the ROM 805 stores a program and data for causing the CPU 803 to execute the processing in the apparatus according to the present embodiment. Further, the RAM 807 temporarily stores the program and the data when the processing of the CPU 803 is executed.

The bus 801 connects the CPU 803, the ROM 805, and the RAM 807 to each other. A storage device 809, a communication interface 811, a sensor 813, an input device 815, a display device 817, and a speaker 819 are further connected to the bus 801. Bus 801 includes, for example, a plurality of types of buses.

As an example, bus 801 includes a high-speed bus connecting the CPU 803, ROM 805, and RAM 807, and one or more other buses slower than the high-speed bus.

The storage device 809 stores data to be temporarily or permanently stored in the device according to the present embodiment. The storage device 809 may be, for example, a magnetic storage device such as a hard disk, or an EEPROM (Electrically).
Erasable and Programmable Read Only Memory, flash memory
memory), MRAM (Magnetoresistive Random Access Memory), FeRAM (Ferroelectric Random Access Memory) and PRAM (Phase)
Nonvolatile memory (nonvolatile) such as change Random Access Memory)
memory) may be used.

The communication interface 811 is a communication means included in the device according to the present embodiment, and communicates with an external device via a network (or directly). The communication interface 811 may be an interface for wireless communication, and in this case, for example, a communication antenna, an RF circuit, and other circuits for communication processing may be included. Further, the communication interface 811 may be an interface for wired communication, and in this case, for example, a LAN terminal, a transmission circuit, and other circuits for communication processing may be included.

Sensor 813 is, for example, a camera, microphone, other sensor, or a combination thereof. The camera captures a subject and includes, for example, an optical system, an image sensor, and an image processing circuit. A microphone picks up ambient sound, converts the sound into an electric signal, and converts the electric signal into digital data.

The input device 815 is a touch panel, a mouse, a line-of-sight detection device, and the like. The display device 817 displays an output image (that is, a display screen) from the device according to the present embodiment, and is, for example, a liquid crystal display, an organic EL (Organic Light-Emitting Diode), or a CRT (Cathode).
It can be realized by using Ray Tube) or the like. The speaker 819 outputs voice, converts digital data into an electric signal, and converts the electric signal into voice.

Next, an example of the functional configuration of the "sensor terminal 200" according to the present embodiment will be described.

The sensor terminal 200 externally measures the behavior and physiological reaction of the user 900, and transmits the data acquired by the measurement to the emotion estimation server 100 described later. The behavioral and physiological reaction data measured by the sensor terminal 200 and the sensor 813 are, for example, image data including facial expressions and body posture state contents measured by a camera, voice data measured by a microphone, and measurement by an acceleration sensor. Acceleration data of body movements, device operation data measured by input devices such as keyboards, mice, and touch panels, and autonomic nervous system activity (heartbeat activity, skin electrical activity, blood pressure, sweating, breathing, etc.) measured by various biosensors. Includes data on skin / core body temperature, etc.), data on central nervous system activity (brain waves, cerebral blood flow, etc.), and data on line-of-sight movement, pupil diameter, number of blinks, etc. measured by a line-of-sight measuring device.

The sensor of the sensor terminal 200 may be a non-contact type or a contact type to the user. When the sensor of the sensor terminal 200 is a contact type, for example, the sensor terminal 200 may be a wearable terminal in the shape of a wristwatch or the like. Further, the sensor terminal 200 may be a smartphone terminal including a camera, a microphone, an acceleration sensor, a communication unit, and the like. It should be noted that the sensor terminal 200 is shown in FIG. 1 so that there is one sensor terminal 200 corresponding to one user 900. However, one sensor terminal 200 may acquire sensor data of a plurality of users 900. In that case, the sensor terminal 200 performs a one-to-one correspondence process of the sensor data and the identification data of the measurement target person for each of the plurality of users 900, and obtains each combination of the sensor data and the identification data of the measurement target person. Remember or send. The identification data of the person to be measured may be obtained in any way, and as an example, it may be identification data of a person identified by face recognition from an image captured by a camera.

Further, the sensor terminal 200 measures sensor data corresponding to a communication action performed by the user 900 with another user 900. The communication behavior is, for example, a face-to-face conversation behavior. In order to sense the communication behavior, the sensor terminal 200 uses information on changes in the surrounding environment of a user 900, such as generation and continuation of voice around the user 900, approach / detachment and stay of another person, and user's. It is possible to detect changes in body posture, movement of places, and the like. These can be realized by a microphone, a camera, an acceleration sensor, or the like of the sensor 813 provided as hardware in the sensor terminal 200 as described above.

Further, the sensor terminal 200 may be provided with an infrared transmitter / receiver, a wireless communication device such as WiFi (registered trademark) or GPS (Global Positioning System) as hardware. With these hardware functions, information indicating the relationship between the mutual positions and orientations of the bodies of the plurality of users 900 can be acquired based on the communication between the sensor terminals 200 worn by the plurality of users 900 on their bodies.

