JP2012113589A - Action motivating device, action motivating method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve effect of motivating an action of a user who browses comments input by another user.SOLUTION: A comment obtaining section 11 obtains comments and action information of an input user from a comment input terminal 6, and a user information obtaining section 13 obtains biological information, voice data and image data of the input user. An emotional change analyzing section 15 determines a psychological state of the input user according to predetermined conditions based on the biological information, the voice data and/or the image data of the input user corresponding to the comments of the input user to calculate an emotion value. An information selecting section 16 selects comments to be displayed in descending order of the emotion value calculated by the emotional change analyzing section 15 to determine display order. A display generating section 17 generates a display screen for displaying the comments in the order selected and determined by the information selecting section 16 and transmits it to a comment display terminal 7 when receiving a comment display instruction from the comment display terminal 7.

Description

  The present invention relates to an information processing system for a comment on a topic, and relates to an action promoting device, an action promoting method, and a program for promoting an action of a viewer of the comment.

  With the development of the Internet, there are an increasing number of sites where you can freely write comments (opinions, impressions, reviews, etc.) on various topics on the Internet and share them with others (eg, blogs and bulletin boards). , Public Opinion site, word of mouth site). It is known that such comment information on another topic strongly affects the behavior of comment viewers on the topic (see Non-Patent Document 1). For example, in the field of the market, it has been pointed out that it affects the purchasing behavior of the viewer who saw the word of mouth comment, and the expectation as a new advertising means is also increasing.

  Also, on the movie criticism site, the comments of those who have already seen the movie are clues for decision making for those who have not yet seen the movie. Comment information such as word of mouth includes comments such as adding new comments and communicating information to someone else in addition to actions directly related to topics such as buying things and watching movies as described above. It also affects various indirect actions beyond “see”. Since writing and browsing of sites like these are possible not only from PCs but also from information communication devices such as mobile phones, it is easy to collect more opinions from more people. Yes.

  Patent Documents 1 and 2 are techniques related to sentence analysis for detecting the magnitude of evaluation included in comments of commenters. Moreover, there are Patent Documents 3 to 8 as techniques related to the detection of emotions of comment speakers.

  Patent Document 1 discloses an information processing apparatus that can automatically extract an expression representing emotionality from a sentence. In the technique of Patent Document 1, a word representing emotion and a word for emphasizing or denying it are prepared in advance as a dictionary. Then, morphological analysis and syntactic analysis of the input sentence, and using the resulting syntactic structure and the meaning of the word, the word that is the center of emotionality and expressions that emphasize or deny it Extract.

  Patent Document 2 discloses a method for obtaining a reputation result of a content in a relatively small configuration and in a relatively short time. In the technology of Patent Document 2, a morphological analysis is performed on a sentence in which an opinion / impression is written, a positive reputation expression word such as “It was interesting”, a negative reputation such as “It was boring” A negative evaluation value is assigned to the expression word, and the evaluation value of each opinion / impression is calculated by weighting them with the expression word such as “very” or “very”.

  Patent Document 3 discloses an information providing apparatus that provides the psychological state of a subject as information in order to evaluate an intellectual activity in a meeting or a meeting. In the technique of Patent Document 3, emotions such as “sympathy / disgust / indifference” are detected by using physiological data and behavioral data for an actual discussion scene with respect to the emotion detection of a speaker.

  Patent Document 4 discloses a behavior control system for a robot that selects a behavior by comprehensively judging the situation where a robot is placed such as recognition results of external environments such as vision and hearing, and internal states such as instinct and emotion. is doing. In the technique of Patent Document 4, several types of emotions such as instinct and emotion are managed as mathematical models.

  Patent Document 5 expresses emotions that change over time in an easy-to-understand manner without being restricted by changes in human mental state or weathering of emotions, and the emotion information is useful for searching information content such as daily experiences. A presentation method, emotion information display method, and information content search method are disclosed. In the technique of Patent Document 5, emotion analysis is performed on emotions that change with time based on biological information such as heart rate and blood pressure, converted into emotion variables, and information content corresponding to the emotion variables is searched.

