Emotion Presumption Device, Method, and Program

Abstract

A device for estimating emotion according to the embodiment includes an estimation unit that extracts a degree of prediction and a degree of expectation which correspond to a scene that a user is currently relating to, the degree of prediction indicating a magnitude of a probability of a change from a first situation to a second situation, the degree of expectation indicating a magnitude of user expectation to the change, extracts at least one of an amount of knowledge and an amount of experience of the user corresponding to the scene that the user is currently relating to, and estimates, on the basis of a result of the extraction, an amount of emotion creation indicating the intensity of emotional expression to be created when the change occurs.

Description

Technical Field

[0001] Embodiments of the present invention relate to an emotion estimation device, method, and program.

Background Art

[0002] Non-Patent Document 1 discloses a method for investigating, collecting, and systematically classifying situations in which a user's emotion occurs. Further, Patent Document 1 discloses a technique for modeling and estimating emotions from a lifelog in which a user's life record is managed. Similarly, Non-Patent Document 2 discloses a technique for supporting communication using the modeled emotions.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Non-Patent Documents

[0004]

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the method disclosed in Non-Patent Document 1 above, for example, momentary emotions associated with an event such as "being nervous during an interview" can be handled, but it is not possible to handle differences in the degree of emotion due to individual users' experiences, such as "being nervous during the first interview".

[0006] Also, in the techniques disclosed in Patent Document 1 or Non-Patent Document 2, for example, the characteristics of the event itself are associated with emotions such as "music live = fun", and it is difficult to estimate emotions for first-time events. Further, in these techniques, differences in the degree of emotion caused by the user's knowledge, such as "the fun increases more for a live performance of an artist well-known since childhood", are not considered.

[0007] This invention has been made in view of the above circumstances, and an object thereof is to provide an emotion estimation device, method, and program capable of appropriately estimating a user's emotion.

Means for Solving the Problems

[0008] An emotion estimation device according to an aspect of the present invention includes a first situation, a second situation changed from the first situation, emotion information representing a user's emotional expression when changing from the first situation to the second situation, a prediction degree indicating the magnitude of the possibility of changing from the first situation to the second situation, and an expectation degree indicating the magnitude of the user's expectation for changing from the first situation to the second situation, which are managed together with information indicating a scene in which the user is involved as first management information; second management information in which at least one of a knowledge amount indicating the amount of knowledge the user has regarding the scene and an experience amount indicating the amount of experience the user has regarding the scene is managed; and a calculation definition of an emotion creation amount indicating the strength of an emotional expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information, which is managed together with information indicating the scene as third management information, and is a device that estimates an emotion generated in the user. The device extracts the prediction degree and the expectation degree corresponding to the scene in which the user is currently involved from the first management information, extracts the second management information corresponding to the scene in which the user is currently involved, and based on the extracted results and the calculation definition of the emotion creation amount corresponding to the scene in which the user is currently involved, estimates an emotion creation amount indicating the strength of an emotional expression created by the user when changing from the first situation to the second situation, and includes an estimation unit.

[0009] An emotion estimation device according to one aspect of the present invention manages, together with information indicating a scene in which the user is involved, first management information including a first situation, a second situation changed from the first situation, emotion information representing the user's emotional expression when changing from the first situation to the second situation, a prediction degree indicating the likelihood of changing from the first situation to the second situation, and an expectation degree indicating the magnitude of the user's expectation for changing from the first situation to the second situation; second management information in which at least one of a knowledge amount indicating the amount of knowledge the user has about the scene and an experience amount indicating the amount of experience the user has about the scene is managed; and third management information in which a calculation definition of an emotion creation amount indicating the intensity of an emotional expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information is managed. The device estimates the emotion that occurs to the user using the third management information. The device includes an estimation unit that extracts the prediction degree and the expectation degree corresponding to the emotion information representing the emotional expression that the user wants to share with a partner from the first management information, extracts the information indicating the scene managed together with the extracted result in the first management information, extracts management information in which at least one of the knowledge amount and the experience amount among the second management information corresponding to the extracted scene satisfies a condition, and estimates an emotion creation amount indicating the intensity of the emotional expression created by the user based on the extracted prediction degree, the expectation degree, the extracted second management information, and the calculation definition of the emotion creation amount.

[0010] The emotion estimation method according to one aspect of the present invention includes a first situation, a second situation changed from the first situation, emotion information representing the user's emotion expression when changing from the first situation to the second situation, a prediction degree indicating the magnitude of the possibility of changing from the first situation to the second situation, and an expectation degree indicating the magnitude of the user's expectation for changing from the first situation to the second situation, which are managed together with information indicating the scene with which the user is involved as first management information, second management information in which at least one of the amount of knowledge indicating the amount of knowledge the user has about the scene and the amount of experience indicating the amount of experience the user has about the scene is managed, and a calculation definition of an emotion creation amount indicating the intensity of the emotion expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information, which is performed by an emotion estimation device that estimates the emotion generated in the user using third management information managed together with information indicating the scene, wherein the estimation unit of the emotion estimation device extracts the prediction degree and the expectation degree corresponding to the scene with which the user is currently involved from the first management information, extracts the second management information corresponding to the scene with which the user is currently involved, and based on the extracted results and the calculation definition of the emotion creation amount corresponding to the scene with which the user is currently involved, estimates the emotion creation amount indicating the intensity of the emotion expression created by the user when changing from the first situation to the second situation.

[0011] The emotion estimation method according to one aspect of the present invention includes a first situation, a second situation changed from the first situation, emotion information representing the user's emotion expression when changing from the first situation to the second situation, a prediction degree indicating the magnitude of the possibility of changing from the first situation to the second situation, and an expectation degree indicating the magnitude of the user's expectation for changing from the first situation to the second situation, which are managed together with information indicating the scene with which the user is involved as first management information, at least one of the amount of knowledge indicating the amount of knowledge the user has about the scene and the amount of experience indicating the amount of experience the user has about the scene is managed as second management information, and a calculation definition of the amount of emotion creation indicating the intensity of the emotion expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information is managed together with the information indicating the scene as third management information. A method performed by an emotion estimation device for estimating the emotion generated in the user, wherein the estimation unit of the emotion estimation device extracts the prediction degree and the expectation degree corresponding to the emotion information representing the emotion expression the user wants to share with the partner from the first management information, extracts the information indicating the scene managed together with the extracted result in the first management information, extracts the management information in which at least one of the amount of knowledge and the amount of experience in the second management information corresponding to the extracted scene satisfies the condition, and estimates the amount of emotion creation indicating the intensity of the emotion expression created by the user based on the extracted prediction degree, expectation degree, the extracted second management information, and the calculation definition of the amount of emotion creation.