In summary, the sensor data corresponding to the communication behavior includes electronic data corresponding to information on who, who, when, where, and what kind of body position and orientation and conversation volume were used for the communication behavior. As a specific example, the sensor data corresponding to the communication behavior is "User 900A and user 900B are standing in front of each other in the passage V on the 4th floor from XX hours XX minutes to XY hours YY minutes on XX days in 20XX years. It is sensor data of behavior such as taking a communication action to talk.

It should be noted that the plurality of users 900 may not actually have a conversation or the like. For example, the state of a user (awareness cognitive state) who senses the atmosphere and situation of others around him through peripheral vision and noise may be regarded as one of communication behaviors in a broad sense. In that case, the sensor terminal 200 may use long-term time-lapse behavior data, for example, over several hours, as sensor data corresponding to the communication behavior.

Further, the sensor terminal 200 may include a storage unit for temporarily storing the measured data. Further, since the sensor terminal 200 may perform not only measurement but also simple analysis, it may be provided with a control unit for analysis calculation. Further, the sensor terminal 200 may include a communication unit for transmitting the measured or simple-analyzed data to the emotion estimation server 100. The storage unit may be mounted by the storage device 809, the control unit may be mounted by the CPU 803, ROM 805 and RAM 807, and the communication unit may be mounted by the communication interface 811.

Next, an example of the functional configuration of the “PBX300” according to the present embodiment will be described with reference to FIG.

FIG. 3 is a block diagram showing an example of the functional configuration of the PBX 300 according to the present embodiment. Referring to FIG. 3, the PBX 300 includes a communication unit 310, a storage unit 320, and a control unit 330. The PBX 300 supports data communication between users 900 using various communication tools. The various communication tools include, for example, IP (Internet Protocol) telephone, Web conferencing, application sharing, groupware, e-mail, chat, instant messaging, and in-house SNS (Social Networking Service). As mentioned above, the PBX 300 does not have to be newly prepared for this emotion estimation system. As the PBX 300, an existing server facility having a communication support function or a scheduler function such as a PBX or SIP (Session Initiation Protocol) server already owned by a user or an organization using this emotion estimation system may be used.

The communication unit 310 communicates with another device. For example, the communication unit 310 is directly connected to the LAN 50 and communicates with the sensor terminal 200 described above and the emotion estimation server 100 described later. Further, the communication unit 310 communicates with various communication tools and communication devices (PCs and smartphones) used by the user 900 in order to realize the above-mentioned communication support function. The communication unit 310 may be implemented by the communication interface 811.

The storage unit 320 stores communication history data related to the operation of the above-mentioned communication support function and scheduler function. The communication history data indicates information on the communication action performed via the PBX300, for example, who, who, when, where, what kind of tool was used, and what kind of communication action was taken. Includes electronic data. As a specific example, the communication history data is "a user 900A and a user 900B, 20XX year XX month XX day XX hour XX minute to YY hour YY minute, at their own seat, using an IP phone, content related to the theme ZZZ" communication. It is electronic data showing the communication history of use such as the action being performed via the PBX300.

As another example, the communication history data is communication of schedule information recorded on the cloud by the scheduler tool via the PBX300 itself or the PBX300, for example, who, who, when, where, what kind of content. Includes electronic data as to what was planned for. As a specific example, we are planning to hold a meeting with user 900B on the schedule of user 900A from XX hours XX minutes to YY hours YY minutes on XX days of 20XX in the W conference room. It is electronic data indicating the communication history of the schedule record such as "yes".

Using the above communication history data, the emotion estimation server 100 described later is recorded in the PBX 300 by the information regarding the electronic communication behavior between the users 900 via the communication of the PBX 300 and the scheduler tool input of the user 900. Communication behavior can be grasped based on at least one of the schedule information. The storage unit 320 may be mounted by the storage device 809.

The control unit 330 provides various functions of the PBX 300. Further, the control unit 330 transmits the communication history data stored in the above-mentioned storage unit 320 to the communication device via the communication unit 310 in response to a request from another communication device, for example, the emotion estimation server 100. Has a function. The control unit 330 may be mounted by the CPU 803, the ROM 805, and the RAM 807.

Next, an example of the functional configuration of the “emotion estimation server 100” according to the present embodiment will be described with reference to FIG. FIG. 4 is a block diagram showing an example of the functional configuration of the emotion estimation server 100 according to the present embodiment. Referring to FIG. 4, the emotion estimation server 100 includes a communication unit 110, a storage unit 120, and a control unit 130.

The communication unit 110 communicates with another device. For example, the communication unit 110 is directly connected to the LAN 50 and communicates with the sensor terminal 200 and the PBX 300. The communication unit 110 may be implemented by the communication interface 811.

The storage unit 120 stores programs and data for the operation of the emotion estimation server 100. The storage unit 120 may be mounted by the storage device 809.

The control unit 130 provides various functions of the emotion estimation server 100. The control unit 130 includes an individual emotion estimation unit 131, a social relationship information management unit 133, and a team emotion estimation unit 135. The control unit 130 may be mounted by the CPU 803, the ROM 805, and the RAM 807. The functions of the individual emotion estimation unit 131, the social relationship information management unit 133, and the team emotion estimation unit 135 will be described later.