  Patent Document 6 discloses an emotion image creation system that can express emotions expressed by sentences and words in a meticulously rich manner with respect to the sentences and words of an input sentence. In the technique of Patent Document 6, a predetermined image expressing emotion is held in association with an emotion vector having emotion as an element value, and the word representing the emotion and the word representing the emotion by searching the emotion dictionary for the input sentence The emotion vector of the word is obtained, and the image having the smallest vector difference is selected and displayed.

  Patent Document 7 discloses a sentence creation device and a sentence creation auxiliary device that create a sentence including characters that are objectively input based on a user's biological information and information about the environment around the apparatus. In the technique of Patent Document 7, an emotion or a surrounding situation is inferred based on a measurement value of a person's biological information or a measurement value of an environment around the text creation device, and character data is generated.

  Patent Document 8 discloses a content providing system that selects an advertisement or the like to be displayed based on a speaker's emotion in order to enhance the advertising effect. In the technique of Patent Literature 8, a word is extracted from conversational speech, and content with a high priority added to the word is read. In addition, the audio feature amount is calculated, the emotion type is determined based on the calculated audio feature amount, and the content associated with the emotion type is reproduced.

JP 2003-157253 A JP 2009-15407 A JP 2004-112518 A JP 2003-334785 A JP 2005-124909 A Japanese Patent Laid-Open No. 2005-208992 Japanese Patent Laid-Open No. 2007-156776 JP 2008-170820 A

Takemura Kazuhisa, 2007, Social psychology of consumer behavior, Kitaoji Shobo

  When a large number of comments are gathered on a topic in word-of-mouth information or the like, it is necessary to devise comment information processing means that considers the influence on the action of the comment viewer. If a large number of comments are simply listed, the amount of data is too large to be understood by the comment viewer, and as a result, there is a high possibility that the effect on the comment viewer's behavior will be reduced. That is, in order to support the comment viewer's action decision making, it is important to control the display of the comment of the comment speaker who pays attention to the influence on the comment viewer.

  As for the influence of comments, the magnitude of evaluation indicated by comments can be an indicator. For example, if a comment about a movie was “very interesting” and simply “it was interesting”, the comment viewer's action decision (whether to go to the movie or not) Therefore, it is speculated that the former comment has a stronger influence.

  In addition to the size of the impression / evaluation included in the comment, the emotion of the comment speaker can be cited as an index of the influence on the comment viewer. About the characteristics of communication (persuasive communication) that aims to change the behavior of the recipients of comments and messages and the direction of opinions (persuasive communication), the objective that excludes the emotional elements is included in those messages that contain emotional elements Social psychology surveys show that the persuasive impact on the recipient is higher than typical reports (for example, Nobutaka Gushi, Kaori Karasawa, 2007, Persuasive effects of emotional messages and counter-thoughts, Experimental Social Psychology Research 46 (See (1), pp. 40-52).

  The techniques described in Patent Documents 1 to 8 are not intended to promote the action of the user who has viewed the comment, but display the comment indicating the size of the evaluation / impression described in the comment sentence and the emotion of the commenter. It is not used as a control index.

  This invention is made | formed in view of the above situations, and it aims at providing the action promotion apparatus, the action promotion method, and program which improve the effect of promoting the action of the user who browsed the comment.

The action promoting device according to the first aspect of the present invention provides:
First acquisition means for acquiring a comment input by an input user;
Second acquisition means for acquiring at least two of the input user's biological information, audio data, image data, and action information indicating the action of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis means for calculating the emotional value by judging the psychological state of time,
Generating means for generating a display screen for changing the display feature amount for each comment based on the emotion value;
A transmission means for transmitting the display screen to the terminal of the browsing user;
It is characterized by providing.

The action promotion method according to the second aspect of the present invention includes:
An action promoting method executed in a system in which an input user inputs a comment via a communication network and a browsing user browses the comment,
A first acquisition step of acquiring the comment input by the input user;
A second acquisition step of acquiring at least two of the input user's biological information, voice data, image data, and behavior information indicating the behavior of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis step to determine the psychological state of the time and calculate the emotion value;
Based on the emotion value, a generation step for generating a display screen to be displayed by changing the display feature amount for each comment;
A transmission step of transmitting the display screen to the terminal of the browsing user;
It is characterized by providing.