Effect of the Invention

[0012] According to the present invention, the emotion of the user can be appropriately estimated.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

[0014] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In one embodiment of the present invention, emotion information representing the emotional expression of the user (hereinafter sometimes simply referred to as emotion) generated by a change in the situation related to the user and the degree of the intensity of this emotion are modeled according to the user's knowledge and experience. Thereby, based on the situation in which the user is placed, it is possible to estimate the emotion and the degree of this intensity (degree of emotion), or to promote empathy with others by reproducing and presenting a similar situation that is easy for the user to understand. In this embodiment, for example, all emotions caused by changes in situations, such as "wasteful", "frustrating", and "thrilling", are the targets. For example, it is assumed to be used for performances tailored to the emotions of the audience at various events such as music live (sometimes simply referred to as live), or for the mutual understanding of emotions in remote or cross-cultural communications.

[0015] In this embodiment, in addition to the situations before and after the user is placed, by considering the state before the occurrence of each person's emotion, such as the amount of knowledge, experience, and degree of expectation described later, it is possible to estimate different emotions or the degree of intensity of emotions for each individual at the moment the emotion occurs. Thereby, it is expected that by customizing the service in real time or presenting similar situations that are easy for each person to understand to the target person, the target person can understand and empathize with the feelings of others.

[0016] For example, when used in telemedicine, by grasping the "tension level" of the patient when interacting with the doctor at all times, it can be used to improve the psychological safety of the patient.

[0017] Also, for example, when used during customer service using a tablet or the like at a travel agency, it is possible to present and have the customer understand cases such as the satisfaction of others at the travel destination.

[0018] (First Embodiment) Next, the first embodiment of the present invention will be described. FIG. 1 is a diagram showing an application example of an emotion estimation device according to the first embodiment of the present invention. As shown in FIG. 1, the emotion estimation device 100 according to the first embodiment of the present invention includes a scene / attribute input unit 10, an emotion calculation unit (emotion estimation unit) 20, a storage unit 30, an output unit 40, and a DB (database) input unit 50. In the memory unit 30, as various databases, a situation change list DB 31, a life log DB 32, and an emotion creation amount calculation definition DB 33 are provided. Details of each part shown in FIG. 1 will be described later.

[0019] In this embodiment, it is assumed that the information in the situation change list DB 31, the life log DB 32, and the emotion creation amount calculation definition DB 33 is created in advance and various databases are constructed by an operation related to the DB input unit 50 by a service provider using the emotion estimation device 100.

[0020] FIG. 2 is a diagram showing an example of the functional configuration of the DB input unit. The DB input unit 50 receives an input operation of information related to the situation change list DB 31, the life log DB 32, and the emotion creation amount calculation definition DB 33 from a service provider for an input / output UI (User Interface) screen. As shown in FIG. 2, the DB input unit 50 includes a scene creation / input unit 51, a situation / emotion data input (acquisition) unit 52, a prediction degree input (acquisition) unit 53, an expectation degree definition input unit 54, a knowledge / experience data acquisition unit 55, a calculation definition input unit 56, and a parameter specification unit 57. Details of each part in the DB input unit 50 will be described later.

[0021] The scene / attribute input unit 10 shown in FIG. 1 can input information indicating a scene in which the user is located, for example, "content viewing" as a major classification of the scene, "music live" as a medium classification of the scene, and "○×stars" (music artist name) as a minor classification of the scene, by input or selection on the screen by the user himself / herself. The information of the above scene may be obtained by sensing or the like. Hereinafter, the input or acquired content will be referred to as a scene.

[0022] The input method by the scene / attribute input unit 10 may be manual input of the scene on the UI screen or voice input by means of the voice recognition function of a smart speaker or the like. Also, the user may select a desired scene from the pull-down menu displayed on the UI screen, or select a scene by voice input by the user from among categories of various enumerated scenes.

[0023] In addition, the scene / attribute input unit 10 can extract, as the input target scene, a scene that matches the input content by manual input of characters or voice on the UI screen or a scene that is most similar to the input content. The scene / attribute input unit 10 can extract the similar scenes by, for example, preparing a pre-association setting file in advance and aggregating them into a category such as "Music Live" for an input such as "Festival", "○×Rock" (event name), or "Outdoor Live".

[0024] When scene information is obtained by the above sensing, if the major classification of the scene is content browsing, the scene information can be automatically obtained from, for example, the browsing information of the user on a PC (Personal Computer) or a TV (television). Also, when the major classification of the scene includes location information, this location information can be obtained from a mobile terminal.

[0025] In addition, the scene / attribute input unit 10 can automatically obtain the user's attributes, such as a personal ID (Identifier), by means of manual input or selection by the user on the UI screen or on a mobile terminal.

[0026] Figure 3 is a diagram showing an example of an input screen related to the parameter specifying unit. The DB input unit 50 shown in FIG. 2 causes the UI screen to display an input screen G1 for situation / feeling data, prediction degree data, calculation definitions for expectation degrees (hereinafter sometimes referred to as expectation degree calculation definitions), and calculation definitions for amounts of emotion creation (hereinafter sometimes referred to as expectation degree calculation definitions), which are parameters corresponding to the input scene. FIG. 3 shows the input screen G1 when the "major classification / medium classification / minor classification" of the input scene is "content viewing / music live / ○×stars".

[0027] The situation / feeling data includes "Situation A", which is the situation before the change, "Situation B", which is the situation after the change from the "Situation A", and the user's emotion. This emotion is the user's emotional expression when changing from "Situation A" to "Situation B". Through an input operation by the service provider on the UI screen or the like, the parameter specifying unit 57 can specify parameters corresponding to the input scene according to the input operation by the service provider.

[0028] The scene creation / input unit 51 of the DB input unit 50 receives an input operation on the UI screen by the service provider, and creates a list of scenes that can be selected by the user when using the emotion estimation device 100 according to this input operation. The list of scenes includes the major classification of the scene, the medium classification of the scene, and the minor classification of the scene. The major classification of the scene is, for example, "content viewing" or the like, and may include location information. The medium classification of the scene is, for example, "music live", "news / weather", or "means of transportation" or the like. The minor classification of the scene is, for example, "artist name", "news category" or the like, and this minor classification may not be set. The classification of the scene is not limited to the above classification and may be any structure of classification, and this list of scenes can be created by citation of documents or the like or manual input by the service provider on the UI screen.