Next, an example of the functional configuration of the "information presentation terminal 400" according to the present embodiment will be described with reference to FIG. FIG. 5 is a block diagram showing an example of the functional configuration of the information presentation terminal 400 according to the present embodiment. Referring to FIG. 5, the information presenting terminal 400 includes a communication unit 410, a storage unit 420, a control unit 430, an operation unit 440, and a presentation unit 450.

The communication unit 410 communicates with another device. For example, the communication unit 410 is directly connected to the LAN 50 and communicates with the emotion estimation server 100. For example, the communication unit 410 acquires information indicating team emotions from the emotion estimation server 100. The communication unit 410 may be implemented by the communication interface 811.

The storage unit 420 stores programs and data for the operation of the information presenting terminal 400. For example, the storage unit 420 temporarily stores information indicating team emotions acquired from the emotion estimation server 100 by the communication unit 410. The storage unit 420 may be mounted by the storage device 809.

The control unit 430 provides various functions of the information presentation terminal 400. For example, the control unit 430 visualizes the information indicating the team emotion, and controls the presentation unit 450 so that the information indicating the team emotion is presented by the presentation unit 450. The control unit 430 may be mounted by the CPU 803, the ROM 805, and the RAM 807.

The operation unit 440 receives an operation input from the user. When the operation unit 440 outputs the received operation to the control unit 430, the operation is used for various operations by the control unit 430. The operation unit 440 may be mounted by the input device 815.

The presentation unit 450 presents information to the user according to the control by the control unit 430. For example, the presentation unit 450 presents to the user information indicating team emotions visualized by the control unit 430. The presentation unit 450 may be mounted by the display device 817.

The functions of the individual emotion estimation unit 131, the social relationship information management unit 133, and the team emotion estimation unit 135 will be described. The personal emotion estimation unit 131 estimates personal emotions by estimating the personal emotions of each user 900 based on the measurement data of behaviors and physiological reactions acquired from the user 900 via the sensor terminal 200 and the communication unit 110. Generate information that indicates personal feelings). In addition, the personal emotion estimation unit 131 has a function of storing the generated data in the storage unit 120. In particular, the personal emotion estimation unit 131 may store information indicating personal emotions estimated from a user 900 in the storage unit 120 in association with the ID of the user 900 (see, for example, FIG. 10 described later). ).

Here, the explanation of personal feelings and their estimation methods is supplemented. As an example, personal emotions are "evaluative reactions to people, things, events, and the environment among various information processing performed by humans in the mental process" (Ortony et al., 1988; Ohira, 2010). Is defined as. Specific types of emotions include happiness, surprise, fear, anger, disgust, sadness, and psychologist James A. on a basic emotion-based discrete model that corresponds to facial expressions by psychologist Paul Ekman. The quadrants of emotions and sorrows in the emotional dimension-based continuous model of pleasure and alertness by Russel are known. Other continuous models include Watson's positive or negative emotions, Wundt's three-axis model (pleasure, excitement, tension), and Plutchik's four-axis model. Other applied / complex emotions include confusion, interest, mental stress, concentration, fatigue, busyness, creativity, relaxation / tension, motivation, empathy, and reliability. Furthermore, the liveliness that is experienced as a group atmosphere in business activities can be said to be a type of higher-level emotion.

The emotions and degree of a person can be determined by, for example, using the questionnaire method, by linguistic reporting of the person's letters, sentences, and symbols. As the questionnaire, "Affect Grid" and the like are well known.

However, since the measurement method using the questionnaire requires answering work, the measurement itself may interfere with the original purpose work in daily life where some other work such as work is performed. be.

Therefore, in this emotion estimation system, the personal emotion estimation unit 131 mechanically estimates and processes emotions (obtained by the questionnaire method or the like) based on the behavioral and physiological reaction data measured by the sensor terminal 200 described above. .. In order to perform the estimation process, the data of the emotion estimation model generated by the learning process in advance is required. The emotion estimation model is generated from, for example, a group of data in which the data of the behavior or physiological reaction at a certain time point / situation is associated with the learning data consisting of the answer data of the questionnaire. For example, ask questions about the user's facial expression, voice, heartbeat activity, skin electrical activity, and other behavioral / physiological data measured from countless cameras, microphones, and wearable activity meters embedded in the office, and the user's subjective emotions. The correct answer data that was answered on paper is associated with the data for learning. The behavior / physiological data may be feature data for learning processing in which the measured values from the sensors are converted.

The feature amount data may be the position of a representative feature point of the face, the distance of a straight line connecting each of the two points, or the angle formed. Alternatively, the feature data may be the fundamental frequency, power, average speech speed, maximum value and standard deviation of the first-order cepstrum coefficient of speech. Alternatively, the feature amount data may be the average value of the heart rate or the beat interval, the standard deviation, or the heart rate variability. Alternatively, the feature amount data may be an average value of the skin conductance level, a standard deviation, an increase / decrease rate, or the like. These feature data may be used in any way, as an absolute value at a certain time point, or as a relative rate of change between two time points.