A program according to the third aspect of the present invention is stored in a computer.
A first acquisition step of acquiring a comment input by an input user;
A second acquisition step of acquiring at least two of the input user's biological information, voice data, image data, and behavior information indicating the behavior of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis step to determine the psychological state of the time and calculate the emotion value;
Based on the emotion value, a generation step for generating a display screen to be displayed by changing the display feature amount for each comment;
A transmission step of transmitting the display screen to a browsing user's terminal;
Is executed.

  According to the present invention, by analyzing the emotion of a user who has input a comment more precisely and controlling the display of the comment based on the analysis result, the effect of promoting the action of the user who viewed the comment is improved. It becomes possible.

It is a figure which shows the structural example of the action promotion system which concerns on Embodiment 1 of this invention. It is a figure which shows the example of the input of the comment which concerns on Embodiment 1, and browsing. It is a figure explaining the example of the estimation reference | standard of an excited state or a calm state. It is a figure explaining the analysis of the time change of the biometric information before and behind comment input. It is a figure explaining the analysis of the temporal change of a long-term emotion value. 3 is a flowchart illustrating an example of an action promoting operation according to the first embodiment. It is a figure which shows the structural example of the action promotion system which concerns on Embodiment 2 of this invention. It is a figure explaining the 1st example of visualization of the emotion which concerns on Embodiment 2. FIG. It is a figure explaining the 2nd example of visualization of the emotion which concerns on Embodiment 2. FIG. It is a figure explaining the 3rd example of visualization of the emotion which concerns on Embodiment 2. FIG. It is a figure explaining the example of definition of an emotion value. 10 is a flowchart illustrating an example of an action promoting operation according to the second embodiment. It is a figure which shows the structural example of the action promotion system which concerns on Embodiment 3 of this invention. It is a figure explaining the example of selection of the comment which concerns on Embodiment 3, rearrangement, and visualization of an emotion. 10 is a flowchart illustrating an example of an action promoting operation according to the third embodiment. It is a figure which shows an example of the hardware constitutions of the action promotion apparatus which concerns on embodiment of this invention.

  In the present invention, the input user is a user who inputs a comment on a certain topic. A browsing user is a user who browses a comment input by an input user. The biological information is, for example, values such as heart rate, blood pressure, body temperature, and sweating. The behavior information is information indicating the behavior of the input user, and is, for example, an elapsed time from login to comment input, an operation log, an access log, or the like. The emotion value is a value indicating a psychological state at the time of inputting an input user's comment, and mainly represents whether the user is in an excited state or in a calm state. Display features include display order, size (relative value for each comment), color, typeface (font), text decoration, frame, background, icons to be displayed, emphasis characters (exclamation marks and emoticons) Etc.).

  EMBODIMENT OF THE INVENTION Below, the form for inventing is demonstrated in detail with reference to drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals.

(Embodiment 1)
FIG. 1 is a diagram showing a configuration example of an action promotion system according to Embodiment 1 of the present invention. The behavior promotion system 101 includes the behavior promotion device 1, a comment input terminal 6, and a comment display terminal 7. The action promoting device 1, the comment input terminal 6, and the comment display terminal 7 can communicate with each other via a network such as the Internet or a mobile communication network (not shown). The action promoting device 1 includes a comment acquisition unit 11, a comment storage unit 12, a user information acquisition unit 13, a user information storage unit 14, an emotion change analysis unit 15, an information selection unit 16, and a display generation unit 17.

  The comment acquisition unit 11 acquires a comment input by the input user to the comment input terminal 6 and stores it in the comment storage unit 12. The action promoting device 1 may not acquire the comment directly via the network, but the management server having the action promoting device 1 may acquire the comment and receive it internally.

  The comment storage unit 12 stores the comment acquired by the comment acquisition unit 11 in association with the acquired time. When the comment acquired by the comment acquisition unit 11 is already associated with the time input by the comment input terminal 6, the comment storage unit 12 stores the comment as it is. The comment storage unit 12 stores the comments selected by the information selection unit 16 and determined in order.