[0029] Next, the input of information in each DB, that is, the situation change list DB 31, the life log DB 32, and the emotion creation amount calculation definition DB 33, by each unit in the DB input unit 50 shown in FIG. 2 will be described.

[0030] In a state where the input screen G1 as shown in FIG. 3 is displayed according to the list of scenes created by the scene creation / input unit 51, the situation / feeling data input (acquisition) unit 52 inputs situation / feeling data (which may be referred to as situation and feeling data) by an operation on the parameter specifying unit 57, and writes it as situation / feeling data related to the scene displayed on the input screen G1 into the situation change list DB31. This situation / feeling data may be manually input for the input screen G1, or may be automatically acquired by crawling web information, etc.

[0031] In the present embodiment, as shown in FIG. 3, the situation / feeling data consists of "Situation A", "Situation B", and "Feeling". "Situation A" is the situation in which the user is placed before changing to "Situation B" under the conditions of a certain scene. "Situation B" is the situation in which the user is placed after changing from "Situation A" under the conditions of the above scene. "Feeling" is one or more types of feelings that occurred to the user in the change from the above "Situation A" to "Situation B" under the conditions of the above scene. This situation / feeling data may be input after being extracted and aggregated by a general-purpose sentiment analysis tool such as ML-Ask (see, for example, https: / / gaaaon.jp / blog / mlask) from the text displayed in a social network service, or as disclosed in the above Non-Patent Document 1, cases where each feeling occurs may be collected from a questionnaire and input.

[0032] In a state where the input screen as shown in FIG. 3 is displayed, the prediction degree input (acquisition) unit 53 accepts an input operation related to the prediction degree by the service provider for the UI screen, and writes this prediction degree as the prediction degree related to the scene displayed on the input screen into the situation change list DB31. The prediction degree indicates the likelihood of the situations and emotions in the situation-emotion data written in the situation change list DB31, and is, for example, a numerical value between 0 and 1.

[0033] The prediction degree data may be input manually by the service provider on the UI screen, or may be automatically acquired by, for example, an AI (Artificial Intelligence) chatbot. In the case of manual input, from information on the web, etc., for example, the probability that a new song will be suddenly revealed at a music live, or the probability that ballad songs will be continuously played at a music live, etc. are investigated in advance, and a prediction degree corresponding to the investigation result may be input.

[0034] In a state where an input screen as shown in FIG. 3 is displayed, the expectation degree definition input unit 54 receives an input operation by the service provider for the UI screen regarding the expectation degree calculation definition (sometimes referred to as the expectation degree definition), and writes this expectation degree calculation definition to the situation change list DB31 as the expectation degree calculation definition related to the scene displayed on the input screen.

[0035] The expectation degree indicates the strength of the user's feeling that the situations and emotions in the situation-emotion data written in the situation change list DB31 should occur, and is, for example, a numerical value between 0 and 1. For the user, when it doesn't matter whether the above situations and emotions occur or not, the expectation degree is "0", and when the user wants the above situations and emotions to occur, the expectation degree is "1". Also, when the above feeling includes the strength of the feeling that the user doesn't want the above situations and emotions to occur, the expectation degree is, for example, a numerical value between "-1" and "1". Specifically, for the user, when the user doesn't want the above situations and emotions to occur no matter what, the expectation degree is "-1", when it doesn't matter either way, the expectation degree is "0", and when the user wants the above situations and emotions to occur no matter what, the expectation degree is "1".

[0036] The definition of expected value calculation may be set by manual input by the service provider on the UI screen, or may be set by selection on the screen by the service provider. In this embodiment, the user's degree of expectation indicating the strength of the desire for a change from "Situation A" to "Situation B" under the conditions of a certain scene is used, and an expected value calculation formula with the knowledge amount T_U and the experience amount K_U as variables is set as the definition of expected value calculation. Also, for example, selection items such as "The greater the knowledge and experience, the stronger the expectation" and "The greater the knowledge and experience, the weaker the expectation" may be displayed on the screen, and the calculation formula corresponding to the item selected by the service provider may be set as the definition of expected value calculation.

[0037] The knowledge / experience data acquisition unit 55 acquires the knowledge amount and experience amount for each scene from sensing data such as the user's search history and purchase history, for example, and accumulates the results of this acquisition in the life log DB 32. These knowledge amount and experience amount are values obtained by parameterizing the magnitude of the user's knowledge and experience with respect to the situation and emotion in the situation / emotion data written in the situation change list DB 31, and are, for example, numerical values between 0 and 1.

[0038] The calculation definition input unit 56 receives the input of a calculation formula for calculating the amount of emotion creation for the UI screen. This amount of emotion creation indicates the strength of the emotion created for the user with this change as a trigger when the situation changes from the above "Situation A" to "Situation B" under the conditions of a certain scene, and is, for example, a numerical value between 0 and 1. This definition of the amount of emotion creation calculation may be set by manual input by the service provider for the UI screen, or may be set by selection on the screen by the service provider.

[0039] Specifically, the definition of the amount of emotion creation calculation can be set as a calculation formula for the amount of emotion creation with the strength of the user's emotion in the change from the above "Situation A" to "Situation B" under the conditions of a certain scene, using some or all of the prediction degree Y, the expected value E_U, the knowledge amount T_U, the experience amount K_U, and the external factor G as variables. Also, for example, selection items such as "the lower the prediction degree, the higher the expectation degree, and the more knowledge and experience, the stronger the emotion" and "the higher the prediction degree, the lower the expectation degree, the stronger the emotion" may be displayed on the screen, and a calculation formula corresponding to the item selected by the service provider may be set as the emotion creation amount calculation definition.

[0040] FIG. 4 is a diagram showing an example of the content of the situation change list DB according to the first embodiment of the present invention in tabular form. In the situation change list DB 31 shown in FIG. 1 and the like, data is individually created for each combination of classifications of scenes. As shown in FIG. 4, under the conditions of a certain scene, (1) "Situation A", (2) "Situation B", (3) "Emotion" that occurs to the user when the situation changes from "Situation A" to "Situation B", (4) "Prediction degree Y" which is the probability that the situation changes from "Situation A" to "Situation B", (5) "Expectation degree E_U calculation definition", and (6) "External factor G" that affects the intensity of emotion. Situation change list information is created and accumulated for each combination of classifications (directly the minimum classification of the scene) from the major classification to the minimum classification of all scenes. For example, when no sub-classification is set, the situation change list information is created and accumulated for each combination of the major classification and the middle classification (directly the middle classification).