For the generation of the emotion estimation model using the learning data, for example, a known SVM (Support Vector Machine) or deep learning method may be used as a learning method, or a simple regression analysis method may be used. May be used. Further, the learning model may be generated for each individual user, or a model common to humans may be generated using the learning data of a plurality of users. The individual emotion estimation unit 131 can estimate individual emotions from the behavior / physiological data of a certain person by using the obtained emotion estimation model data.

The emotion estimation server 100 and the individual emotion estimation unit 131 may have a function of generating learning data for the above-mentioned individual emotion estimation process and the emotion estimation model itself. Further, the above-mentioned feature amount data for the learning data is generated not by the emotion estimation server 100 but by the sensor terminal 200, and the sensor terminal 200 transmits the feature amount data to the emotion estimation server 100. You may.

The social relationship information management unit 133 is provided with information indicating social relationships based on social network (social NW) information, which is necessary for the team emotion estimation unit 135, which will be described later, to estimate and process team emotions (hereinafter, “social relationship information”). Also called). Therefore, the social relationship information management unit 133 acquires and acquires the above-mentioned "sensor data corresponding to the communication behavior by the sensor terminal 200" or the above-mentioned "communication history data by PBX 300" via the communication unit 110. It has a function of generating social NW information described later based on the data and storing the generated social NW information in the storage unit 120.

Here, "social NW information" will be described. Social networks can be defined as "people's connections defined by the quantity or quality of communication behavior with each other." In addition, the strength of the social network can be determined according to the size or height of the quantity or quality of the social network.

The structure of an abstracted social network is such that the user 900 belonging to it is represented as a circular node (element), and the relationship between the user 900 is represented as a line segment-shaped edge (relationship between elements). And functions can be shown as a simplified model. The model is based on indicators such as degree (number of edges connected to a node), strength (strength of connection between nodes), and cluster coefficient (presence ratio of triangular structure formed by nodes and edges). Its structure and function can be analyzed by the Social Relations Information Management Department 133.

6, 7, and 8 are explanatory diagrams showing an example of an abstracted social network. FIG. 6 shows an example of a social network in which the “order” index is reflected as a basic model of the social network.

In the figure, nodes 910A, 910B, 910C, 910D and 910E are present corresponding to the users 900A, 900B, 900C, 900D and 900E. In addition, an edge 920 exists as a part of each combination of the two nodes. In this example, Edge 920 indicates that there was a past occurrence of communication behavior. Since the user 900A has communicated with each of 900B, 900C, 900D and 900E in the past, there is an edge between node 910A and node 910B, 910C, 910D and 910E. On the contrary, since the user 900D has never communicated with the 900E, there is no edge between the node 910D and the node 910E.

Here, the order is the number of edges connected to a certain node. In the example shown in FIG. 6, the order of the node 910A is 4, and the order of the node 910E is 1. The social relations information management unit 133 can obtain the breadth of the interaction relations of a certain user as quantitative data by the size of the degree index.

More specifically, the social relationship information management unit 133 may determine that the higher the order of the user 900, the stronger the social relationship with the other user 900. For example, the social relationship information management unit 133 sets the social relationship of the user 900 having a degree of 0 as "none" and the social relationship of the user 900 whose order is lower than the first order as "weak", and sets the order as "weak". May be "medium" for the social relationship of the user 900 having the second order or higher and less than the second order, and "strong" for the social relationship of the user 900 having the second order or higher. The threshold value for the order does not have to be three. That is, the social relationship based on the order does not have to be four stages, may be two stages, may be three stages, or may be five or more stages. Further, the size of each threshold value is not limited.

FIG. 7 illustrates an example of a social network that reflects the “strength” index.

Here, the strength corresponds to the strength of the social network mentioned above. For example, strength refers to the frequency of communication behavior, the length of time it occurs, the novelty of the time of occurrence, the amount of communication content, the amount and depth of private content of communication content, and the media richness of the communication tool used. It is defined according to the degree, the volume and level of conversation, the distance between each other's bodies, the angle formed by the direction of each other's front of the body, the degree of positive / negative emotions, and so on.

The angle formed by the orientation of the front of the body may be detected in any way. For example, a plurality of infrared tags to which different direction IDs are assigned in advance may be attached in four directions of front, back, left, and right with the front of the body of each of the plurality of users in front. In such a case, the angle formed by the orientation of the front of the body may be detected by the direction ID exchanged by the infrared tag among the plurality of users. For details, see "Yoshi Nakamura and 4 others," Estimating Topological Position and Direction from Relative Positional Relationships Using Multiple Infrared Tags, "March 2007, IPSJ Journal, Vol. 48, No. 3 ”and so on. Further, as described in Japanese Patent No. 5907549, whether the positional relationship of a plurality of users is face-to-face or non-face-to-face may be detected.