  The user information acquisition unit 13 acquires biometric information, audio data, and image data of the input user detected by the comment input terminal 6 and stores them in the user information storage unit 14. Instead of acquiring the biometric information, voice data, and image data of the input user directly via the network, the behavior promoting device 1 may be acquired by the management server having the behavior promoting device 1 and received internally. Moreover, it is good also as storing the biometric information of the predetermined period before and behind the time of comment input, and the said comment in correlation.

  The user information storage unit 14 stores the input user's biometric information, voice data, and image data acquired by the user information acquisition unit 13 in association with the acquired time. The user information storage unit 14 stores the biometric information, audio data, and image data acquired by the user information acquisition unit 13 as they are when they are already associated with the time detected by the comment input terminal 6. The user information storage unit 14 stores biological information, sound data, and image data in a normal state of the input user in advance.

  The emotion change analysis unit 15 analyzes emotion based on the input user's biometric information, voice data, and image data corresponding to the time when the comment acquisition unit 11 acquires the input user's comment, and calculates an emotion value. Moreover, the emotion change analysis part 15 is good also as calculating an emotion value in consideration of the result of sentence analysis and action information of an input user's comment. When calculating an emotion value in consideration of behavior information, the user information acquisition unit 13 acquires an operation log or an access log of the input user corresponding to the comment acquired by the user information acquisition unit 13 as the behavior information.

  The information selection unit 16 selects comments to be displayed in descending order of emotion values calculated by the emotion change analysis unit 15 and determines the order of display. The information selection unit 16 causes the comment storage unit 12 to store the comments selected and determined in order.

  Upon receiving a comment display instruction from the comment display terminal 7, the display generation unit 17 reads the comment selected by the information selection unit 16 from the comment storage unit 12. The display generation unit 17 generates a display screen that displays the comments read from the comment storage unit 12 in the order determined by the information selection unit 16 and transmits the display screen to the comment display terminal 7. Instead of the action promotion device 1 directly acquiring a comment display instruction via the network, the management server having the behavior promotion apparatus 1 may acquire a comment display instruction and receive it internally.

  The comment input terminal 6 includes an input unit 61, a detection unit 62, and a display unit 63. The input unit 61 receives an input of a comment by the input user, and transmits the comment to the action promoting device 1 or the management server having the behavior promoting device 1. Alternatively, the input unit 61 may send the input comment to the action promoting device 1 in association with the input time. While the input user accesses the action promoting device 1 or the management server having the action promoting device 1 in order to input a comment, the detecting unit 62 detects the input user's biometric information, voice data, and image data to act. It transmits to the management server which has the promotion apparatus 1 or the action promotion apparatus 1. FIG. The display unit 63 displays a comment input screen.

  The comment display terminal 7 includes an input unit 71 and a display unit 72. The input unit 72 receives an instruction to display a comment from the browsing user and transmits the instruction to the behavior promoting device 1 or the management server having the behavior promoting device 1. The display unit 72 displays the display screen sent from the display generation unit 17 of the behavior promoting device 1.

  FIG. 2 is a diagram illustrating an example of comment input and browsing according to the first embodiment. Here, it is assumed that the topic is a movie, and a movie critique site where an input user who views the movie writes an individual comment. First, it is assumed that the comments of the user B, user C, user D, user E, and user F who have already watched the movie are selected by the information selection unit 16 and displayed on the movie criticism site in the determined order. It is assumed that there is a limit that only 5 comments can be displayed for one movie on the movie criticism site.

  When a user A (input user) who has newly watched a movie logs in to the movie critique site using the comment input terminal 6 and inputs a comment about the movie to the input unit 61, the input unit 61 The comment of A is immediately transmitted to the action promotion apparatus 1. On the other hand, when the user A logs into the movie criticism site in order to input a comment, the detection unit 62 of the comment input terminal 6 uses the microphone, the biometric sensor, and the camera connected to the comment input terminal 6. Detection of biometric information, audio data, and image data is started. The detection unit 62 sends the detected biological information, voice data, and image data of the user A to the behavior promoting device 1.

  The comment acquisition unit 11 of the action promoting device 1 stores the comment storage unit 12 in association with the time when the comment of the user A acquired from the comment input terminal 6 is acquired. The user information acquisition unit 13 stores the biometric information, voice data, and image data of the user A acquired from the comment input terminal 6 in the user information storage unit 14 in association with the acquired time.