[0041] In FIG. 4, the situation change list information when the "major classification / middle classification / minor classification" of the input scene is "content viewing / music live / ○×stars" (see FIG. 3) is shown. Also, each row of the situation change list information shown in FIG. 4 may be referred to as a record in the situation change list information. As shown in FIG. 4, an individual correspondence number (also referred to as a record number) is assigned to each row of the situation change list information.

[0042] The prediction degree Y is the likelihood or probability that the situation changes from "Situation A" to "Situation B" under the conditions of a certain scene, and can be expressed as a value in the range of 0 to 1. In the situation change list information, if, for example, the emotion of "wasteful" in the scene of "news" is set in the situation change list DB31, then, for example, the prediction degree Y (for example, 0.27) as the frequency of unsold food being discarded can be set in the record where the emotion is set. Also, in the situation change list information, if the emotion of "wasteful" in the scene of "meeting", for example, is set in the situation change list DB31, then, for example, the prediction degree Y (for example, 0.3) as the probability that a participant with knowledge related to the speech misses the opportunity to speak can be set in the record where the emotion is set.

[0043] External factor G is defined as other factors that affect the fluctuation of the emotion creation amount. The external factor G is, for example, (1) acquisition of advance prediction information that the situation changes from "Situation A" to "Situation B", (2) BGM (background music) when the situation occurs, or (3) presence or absence of speech by fellow participants, etc., which are factors that strengthen or weaken the user's emotion. In the situation change list DB31, the presence or absence of events targeted by the external factor G and parameters are accumulated.

[0044] If the external factor G is a factor that strengthens the emotion, the parameter of this external factor G can be set to "+0.1". If the external factor G is a factor that weakens the emotion, the parameter of this external factor G can be set to "-0.1".

[0045] FIG. 5 is a diagram showing an example of the content of the life log DB in tabular form. In the life log DB32 shown in FIG. 1 and the like, life logs created for each individual user or user attribute are accumulated. This life log is associated with (1) the major classification of the scene, (2) the middle classification of the scene, (3) the minor classification of the scene, (4) the user's knowledge amount T_U for the combination of each classification of the scene (directly the minimum classification of the scene), and (5) the user's experience amount K_U for the combination of scenes of each classification. Also, each row of the life log shown in FIG. 5 may be referred to as a record in the life log.

[0046] In the example of FIG. 5, there is a life log created for a certain user, including (1) a record of the life log when the "major classification / medium classification / minor classification" of the input scene is "content viewing / music live / ○× stars", (2) a record of the life log when the "major classification / medium classification / minor classification" of the input scene is "content viewing / music live / Yamada ○× (artist name)", and (3) a record of the life log when the "major classification / medium classification / minor classification" of the input scene is "conference / academic society / international conference".

[0047] The amount of knowledge T_U in the life log is the amount of knowledge that user U has about the corresponding scene. For example, regarding the amount of knowledge about the music artist "○× stars" shown in FIG. 5, for example, taking the combined value of (1) the number of purchase histories of songs etc. related to the above artist in the user's purchase history and (2) the number of times the name of the above artist is entered and the presence or absence of bookmark registration in the user's search history as T. And, the maximum value of the above combined value among all other users, that is, the combined value of the user with the most combined value among all users is T max The ratio T / T when max is taken as the amount of knowledge T_U of user U, and the amount of knowledge T_U can be calculated by methods such as this.

[0048] The amount of experience K_U in the life log is the amount of times user U has actually experienced the corresponding scene. For example, regarding the amount of experience that the user has about the music artist "○× stars", for example, taking the combined value of (1) the number of purchase histories of live tickets of the above artist in the user's purchase history and (2) the number of times of participating in the live of the above artist in the user's schedule data as K.

[0049] And, the maximum value of the above combined value among all other users, that is, the combined value of the user with the most combined value among all users is K max The ratio K / T when maxThe experience amount \(K_U\) of user \(U\) can be calculated by methods such as using the experience amount \(K_U\).

[0050] Also, using the knowledge amount \(T_U\) and experience amount \(K_U\) of the above life log, the expectation degree \(E_U\) is calculated based on the expectation degree calculation definition in the situation change list DB31. This expectation degree \(E_U\) is the strength of the expectation that user \(U\) desires a change from "Situation A" to "Situation B".

[0051] For example, the expectation degree for the first public performance of a new song by the artist "○× Stars" (refer to the row with the corresponding number "1" in Figure 4) is assumed to be higher as the user knows the artist well and has more live experience. Therefore, the expectation degree \(E_U\) of user \(U\) can be calculated by the following formula (1). \(E_U = T_U + K_U\) … Formula (1)

[0052] Conversely, the expectation degree for the band dissolution announcement by the artist "○× Stars" (refer to the row with the corresponding number "2" in Figure 4) is assumed to be lower as the knowledge amount and experience amount of the users related to the artist are higher. Therefore, the expectation degree \(E_U\) of user \(U\) can be calculated by the following formula (2). \(E_U = 1 / (T_U + K_U)\) … Formula (2)

[0053] In this way, the expectation degree can be calculated as an increase or decrease in emotion using the knowledge amount and experience amount for each situation. Also, in the life log, when there is no influence on emotion due to the knowledge amount and experience amount in a certain situation change (for example, refer to the row with the corresponding number "3" in Figure 4), the related expectation degree \(E_U\) is set to 0.

[0054] Figure 6 is a diagram showing an example of the content of the emotion creation amount calculation definition DB in tabular form. In the emotion creation amount calculation definition DB33 shown in Figure 1 and the like, emotion creation amount calculation definition information created individually for each combination of classifications of scenes is accumulated, with the emotion creation amount calculation definition and the maximum value of the emotion creation amount associated.

[0055] In the example of FIG. 6, multiple types of emotional creation amount calculation definition information are shown, including (1) emotional creation amount calculation definition information (reference sign a in FIG. 6) when the "major classification / medium classification / minor classification" of the input scene is "content viewing / music live / ○× stars" (see FIG. 5), (2) emotional creation amount calculation definition information (reference sign b in FIG. 6) when the "major classification / medium classification / minor classification" of the input scene is "content viewing / music live / Yamada ○×" (see FIG. 5), and (3) emotional creation amount calculation definition information (reference sign c in FIG. 6) when the "major classification / medium classification / minor classification" of the input scene is "meeting / academic society / international conference" (see FIG. 5).