For example, the social relations information management unit 133 may set the social relation between the two users 900 as "strong" when the conversation occurrence frequency between the two users 900 is equal to or higher than the first frequency. good. Further, when the frequency of conversation occurrence between the two users 900 is equal to or higher than the second frequency and lower than the first intensity, the social relations information management unit 133 states that the frequency between the two users 900 Social relations may be set to "medium". Further, when the frequency of conversation occurrence between the two users 900 is less than the second frequency, the social relationship information management unit 133 may set the social relationship between the two users 900 to "none". good.

As an example, when the frequency of conversation occurrence in the last 10 days is used as an index, the social relations information management unit 133 considers the social relation to be "strong" if the number of conversation occurrences is 10 or more, and the number of conversation occurrences. If the number of times is 5 or more, the social relationship is "medium", if the number of conversations is 1 or more, the social relationship is "weak", and if the number of conversations is 0, the social relationship is "medium". It may be "none". In FIG. 7, the edge 920AB between the node 910A and the node 910B is thick with the social relationship “strong”, the edge 920AE between the node 910A and the node 910E is slightly thick with the social relationship “medium”, and the others. Edge 920 is illustrated thinly with social relations "weak".

The period regarding the frequency of conversation occurrence is not limited to the last 10 days. For example, the period regarding the frequency of conversation occurrence may be the latest one month or another period. Further, the threshold value regarding the number of conversation occurrences does not have to be three. That is, the social relationship based on strength does not have to be four stages, may be two stages, may be three stages, or may be five or more stages. Further, the size of each threshold value is not limited.

Further, here, a method of obtaining the strength based on the frequency of occurrence of communication behavior has been described as an example, but similarly, a social relationship based on the above other information can be obtained. At this time, the longer the communication behavior occurs, the stronger the social relationship may be. In addition, the newer the time of occurrence of communication behavior, the stronger the social relationship. In addition, the greater the amount of communication content, the stronger the social relationship.

In addition, the larger the amount of private communication content, the stronger the social relationship. In addition, the deeper the content related to private life, the stronger the social relationship. In addition, the stronger the degree of media richness of the communication tool used, the stronger the social relationship. In addition, the lower the volume of conversation, the stronger the social relationship. In addition, the lower the sound of conversation, the stronger the social relationship. In addition, the closer the distance between the bodies, the stronger the social relationship. In addition, the smaller the angle formed by the orientation of each other's front of the body (that is, the closer to the non-opposite orientation), the stronger the social relationship may be. In addition, the stronger the degree of positive / negative emotions, the stronger the social relationship.

FIG. 8 shows an example of a social network in which the “cluster coefficient” index is reflected as another index example. The cluster coefficient is an index related to the size of the cluster property of the network, but since it is a known theory, detailed description thereof will be omitted here. As an example, the social relations information management unit 133 can obtain the breadth of the user's interaction relations as a whole social network as quantitative data by the size of the cluster coefficient.

More specifically, the social relations information management unit 133 may determine that the larger the cluster coefficient, the stronger the social relations of the social network as a whole. For example, the Social Relations Information Management Department 133 sets the social relations of the social network having a cluster coefficient of 0 as "none" and the social relations of the social network having a cluster coefficient lower than the first value as "weak". , The social relationship of the social network having the cluster coefficient of the second value or more and less than the second value may be set to "medium", and the social relationship of the social network having the cluster coefficient of the second value or more may be set to "strong". The threshold value for the cluster coefficient does not have to be three. That is, the social relationship based on the cluster coefficient does not have to be four stages, may be two stages, may be three stages, or may be five or more stages. Further, the size of each threshold value is not limited.

FIG. 9 is an explanatory diagram showing an example of information showing a social relationship based on communication behavior between users 900A to 900E. FIG. 9 shows a table 121 of social relationship information based on social NW information generated from communication behavior data between users 900A to 900E. As shown in FIG. 9, the table 121 of the social relationship information based on the social NW information shows, for example, an employee ID 1 that uniquely identifies a certain user 900, an employee ID 2 of another user 900, and the strength of the social relationship between the two. including. Here, it is assumed that the strength of social relations is set based on the "strength" index of the edge in the above-mentioned social network. Table 121 in FIG. 9 can be said to represent the social network shown and described in FIG. 7 in tabular form (for example, in the first row of FIG. 9, the edge 920AB between the nodes 910A and 910B in FIG. 7 is a social relationship. Corresponds to the expression shown thickly with the gender "strong").

In the example of FIG. 9, the strength of social relations is divided into four stages of "strong", "medium", "weak", and "none", but as already mentioned, social relations are classified based on strength. Of course, the number of sexual strengths may be greater or less than this, and the strengths of social relationships classified based on intensity may be continuous quantitative data. In addition, the threshold value used for associating and classifying the strength index of social NW information with the strength of social relations may be arbitrarily set in advance by the administrator of this emotion estimation system, and the setting parameters are stored in advance. It may be stored in the unit 120.

In the example shown in FIG. 9, the strength of social relations is set based on the strength index. In such an example, one strength of social relationship is set for two users 900. However, as described above, the strength of social relations may be set based on the degree index, or the strength of social relations may be set based on the cluster coefficient. When the strength of social relations is set based on the degree index, the strength of social relations is set for each user 900. On the other hand, when the strength of social relations is set based on the cluster coefficient, the strength of social relations is set for the entire social network.