  The emotion change analysis unit 15 calculates an emotion value at the time of inputting the comment of the user A based on the biological information, voice data, and image data of the user A who inputs the comment. The emotion change analysis unit 15 may analyze the sentence of the user A's comment and add the result to the calculation of the emotion value. Furthermore, the emotion change analysis unit 15 may add the action information of the user A to the calculation of the emotion value.

  The user information storage unit 14 also stores emotion values of the users B, C, D, E, and F who have already input comments. The information selection unit 16 rearranges the comments in descending order of emotion values. When the magnitude of the emotion value is user A> user B> user C> user D> user E> user F, the information selection unit 16 sets the user A, user B, user C, user D, User E's five comments are selected, and the display order is determined in descending order of emotion values. Such selection / rearrangement of comments using emotion values may be automatically performed by the action promoting device 1 in advance before the browsing user browses, or a filter button such as “rearranged in descending order of emotion values”. By providing it in a Web page for browsing, the browsing user may actively perform the browsing.

  In addition, a web page for browsing buttons for voting and applauding each comment is prepared, and a movie can be evaluated by voting and applauding the comment even if the viewing user does not input the comment. You may be able to do it. In this case, the number of votes and applause performed by other users are given to each comment.

  When a user G (viewing user) who has not yet watched the movie uses the comment display terminal 7 to access the comment web page of the movie on the movie criticism site, the comment display terminal 7 promotes action to display a comment. Transmit to device 1. Upon receiving a comment display instruction from the comment display terminal 7, the display generation unit 17 displays a display screen in which comments are displayed in the order of user A, user B, user C, user D, and user E determined by the information selection unit 16. It is generated and transmitted to the comment display terminal 7. The display unit 72 of the comment display terminal 7 displays the display screen received from the display generation unit 17.

  FIG. 3 is a diagram illustrating an example of an estimation criterion for an excited state or a calm state. In the example of FIG. 3, the estimation criteria include “heart rate”, “pulse rate”, “body temperature”, “blood pressure”, “line of sight”, “speech (voice volume)”, “speech (speed)”, “speech (high / low) ) ”Item. The user information storage unit 14 stores in advance biometric information, voice data, and image data of an input user in a normal calm state by storing a medical examination chart and data acquired on a daily basis.

  Regarding the heart rate acquired by the detection unit 62 of the comment input terminal 6 using a biometric sensor, the emotion change analysis unit 15 of the behavior promoting device 1 increases from the average heart rate indicating a calm state by a predetermined threshold value X or more. If it is detected, it is estimated to be excited. In addition, the emotional change analysis unit 15 estimates a calm state when the increase in heart rate is smaller than the threshold value X and when the value of the heart rate is smaller than the average heart rate. Similarly, the emotional change analysis unit 15 estimates that the pulse rate, blood pressure, and body temperature are in an excited state when they are increased by a predetermined threshold or more from values indicating a calm state. Further, when the increase value is smaller than a predetermined threshold and when the values of the pulse rate, blood pressure, and body temperature are smaller than the values indicating the calm state, the calm state is estimated.

  In addition, the detection unit 62 acquires the voice data using a microphone if there is a voice reaction when the comment of the input user is input. Alternatively, when the user inputs a comment by voice, the detection unit 62 acquires the voice data using a microphone. The emotion change analysis unit 15 compares the voice data with the voice level, speed, and height of the voice data stored in the user information storage unit 14 according to a predetermined value from the value indicating the voice calm state. If it is higher than the threshold value, it is estimated as an excited state. That is, when the loudness of the voice is greater than or equal to a predetermined threshold, the utterance speed is faster than the predetermined threshold, or when the voice is higher than the predetermined threshold, the state of excitement is estimated. Further, when the increase value is smaller than the predetermined threshold value and when the value of the voice data is smaller than the value indicating the calm state, the calm state is estimated.

  Furthermore, the detection unit 62 acquires image data when the input user's comment is input using the camera. The emotion change analysis unit 15 estimates the excited state by analyzing the line-of-sight movement of the input user regarding the image data of the input user acquired by the detection unit 62 using the camera. For example, if the line of sight is fixed with respect to one's own comment at the time of comment, it is estimated as an excited state, and if the line of sight is shaken, it is estimated as a calm state. Alternatively, the excited state may be estimated by comparing with image data in a calm state stored in the user information storage unit 14.