[0056] Also, each row of the emotional creation amount calculation definition information shown in FIG. 6 may be referred to as a record in the emotional creation amount calculation definition information. Also, a corresponding number is assigned to each row of the emotional creation amount calculation definition information. The corresponding numbers "D1-1", "D1-2", "D1-3"... assigned to each row of the emotional creation amount calculation definition information shown in FIG. 6 correspond to the corresponding numbers "1", "2", "3"... assigned to each row of the situation change list information shown in FIG. 4.

[0057] The emotional creation amount calculation definition is a calculation definition of a predicted value indicating to what extent the user's emotion is created in the corresponding scene based on part or all of the prediction degree Y, expectation degree E_U, knowledge amount T_U, experience amount K_U, and external factor G. In the emotional creation amount calculation definition DB33, the maximum value of the values that each definition can take to express the degree of the calculated value of the emotional creation amount is accumulated.

[0058] For example, assuming that the amount of excitement created in the world premiere of a new song by a music artist called "○× Stars" shown in Figure 6 etc. (refer to the row with the corresponding number "1" in Figure 4) is higher the more unexpected the new song premiere is (for example, the prediction degree Y in the list of situation changes shown in Figure 4 is small), higher the greater the user's expectation (for example, the expectation degree E_U in the list of situation changes shown in Figure 4), and higher the better the user knows the artist (for example, the knowledge amount T_U and experience amount K_U in the life log shown in Figure 5 are large), the amount of excitement created can be calculated by, for example, "((E_U + 1)·(T_U + K_U + 1)) / (Y + 1)".

[0059] Also, considering the external factor G, the calculation formula for the amount of excitement created can be defined as "((E_U + 1)·(T_U + K_U + 1)) / (Y + 1)+G" (refer to the row with the corresponding number "D1-1" shown in Figure 6). Also, the "+1" in "E_U + 1" in the said calculation formula is a value for preventing the denominator or numerator from becoming 0. Also, when the expectation degree E_U is less than 0, "E_U + 1" in the above calculation formula can be changed to "E_U".

[0060] Specifically, when calculating the user's excitement in the world premiere of the new song by the above "○× Stars" as the amount of excitement created, from the examples shown in Figures 4 to 6, with "prediction degree Y = 0.3", "expectation degree E_U = 1.4", "knowledge amount T_U = 0.9", "experience amount K_U = 0.5", and "external factor G = +0.1", the amount of excitement created is calculated by the following formula (3). Amount of excitement created=(2.4·2.4) / 1.3+0.1≒4.5 …Formula (3)

[0061] After constructing the excitement creation amount calculation definition DB33, the ratio of the calculated result of the excitement creation amount to the maximum value of this excitement creation amount can be calculated as shown in the following formula (4). Ratio=4.5 / 6.1≒0.74=74[%] …Formula (4)

[0062] The external factor G in the above various databases is not essential, and this external requirement G may be removed from the various databases. In this case, the emotional creation amount calculation definition in the emotional creation amount calculation definition database 33 may be constituted by part or all of the prediction degree Y, the expectation degree E_U, the knowledge amount T_U, and the experience amount K_U.

[0063] Also, in the above situation change list database 31 and the emotional creation amount calculation definition database 33, only one of the knowledge amount T_U and the experience amount K_U may be managed. For example, when only the knowledge amount T_U among the knowledge amount T_U and the experience amount K_U is managed in the life log database 32, the expectation degree E_U calculation definition in the situation change list database 31 may be constituted only by the knowledge amount T_U. Also, when only the knowledge amount T_U among the knowledge amount T_U and the experience amount K_U is managed in this way, the emotional creation amount calculation definition in the emotional creation amount calculation definition database 33 may be constituted by part or all of at least the prediction degree Y, the expectation degree E_U, the knowledge amount T_U, and the external factor G among the prediction degree Y, the expectation degree E_U, the knowledge amount T_U, and the external factor G.

[0064] FIG. 7 is a diagram showing an example of a procedure of a processing operation by an emotion estimation device according to a first embodiment of the present invention. After the construction of the various databases, the emotion calculation unit 20 performs the following processes S11 to S13 on each classification of the scene and the user attributes input by the user.

[0065] · Process S11 The emotion calculation unit 20 detects a change from "Situation A" (for example, "Known song") to "Situation B" (for example, "Premiere of a new song"), and at the same time, detects the presence or absence of the external factor G at that time (for example, "Detection present = +0.1"), and among the situation change list information related to each scene accumulated in the situation change list database 31, identifies the situation change list information related to the combination of each classification of the input scene. The emotion calculation unit 20 identifies a record (reference symbol a in FIG. 7) including the combination of the detected "Situation A", "Situation B", and external factor G in this identified situation change list information. The emotion calculation unit 20 extracts, from the identified record, the emotion "excitement" created by the change in the situation, the prediction degree "0.3", and the definition for calculating the expectation degree E_U, which are associated with "Situation A", "Situation B", and the external factor G. Simultaneously with this extraction, the corresponding number of the identified record in the situation change list DB31 (see, for example, the row of the corresponding number "1" of the record indicated by the reference sign a in FIG. 7) is determined.

[0066] The change from the above "Situation A" to "Situation B" can be detected, for example, by "Extraction of Actors, Actions and Events from Sports Video by Integrating Linguistic and Image Information, Nitta et al., Technical report of IEICE, PRMU, Pattern Recognition and Media Understanding, Vol. 99, No. 709, pp. 75 - 82, 2000." or "A study on selecting scenes that match the news contents for news video summarization, ZHANG et al., IEICE Technical Report, vol. 115, no. 76, MVE2015 - 7, pp. 57 - 58, 2015." when the major classification of the input scene is "content viewing". When the location information is included in the major classification of the scene, the stay location is extracted from the mobile terminal as a situation.

[0067] The presence or absence of the external factor G is detected as the presence or absence of the situation corresponding to the external factor in the situation change list DB31 as a factor affecting the variation in the intensity of the emotion caused by the change from the above - detected "Situation A" to "Situation B". For example, the presence or absence of the external factor G is detected based on the degree of coincidence between the information transmitted before the occurrence of "Situation B", which is obtained by voice recognition or image recognition of the artist's speech or production in the major classification "Content Browsing" of the input scene, and "Situation B". Alternatively, the presence or absence of the external factor G is detected from the user's search history and content browsing history related to the artist before the "Content Browsing".

[0068] At that time, the parameters of the external factor may be increased or decreased according to the degree of coincidence with the situation or the degree of the history. As an example of a specific effect, for example, when it is detected that the user has obtained information with a surprise at a live performance immediately before, the value of the external factor G is amplified, and the fact that the user's excitement level increases knowing the surprise is reflected in the estimated value of the emotional creation amount.