The team emotion estimation unit 135 includes information indicating the above-mentioned individual emotions generated by the individual emotion estimation unit 131 and information indicating the social relationship based on the above-mentioned social NW information generated by the social relationship information management unit 133. Generate information that indicates team feelings based on it. Further, the team emotion estimation unit 135 has a function of storing the calculated information indicating the team emotion in the storage unit 120.

Here, the team feeling is, for example, a representative value in which information indicating individual feelings for each member in a preset group of users is weighted by the strength of social relations with the group, for example, an "average value". It is calculated as "median" and "mode". As an example, in the team emotion estimation unit 135, when the information indicating individual emotions is quantitative data and the average value is adopted as the representative value, and when the strength of social relations is set based on the intensity index. , As an example of the information indicating the team emotion, the team emotion value can be obtained by the following (Equation 1).

・・・（数式１）

... (Formula 1)

However, TE: team emotion value, m: number of combinations of two users targeted in the group, QR: weighting coefficient corresponding to the strength of social relations of the two users, RE: of the two users. It is a representative value (for example, "average value", "median value", "mode value") obtained from each individual emotion value of the user. At this time, the group of users to be calculated for the team emotion value may be all users 900A to 900E in the example of FIG. 6, or may be an arbitrary group of people partially combined and selected from among them. good. By the above-mentioned calculation method, if there is data of the basic individual emotion value and social NW information, the team emotion value can be obtained for an arbitrarily set group. Further, the number of combinations of the above-mentioned users does not necessarily have to be two, and may be a combination of three or more.

Here, the value of the weighting coefficient corresponding to the strength of each social relationship is not limited. For example, the stronger the social relationship, the larger the value of the weighting coefficient may be. More specifically, the weighting coefficient corresponding to the strength "strong" of social relations may be larger than the weighting coefficient corresponding to the strength "medium" of social relations. Further, the weighting coefficient corresponding to the strength of social relation "medium" may be larger than the weighting coefficient corresponding to the strength of social relation "weak". Further, the weighting coefficient corresponding to the strength of social relation "weak" may be larger than the weighting coefficient corresponding to the strength of social relation "none".

Here, the method of obtaining the team emotion value from the strength of social relations based on the strength index is shown. However, instead of the strength of social relations based on the intensity index, the team emotion value may be obtained from the strength of social relations based on the degree index. For example, the team emotion estimation unit 135 determines that the information indicating individual emotions is quantitative data and the average value is adopted as the representative value, and the strength of social relations is set based on the order index. As an example of the information indicating the team emotion, the team emotion value can be obtained by the following (formula 2).

・・・（数式２）

... (Formula 2)

However, TE: team emotion value, n: number of users targeted in the group, SR: weighting coefficient corresponding to the strength of social relations of each user, PE: individual emotion value.

In addition, the team emotion value may be obtained from the strength of social relations based on the cluster coefficient. For example, the team emotion estimation unit 135 determines that the information indicating individual emotions is quantitative data and the average value is adopted as the representative value, and the strength of social relations is set based on the cluster coefficient. As an example of the information indicating the team emotion, the team emotion value can be obtained by the following (formula 3).

・・・（数式３）

... (Formula 3)

However, TE: team emotion value, n: number of users targeted in the group, UR: weighting coefficient corresponding to the strength of social relations as a whole social network, PE: individual emotion value.

In the present embodiment, it is mainly assumed that the individual emotion estimation unit 131 first estimates the individual emotion, and then the team emotion estimation unit 135 estimates the team emotion as a collective data thereof. However, the team emotion estimation unit 135 may estimate the team emotion directly from the sensor data of a plurality of people targeted by the team. In that case, the team emotion estimation unit 135 may perform a weighting calculation process corresponding to the strength of the above-mentioned social relationship with respect to the sensor data value itself or its feature amount data (converted value).

At this time, the value of the sensor data itself or the feature amount data thereof is not particularly limited. As an example, the feature amount data of the sensor data may be the rate of change of the heart rate after being converted from the pulse detected by the pulse sensor. For example, the rate of change in heart rate can be the rate of change in heart rate, for example, the ratio of the current heart rate to the previous heart rate (eg, "heart rate increased by 20%", "heart rate increased". It may be represented by "10% decrease" etc.). The value of the sensor data itself or the feature amount data thereof may be used instead of the information indicating the individual emotion in estimating the team emotion.

FIG. 10 is an explanatory diagram showing an example of information showing individual feelings and information showing team feelings in a certain organization. FIG. 10 shows a table 122 of information indicating individual emotions and information indicating team emotions corresponding to an example of social NW information described with reference to FIG. 7. Here, the positiveness of quantitative data is illustrated as an example of information indicating individual emotions and information indicating team emotions.