  FIG. 4 is a diagram for explaining analysis of temporal changes in biological information before and after comment input. The final determination of emotion values follows analysis of temporal changes in biometric information before and after comment input. The emotion change analysis unit 15 analyzes temporal changes in the biometric information before and after inputting a comment for an item that is estimated to be in an excited state according to the reference in FIG. For example, in both FIGS. 4A and 4B, the heart rate of the input user at the time of inputting a comment is estimated to be an excited state because it shows an increase of the threshold value X or more from the normal calm state, but the emotion change analysis unit 15 Determines whether or not the comment input action itself is excited by referring to the state at the time of login to the site where the comment is input. In FIG. 4 (a), it was close to a calm state at the time of login, but since the heart rate suddenly increased immediately before the comment input, the input user is likely to be excited about the content of the comment input. On the other hand, in FIG. 4B as well, since the heart rate is rather lower than when logging in, it is unlikely that the input user is excited about the content of the comment input.

  At this time, for an item for determining whether the value is an excited state or a calm state, such as the item “heart rate”, the emotion value is calculated as follows, for example. For an item estimated to be in an excited state according to the criteria of FIG. 3, α−β is a rise value from the calm state of the biological information at the time of comment input, and β is a rise value from the calm state of the biological information at the time of login. If> X, +10 points, and if α−β ≦ X, +0 points. Even when it is estimated that the state is calm according to the standard of FIG. Alternatively, the threshold Y is set, and if α / β> Y, +10 points, and if α / β ≦ Y, +0 points.

  If there is only one of the excited state (fixed) and the calm state (swaying) as in the item of “line of sight”, if the excited state is estimated according to the criteria of FIG. When a comment is entered, the excitement state is +10 points, and when the comment is entered and the comment is entered, the excitement state is +0 points. Even when it is estimated that the state is calm according to the standard of FIG. The emotion change analysis unit 15 calculates the number of points of all items in this way, and calculates the total as an emotion value.

  As for the item of “heart rate”, with respect to an item for determining whether it is an excited state or a calm state by providing a threshold value, for example, a first threshold value K and a second threshold value L larger than that are provided, and L> If α−β> K, +10 points, and if α−β ≧ L, +20 points, the larger the value indicating the excited state, the higher the number of points may be. Or it is good also as normalizing each biometric information, audio | voice data, and image data, and calculating an emotion value by adding together. In addition, the calculation of the emotion value is not limited to the increase value, and an increase rate may be used.

  The number of points may also be changed depending on the value of the biological information at the time of inputting a comment for an item that is estimated to be in a calm state in accordance with the criteria of FIG. For example, if the heart rate at the time of comment input is γ, and the value of biological information in a calm state is ω, it is +0 points when X + ω> γ ≧ ω, and −10 points when ω> γ.

  Moreover, the time change after inputting a comment may be an analysis target. In FIG. 4 (a), the excitement state persists even after the comment is input, so it is considered that the relevance of the comment input and the excitement state is high. In FIG. Therefore, it is considered that the relevance between the comment input and the excitement state is low by regarding the input user as having regained calmness by inputting the comment. In this case, for example, α is corrected using the slope of the heart rate change function after inputting a comment as an index. In addition, the value of the biometric information at the time of comment input is compared with the value of biometric information after a predetermined time has elapsed from the time of comment input, and it is determined whether the drop value or the drop rate exceeds a threshold value. The value may be corrected.

  For the calculation of the emotion value, not only biometric information, audio data, and image data, but also the magnitude of evaluation indicated by the comment may be used together. In this case, as a result of sentence analysis of the comment input by the input user, the number of points is defined in advance for a predetermined word extracted from the comment (for example, “Very” or “Very”) and added to the emotion value To do. Alternatively, the weight is corrected so that the emotion value becomes larger.

  As described above, after estimating the excitement state with respect to the excitement state at the time of inputting a comment, the emotion value is calculated based on the temporal change of the biological information before and after that, thereby calculating the reliability of the emotion value of the input user. Can be increased.