[0069] · Process S12 The emotion calculation unit 20 identifies the life log related to the input user attribute among the life logs stored in the life log DB 32, identifies the record related to the combination of each classification of the input scene by the user in this life log, and extracts the knowledge amount and experience amount of the user in this identified record. Then, the emotion calculation unit 20 calculates the expected degree E_U by substituting the results of these extractions into the expected degree E_U calculation definition in the record identified in the process S11 in the information change list DB 31 under the conditions of the input scene.

[0070] · Process S13 The emotion calculation unit 20 identifies the situation change list information related to the combination of each classification of the input scene among the emotion creation amount calculation definition information of each scene stored in the emotion creation amount calculation definition DB 33, and identifies the record (see reference symbol b in FIG. 7, for example) of the corresponding number (for example, the corresponding number "D1-1" in FIG. 7) corresponding to the corresponding number (for example, the corresponding number "1" in FIG. 7) determined in the process S11 in this identified situation change list information. The emotion calculation unit 20 substitutes the prediction degree Y extracted in process S11, the events / parameters of the external factor G, the knowledge amount T_U and experience amount K_U extracted in process S12, and the expectation degree E_U calculated in process S12, in the situation detected in process S11, into the emotion creation amount calculation definition included in the record (refer to symbol b in FIG. 7, for example, correspondence number "D1-1" in FIG. 7) corresponding to the correspondence number (for example, correspondence number "1" in FIG. 7) determined in process S11 and specified above in the emotion creation amount calculation definition DB33, to calculate the emotion creation amount (for example, "0.74 (74%)" shown in FIG. 7), which is the intensity of the emotion created by the user. The emotion calculation unit 20 outputs this calculation result as a message to the user (for example, "Excited with 74% intensity" shown in FIG. 7).

[0071] In the output of the emotion creation amount, the created emotion and its magnitude are included. When the output emotions are not of one type but multiple types, that is, multiple types of emotions can be output by the combination of "Situation A" and "Situation B". For example, each of the multiple types of emotions such as "Excited with 74% intensity" corresponding to "Situation A" and "Nervous with 55% intensity" corresponding to "Situation B" is output in a form where the created emotion and its magnitude are associated. Note that the output method of the above emotion creation amount may be in text form like the above "Excited with 74% intensity", or in vector form like "(Excited, 74)" for example, as long as the magnitudes for multiple types of emotions are noted in the corresponding form. Also, when multiple types of emotions are output, only the emotion with the largest value may be output.

[0072] In one embodiment of the present invention, although an example is described where the magnitude of emotion is calculated as a percentage (%), it may also be expressed as a negative value. As an example of the output of emotion, for example, it may be output in the form of text and sentences such as "Excitement 90%", emojis, illustrations, various symbols, or corresponding codes or combinations thereof. The method of expressing the output emotion may be expressed according to the characteristics of the media.

[0073] For example, in this embodiment, after identifying the created emotion and its magnitude, it is converted into a content format such as video, audio, or image, and presented on the screen of a smart phone, a television screen, a personal computer screen, or a game device screen, or output as audio. By switching the type of audio or the display method of the video according to the type of emotion, the expression of the content can also be changed so that an arbitrary emotional expression effect can be given.

[0074] Also, in this embodiment, by converting the intensity of the emotion into the type, size, color density, gaudiness of the decoration, tone color of the voice read aloud, volume, size of the symbol bar, or type and expressing it, it becomes possible to convey to the user the type of the created emotion and its magnitude in a form that is easy for the user to identify.

[0075] (Second Embodiment) Next, the second embodiment will be described. A detailed description of the same configuration as in the first embodiment in this embodiment will be omitted. In the second embodiment, the following points are different compared to the first embodiment. In the above first embodiment, after the construction of various DBs, the "scene" and the "attributes" of the user are input by the user, but in the second embodiment, after the construction of various DBs, the "attributes" of the user and the "emotion to be shared with the partner" by the user are input by the user.

[0076] In the first embodiment and the second embodiment, the range of processing targets in the situation change list DB 31 and the emotion creation amount calculation definition DB 33 is different. For example, in the first embodiment, only the information related to the scene input by the user in the situation change list DB 31 and the emotion creation amount calculation definition DB 33 is the target of processing, while in the second embodiment, all the situation change list information related to all scenes stored in the situation change list DB 31 and all the emotion creation amount calculation definition information related to all scenes stored in the emotion creation amount calculation definition DB 33 are the targets of processing.

[0077] FIG. 8 is a diagram showing an example of the content of the situation change list DB according to the second embodiment of the present invention in tabular form. FIG. 9 is a diagram showing an example of the procedure of the processing operation by the emotion estimation device according to the second embodiment of the present invention. In the second embodiment, for example, for the purpose of "wanting to convey a distance to a person from a different culture who is not familiar with the emotion of 'wasteful'", the scene / attribute input unit 10 accepts the input of information indicating (1) the attributes of the user and (2) the type of emotion that the user wants to know or the emotion in the specified situation through the input operation of the user on the UI screen. The emotion calculation unit 20 performs the following processes S21 to S23 on the input information.

[0078] · Process S21 The emotion calculation unit 20 identifies one or more records containing the input emotion from each record of all the situation change list information related to all scenes stored in the situation change list DB 31, and extracts the list of scenes related to the identified records and the corresponding numbers included in the records. For example, when "wasteful" is input as the emotion, regardless of the type of scene, one or more records (for example, reference numerals a and b in FIG. 9) containing the emotion "wasteful" (see FIG. 8) are identified from the situation change list information, and the list of scenes (major classification / minor classification) related to the identified records (for example, "content browsing / news") and the list of corresponding numbers (for example, "1", "3",...) are extracted.

[0079] · Process S22 The emotion calculation unit 20 identifies the life logs related to the user attributes input by the user among the life logs stored in the life log DB 32, and identifies one or more records related to the scene extracted in S21 in this life log. The emotion calculation unit 20 identifies, among the identified records, one or more records that satisfy a predetermined condition, for example, include a knowledge amount T_U and / or an experience amount K_U equal to or greater than a predetermined threshold, and extracts a list of scenes of each classification included in this identified record. For example, the emotion calculation unit 20 identifies one or more records in which at least one of the knowledge amount T_U and the experience amount K_U is 0.7 or more, and / or one or more records in which the total of the knowledge amount T_U and the experience amount K_U exceeds 1.5, among one or more records related to the scene extracted in S21 in the life log, and extracts the scenes of each classification included in this record.