Next, an example of the functional configuration of the "information presentation terminal 400" according to the present embodiment will be described. In the information presentation terminal 400, the control unit 430 acquires the information indicating the team emotion and the information indicating the individual emotion estimated by the emotion estimation server 100 via the communication unit 410, visualizes them, and uses the presentation unit 450. And present it to the user. The user may be a user 900 whose team emotions and individual emotions are estimated, or a person different from the user 900 whose team emotions and individual emotions are estimated (for example, an industrial physician in charge of mental health management of an organization). Etc.).

Further, when the control unit 430 visualizes and presents the information indicating the team emotion and the information indicating the individual emotion, the user can easily understand the information by using the information indicating the social relationship based on the above-mentioned social NW information. As described above, the information indicating the team emotions and the information indicating the individual emotions may be "categorized" and presented. For example, when the control unit 430 displays information indicating the team emotions of a certain team on the screen, the control unit 430 combines the information indicating the individual emotions of the members of the team with the information in the vicinity of the information indicating the team emotions, and the presentation unit 450. May be presented via. As a result, the user of the information presenting terminal 400 can intuitively grasp the information indicating the team emotion and the information indicating the individual emotion which is the calculation element thereof.

Further, for example, when the control unit 430 displays information indicating a plurality of team emotions by a plurality of teams on the screen, the team has a strong connection by communication behavior between the members based on the information indicating the social relationship. Information indicating each other's team feelings may be presented via the presentation unit 450 in a region closer to each other. As a result, the user of the information presenting terminal 400 can efficiently view the information indicating the team emotions of the plurality of teams based on the strength of the social relationship between the teams.

FIG. 11 is an explanatory diagram showing an example of an application screen of the information presentation terminal 400. With reference to FIG. 11, information indicating a certain team feeling of a certain team (positive degree of team T1) and information indicating individual feelings as a calculation factor thereof (positive degree of users U11 to U14 belonging to team T1) are close to each other. Information indicating another team feeling of the same team (team T1 fatigue level) and information indicating individual feelings (fatigue degree of users U11 to U14 belonging to team T1) displayed in the area of It is displayed in the nearby area. In this way, the presentation unit 450 presents the information indicating the team emotion and the information indicating the individual emotion for each emotion category, so that the team emotion and the individual emotion can be easily grasped for each category.

As shown in FIG. 11, the control unit 430 may have the past estimated information of team emotions and individual emotions presented as a history, and the information may be presented in a time-series graph form (FIG. 11 shows a graph). The right end of the line is shown as the current estimated information of team emotions and individual emotions, and the estimated information that goes back to the past is shown from the right end to the left).

Subsequently, an example of information processing according to the present embodiment will be described with reference to FIG. FIG. 12 is a flow chart showing an example of a schematic flow of information processing of the emotion estimation system according to the present embodiment. In step S101, the emotion estimation server 100 sets the parameters of the team to be estimated for team emotions and its constituents, and the type of team emotions to be estimated (for example, positiveness, etc.), which is set by the administrator of the emotion estimation system. Corresponds to parameters of negativeness, confusion, etc., calculation method of team emotions (for example, selection of average / median / mode as representative values, number of classifications of strength of social relations, strength of social relations) The parameters of the weighting coefficient value, etc.) are stored in the storage unit 120.

The above parameters may be set in advance by the administrator before the information processing, the setting parameters at the time of the previous information processing may be diverted, or the initial setting parameters may be used as they are. good. In that case, the execution of this step S101 may be omitted.

In step S103, in the emotion estimation server 100, the communication unit 110 acquires sensor data of the behavior and physiological reaction of the user 900 who is a member of the team for which the team emotion is estimated, via the sensor terminal 200. ..

In step S105, in the emotion estimation server 100, the individual emotion estimation unit 131 performs an individual emotion estimation process from the sensor data of the user 900. In the present embodiment, the process in which the individual emotion estimation unit 131 first estimates the individual emotion and then the team emotion estimation unit 135 estimates the team emotion as a collective data of them has been described as an example. However, the team emotion estimation unit 135 may estimate the team emotion directly from the sensor data of a plurality of people targeted by the team. In that case, the team emotion estimation unit 135 may perform a weighting calculation process corresponding to the above-mentioned strength of social relations on the sensor data value itself or its feature amount data. In that case, the execution of this step S105 may be omitted.

In step S107, in the emotion estimation server 100, the communication unit 110 responds to the “communication action by the sensor terminal 200” of the user 900 who is a member of the team to which the team emotion is estimated from the sensor terminal 200 or the PBX 300. Acquiring "sensor data" or "communication history data by PBX300", the social relationship information management unit 133 generates the above-mentioned social NW information based on the data acquired by the communication unit 110, and the generated society. Calculate social relationship information based on NW information.

In step S109, the team emotion estimation unit 135 estimates the team emotion of the team set in step S101 from the information indicating the individual emotion and the social relationship information. In step S111, in the information presentation terminal 400, the control unit 430 causes the presentation unit 450 to present the information indicating the team emotion acquired from the emotion estimation server 100 and the information indicating the individual emotion as the calculation element thereof. If the information presenting terminal 400 does not present the information indicating the estimated team emotion, the execution of this step S111 may be omitted.