[0080] · Process S23 The emotion calculation unit 20 identifies, from the situation change list information stored in the situation change list DB 31 under the condition of the same scene, one or more records corresponding to the correspondence number extracted in S21 among the situation change list information related to the scene extracted in S22, and for each of the identified records, extracts the parameters of the prediction degree Y, the expected degree E_U calculation definition, and the external factor G included in this record.

[0081] The emotion calculation unit 20 extracts the knowledge amount T_U and the experience amount K_U included in each of the one or more records identified in S22 from each record of the life log stored in the life log DB 22. The emotion calculation unit 20 calculates the expected degree E_U related to the record extracted from the situation change list DB 31 by substituting the knowledge amount T_U and the experience amount K_U extracted from the life log DB 22 into the expected degree E_U calculation definition of the record extracted from the situation change list DB 31 under the condition of the same scene.

[0082] Under the condition of the same scene as described above, the emotion calculation unit 20 identifies one or more records related to the corresponding number (e.g., "D-1" in FIG. 9) corresponding to the corresponding number (e.g., "1" in FIG. 9) extracted in process S21 from each record of the emotion creation amount calculation definition information stored in the emotion creation amount calculation definition DB 33, and extracts the emotion creation amount calculation definition included in the identified records. Under the condition of the same scene as described above, the emotion calculation unit 20 substitutes (1) the event / parameters of the predicted degree Y and external factor G extracted above from the record related to the extracted corresponding number (e.g., "1" in FIG. 9) stored in the situation change list DB 31, (2) the calculated expected degree E_U, and (3) the knowledge amount T_U and experience amount K_U extracted above from the life log DB 32 into the extracted emotion creation amount calculation definition to calculate the emotion creation amount.

[0083] When the number of the identified records is plural in the emotion creation amount calculation definition DB 33 as described above, the emotion calculation unit 20 calculates the emotion creation amount for each of the plural records, and among these records, identifies one or more records (see reference numeral c in FIG. 9) including the emotion creation amount calculation definition used for calculating the largest emotion creation amount, and extracts the corresponding number (e.g., corresponding number "D1-3" in FIG. 9) included in this record.

[0084] The emotion calculation unit 20 identifies the record corresponding to the extracted corresponding number from the situation change list DB 31, and outputs the scene, "Situation A", "Situation B", and emotion included in this record to the user.

[0085] When the information such as the output emotion is plural, that is, when plural records related to the same largest emotion creation amount are extracted, these information are presented together, or only the first extracted piece of information is presented, etc. The expression method when information on various situations and emotions is output to the user is arbitrary. For example, like the "situation of discarding the surplus of a bumper crop" shown in FIG. 9, it can be presented to the user by an expression method such as using the words described in "Situation A", "Situation B", etc. to fit the output result into a fixed sentence and display it as text.

[0086] FIG. 10 is a block diagram showing an example of the hardware configuration of an emotion estimation device according to an embodiment of the present invention. In the example shown in FIG. 10, the emotion estimation device 100 according to the above embodiment is configured by, for example, a server computer or a personal computer, and has a hardware processor 111A such as a CPU (Central Processing Unit). Then, a program memory 111B, a data memory 112, an input / output interface 113, and a communication interface 114 are connected to the hardware processor 111A via a bus 115.

[0087] The communication interface 114 includes, for example, one or more wireless communication interface units, and enables the transmission and reception of information with a communication network NW. As the wireless interface, for example, an interface adopting a low-power wireless data communication standard such as a wireless LAN (Local Area Network) is used.

[0088] An input device 200 and an output device 300 used by a user or the like, which are attached to the emotion estimation device 100, are connected to the input / output interface 113. The input / output interface 113 captures operation data input by a user or the like through an input device 200 such as a keyboard, a touch panel, a touchpad, or a mouse, and performs a process of outputting and displaying the output data to an output device 300 including a display device using liquid crystal or organic EL (Electro Luminescence) or the like. Note that, as the input device 200 and the output device 300, devices built in the emotion estimation device 100 may be used, or input devices and output devices of other information terminals capable of communicating with the emotion estimation device 100 via the network NW may also be used.

[0089] The program memory 111B is used in combination with a non-volatile memory such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) that can be written and read at any time and a non-volatile memory such as a ROM (Read Only Memory) as a non-temporary tangible storage medium, and stores programs necessary for executing various control processes and the like according to one embodiment.

[0090] The data memory 112 is used in combination with a non-volatile memory such as the above and a volatile memory such as a RAM (Random Access Memory) as a tangible storage medium, and is used to store various data acquired and created in the process of performing various processes.

[0091] The emotion estimation device 100 according to one embodiment of the present invention can be configured as a data processing device having a scene / attribute input unit 10, an emotion calculation unit 20, an output unit 40, and a DB input unit 50 shown in FIG. 1 as a software processing functional unit.

[0092] The storage devices used as a working memory and the like by each part of the emotion estimation device 100 and the storage device used as the storage unit 30 can be configured by using the data memory 112 shown in FIG. 10. However, these configured storage areas are not essential components within the emotion estimation device 100. For example, they may be areas provided in an external storage medium such as a USB (Universal Serial Bus) memory or a storage device such as a database server arranged in the cloud.

[0093] The processing function parts in each of the above scene / attribute input unit 10, emotion calculation unit 20, output unit 40, and DB input unit 50 can all be realized by causing the hardware processor 111A to read and execute the program stored in the program memory 111B. Note that some or all of these processing function parts may be realized in other various forms including integrated circuits such as an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA).

[0094] In addition, the methods described in each embodiment can be stored as a program (software means) to be executed by a computer on a recording medium such as a magnetic disk (e.g., a floppy (registered trademark) disk, a hard disk, etc.), an optical disc (e.g., a CD-ROM, a DVD, an MO, etc.), or a semiconductor memory (e.g., a ROM, a RAM, a flash memory, etc.), and can also be transmitted and distributed through a communication medium. Note that the program stored on the medium side includes a setting program for configuring software means (including not only an execution program but also a table and a data structure) to be executed by a computer within the computer. The computer that realizes this device reads the program recorded on the recording medium, and in some cases, constructs software means using the setting program, and executes the above-described processing by controlling the operation by this software means. Note that the recording medium referred to in this specification includes not only a medium for distribution but also a storage medium such as a magnetic disk or a semiconductor memory provided inside a computer or in a device connected via a network.