As described above, according to the embodiment of the present invention, team emotions based on individual emotions are automatically made highly accurate by using information indicating dynamic and "informal" social relationships such as daily communication behaviors. Can be estimated.

Although the preferred embodiments of the present invention have been described in detail with reference to the accompanying drawings, the present invention is not limited to such examples. It is clear that a person having ordinary knowledge in the field of technology to which the present invention belongs can come up with various modifications or modifications within the scope of the technical ideas described in the claims. , These are also naturally understood to belong to the technical scope of the present invention.

For example, in the above, an example in which the control unit 430 of the information presentation terminal 400 determines the layout of various information presented by the presentation unit 450 has been mainly described. However, the layout of various information presented by the presentation unit 450 may be determined by another device (for example, the emotion estimation server 100). At this time, the emotion estimation server 100 may determine the layout of various information presented by the presentation unit 450 for each type of user (for example, the job title of the user) who visually recognizes the information presentation terminal 400.

100 Emotion estimation server 110 Communication unit 120 Storage unit 130 Control unit 131 Individual emotion estimation unit 133 Social relations information management department 135 Team emotion estimation unit 200 Sensor terminal 300 PBX
310 Communication unit 320 Storage unit 330 Control unit 400 Information presentation terminal 410 Communication unit 420 Storage unit 430 Control unit 440 Operation unit 450 Presentation unit

Claims (13)

Social relationship information management that generates information indicating social relationships based on the first sensor data corresponding to communication behaviors between a plurality of users or communication history data including information that can identify users related to the communication behaviors. Department and
Based on the information indicating the social relationship, a weighting coefficient corresponding to each combination of the plurality of target users selected by a part of the plurality of users is calculated, and the second sensor data of each of the plurality of target users is calculated. Alternatively, a team emotion estimation unit that estimates team emotions based on the information based on the second sensor data and the weighting coefficient, and
Emotion estimation server. The emotion estimation server
A personal emotion estimation unit that estimates the individual emotions of each target user based on the second sensor data of each of the plurality of target users is provided.
The team emotion estimation unit obtains information indicating the individual emotion as information based on the second sensor data.
The emotion estimation server according to claim 1. The team emotion estimation unit obtains feature amount data of the second sensor data of each of the plurality of target users as information based on the second sensor data.
The emotion estimation server according to claim 1. The team emotion estimation unit weights the second sensor data or information based on the second sensor data by the weighting coefficient, and the average value, median value, or mode of the weighted data. Calculate the value as the team feeling,
The emotion estimation server according to claim 1. The social relationship information management unit generates social network information based on the first sensor data or the communication history data, and generates information indicating the social relationship based on the social network information.
The emotion estimation server according to claim 1. The social network information includes a degree indicating the number of other users who have a relationship with each of the plurality of users.
The social relations information management unit generates information indicating the social relations based on the order.
The emotion estimation server according to claim 5. The social network information includes the strength of the relationship between the two of the plurality of users.
The social relations information management unit generates information indicating the social relations based on the strength.
The emotion estimation server according to claim 5. The social network information includes a cluster coefficient of the social network representing the relationship between the plurality of users.
The social relations information management unit generates information indicating the social relations based on the cluster coefficient.
The emotion estimation server according to claim 5. To generate information indicating social relations based on the first sensor data corresponding to the communication behavior between a plurality of users or the communication history data including the information that can identify the user regarding the communication behavior.
Based on the information indicating the social relationship, a weighting coefficient corresponding to each combination of the plurality of target users selected by a part of the plurality of users is calculated, and the second sensor data of each of the plurality of target users is calculated. Alternatively, estimating team sentiment based on the information based on the second sensor data and the weighting coefficient, and
Emotion estimation methods, including. It was calculated based on the first sensor data corresponding to the communication behavior between a plurality of users or the information indicating the social relationship generated based on the communication history data including the information that can identify the user regarding the communication behavior. Based on the weighting coefficient corresponding to each combination of the plurality of target users selected by a part of the plurality of users, and the second sensor data of each of the plurality of target users or the information based on the second sensor data. A presentation unit that presents information that indicates the estimated team sentiment,
A presentation device. The presentation unit presents information indicating the team emotions of each of the plurality of teams in an area closer to each other as the connection between the teams is stronger.
The presentation device according to claim 10. The presentation unit presents information indicating the team emotion and individual emotions of each of the plurality of target users for each emotion category.
The presentation device according to claim 10. Social relationship information management that generates information indicating social relationships based on the first sensor data corresponding to communication behaviors between a plurality of users or communication history data including information that can identify users related to the communication behaviors. Department and
Based on the information indicating the social relationship, a weighting coefficient corresponding to each combination of the plurality of target users selected by a part of the plurality of users is calculated, and the second sensor data of each of the plurality of target users is calculated. Alternatively, a team emotion estimation unit that estimates team emotions based on the information based on the second sensor data and the weighting coefficient, and
Emotion estimation server and
A presentation unit that presents information indicating the team emotions,
With a presentation device
Emotion estimation system.

Images