[0095] Note that the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the gist thereof at the implementation stage. Also, the embodiments may be combined as appropriate, and in that case, the combined effects can be obtained. Furthermore, the above-described embodiments include various inventions, and various inventions can be extracted by combinations selected from a plurality of disclosed constituent elements. For example, even if some constituent elements are deleted from all the constituent elements shown in the embodiments, if the problem can be solved and the effects can be obtained, the configuration from which these constituent elements are deleted can be extracted as an invention.

Explanation of Reference Numerals

[0096] 100... Emotion estimation device 10... Scene / attribute input unit 20... Emotion calculation unit 30... Storage unit 31... Situation change list DB 32... Life log DB 33…Emotion creation amount calculation definition DB 40…Output section 50…DB input section 51…Scene creation / input section 52…Situation / emotion data input (acquisition) section 53…Prediction degree input (acquisition) section 54…Expected degree definition input section 55…Knowledge / experience data acquisition section 56…Calculation definition input section 57…Parameter specification section

Claims

1. First situation, second situation changed from the first situation, emotion information representing the user's emotional expression when changing from the first situation to the second situation, a prediction degree indicating the magnitude of the possibility of changing from the first situation to the second situation, and an expectation degree indicating the magnitude of the user's expectation for changing from the first situation to the second situation, together with information indicating the scene with which the user is involved, are first management information managed; at least one of a knowledge amount indicating the amount of knowledge the user has regarding the scene and an experience amount indicating the amount of experience the user has regarding the scene is second management information managed; and a calculation definition of an emotion creation amount indicating the strength of the emotional expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information, together with information indicating the scene, is used by third management information managed to estimate the emotion generated in the user, a device comprising: An estimation unit that extracts the prediction degree and the expectation degree corresponding to the scene with which the user is currently involved from the first management information, extracts the second management information corresponding to the currently involved scene, and based on the extracted results and the calculation definition of the emotion creation amount corresponding to the scene with which the user is currently involved, estimates an emotion creation amount indicating the strength of the emotional expression created by the user when changing from the first situation to the second situation; An emotion estimation device comprising the above.

2. The first management information includes a variable factor that is a factor affecting the fluctuation of the emotion creation amount. The calculation definition of the emotion creation amount is a calculation definition based on the prediction degree, the expectation degree, the variable factor, and the second management information. The estimation unit Based on the extracted results and the calculation definition of the emotion creation amount, estimates an emotion creation amount indicating the strength of the emotional expression created by the user. The emotion estimation device according to Claim 1.

3. The expectation degree is calculated based on at least one of the knowledge amount the user has regarding the scene and the experience amount indicating the amount of experience the user has regarding the scene. The emotion estimation device according to Claim 1.

4. The first situation, the second situation changed from the first situation, emotion information representing the user's emotional expression when changing from the first situation to the second situation, a prediction degree indicating the likelihood of changing from the first situation to the second situation, and an expectation degree indicating the magnitude of the user's expectation for changing from the first situation to the second situation are first management information managed together with information indicating the scene with which the user is involved, second management information in which at least one of the amount of knowledge indicating the amount of knowledge the user has about the scene and the amount of experience indicating the amount of experience the user has about the scene is managed, and a calculation definition of an emotion creation amount indicating the intensity of the emotional expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information is an apparatus for estimating the emotion generated in the user using third management information managed together with the information indicating the scene, An estimation unit that extracts the prediction degree and the expectation degree corresponding to the emotion information representing the emotional expression that the user wants to share with others from the first management information, extracts the information indicating the scene managed together with the extracted result in the first management information, extracts management information in which at least one of the amount of knowledge and the amount of experience in the second management information corresponding to the extracted scene satisfies the condition, and estimates an emotion creation amount indicating the intensity of the emotional expression created by the user based on the extracted prediction degree, the expectation degree, the extracted second management information, and the calculation definition of the emotion creation amount, An emotion estimation device comprising the same.

5. The first situation, the second situation changed from the first situation, the emotion information representing the user's emotional expression when changing from the first situation to the second situation, the prediction degree indicating the likelihood of changing from the first situation to the second situation, and the expectation degree indicating the magnitude of the user's expectation for the change from the first situation to the second situation are managed together with the information indicating the scene with which the user is involved as first management information, the knowledge amount indicating the amount of knowledge the user has about the scene and the experience amount indicating the amount of experience the user has about the scene are managed as second management information, and a calculation definition of the emotion creation amount indicating the intensity of the emotional expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information is performed by an emotion estimation device that estimates the emotion generated in the user using third management information managed together with the information indicating the scene, extracting, by an estimation unit of the emotion estimation device, the prediction degree and the expectation degree corresponding to the scene with which the user is currently involved from the first management information, extracting the second management information corresponding to the scene with which the user is currently involved, and estimating, based on the extracted results and the calculation definition of the emotion creation amount corresponding to the scene with which the user is currently involved, the emotion creation amount indicating the intensity of the emotional expression created by the user when changing from the first situation to the second situation An emotion estimation method comprising the steps of: **Claim 6** The first situation, the second situation changed from the first situation, the emotion information representing the user's emotion expression when changing from the first situation to the second situation, the prediction degree indicating the magnitude of the possibility of changing from the first situation to the second situation, and the expectation degree indicating the magnitude of the user's expectation for changing from the first situation to the second situation are managed together with the information indicating the scene in which the user is involved as the first management information, at least one of the knowledge amount indicating the amount of knowledge the user has about the scene and the experience amount indicating the amount of experience the user has about the scene is managed as the second management information, and a calculation definition of the emotion creation amount indicating the strength of the emotion expression created by the user when changing from the first situation to the second situation based on the prediction degree, the expectation degree, and the second management information is performed by an emotion estimation device that estimates the emotion generated in the user using the third management information managed together with the information indicating the scene, extracting, by the estimation unit of the emotion estimation device, from the first management information, the prediction degree and the expectation degree corresponding to the emotion information representing the emotion expression that the user wants to share with others, extracting the information indicating the scene managed together with the extracted result in the first management information, extracting the management information in which at least one of the knowledge amount and the experience amount among the second management information corresponding to the extracted scene satisfies the condition, and estimating, based on the extracted prediction degree, expectation degree, the extracted second management information, and the calculation definition of the emotion creation amount, the emotion creation amount indicating the strength of the emotion expression created by the user An emotion estimation method comprising the above.

7. An emotion estimation processing program that causes a processor to function as each part of the emotion estimation device according to any one of Claims 1 to 4.

Images