  FIG. 5 is a diagram for explaining analysis of temporal changes in emotion values over a long period of time. The analysis of the time change may be a long-term analysis as well as a short-term analysis from login to comment input. In the example of FIG. 5, for example, when both the user A and the user B pay attention only to the comment on the topic 3, both have high emotion values when inputting the comment. When the sentence analysis of the comment is performed, since the emphasis “very” is included in the comment of the user B, the thanks value is weighted toward the comment of the user B.

  However, when the time change of the emotion value of the comments from topic 1 to topic 3 is analyzed for each user, user A has a high emotion value only for topic 3, whereas user B always has a high emotion value. From this, it is estimated that the user A feels the comment regarding the topic 3 as a special one. Therefore, the emotion change analysis unit 15 uses a predetermined period before the comment is posted (for example, one month) or an emotion value of a predetermined number of comments (for example, two) before the comment is posted. If the emotion value of the comment is larger than a predetermined threshold, the emotion value of the comment is corrected so as to increase. Alternatively, the emotion change analysis unit 15 has a predetermined period before the comment is posted or the emotion value of a predetermined number of comments before the comment is posted and the emotion value of the comment are equal to or greater than a predetermined threshold. If there is no difference, correction is made so that the emotion value of the comment becomes smaller.

  For calculating the emotion value, not only biological information, audio data, and image data, but also action information of the input user at the comment input terminal 6 may be used in combination. For example, the input time indicating the elapsed time from the input user login to the comment input is stored and used. Alternatively, the operation log or access log of the input user acquired by the user information acquisition unit 13 (for example, a web page viewed immediately before inputting a comment) is used. For example, behavioral characteristics (such as typing speed and strength) are calculated from the operation log and used.

  In this case, similarly to the emotion analysis described with reference to FIGS. 3 to 5, the emotion change analysis unit 15 is configured so that the user information storage unit 14 previously stores the elapsed time from the input user login to the comment input in the normal calm state. Remember typing speed and strength. Then, the emotion change analysis unit 15 preliminarily displays the elapsed time from the input user login to the comment input and the typing speed and strength in the normal calm state, the elapsed time from the login to the comment input at the time of the comment input, and The speed and strength of typing are compared to determine whether the state is excited or calm. Then, the time change is analyzed to calculate the number of points and added to the emotion value. Alternatively, correction is made so that the emotion value becomes larger by weighting a comment in which the elapsed time until the input of the comment, the typing speed and strength, etc. exceed a predetermined threshold.

  In addition, the emotion change analysis unit 15 determines whether or not the web page that the input user was viewing immediately before inputting the comment is a web page related to the topic of the comment, and the web page related to the topic is If it does, a predetermined number of points is added to the emotion value. Alternatively, if the web page is not related to a topic, a predetermined number of points is subtracted from the emotion value. Whether or not a web page is related to a topic is determined based on whether or not the keyword of the topic is included in the web page.

  FIG. 6 is a flowchart illustrating an example of an action promoting operation according to the first embodiment. First, the comment acquisition unit 11 acquires an input user's comment from the comment input terminal 6 (step S11). The comment acquisition unit 11 stores the acquired comment of the input user in the comment storage unit 12 in association with the acquisition time. The user information acquisition unit 13 acquires the input user's biometric information, audio data, and image data from the comment input terminal 6 (step S12).

  The user information acquisition unit 13 stores the acquired biometric information, voice data, and image data of the input user in the user information storage unit 14 in association with the acquired time. The emotion change analysis unit 15 calculates an emotion value based on the input user's biometric information, voice data, and image data corresponding to the comment acquired by the comment acquisition unit 11 (step S13). The information selection unit 16 selects comments to be displayed in descending order of emotion values calculated by the emotion change analysis unit 15, and determines the display order (step S14). The information selection unit 16 causes the comment storage unit 12 to store a comment that has been selected and whose display order has been determined.

  The display generation unit 17 determines whether or not a comment display instruction has been acquired from the comment display terminal 7 (step S15). If a comment display instruction has not been acquired (step S15; NO), the process ends. When a comment display instruction is acquired (step S15; YES), the display generation unit 17 reads the comments selected by the information selection unit 16 from the comment storage unit 12, and displays them in the order determined by the information selection unit 16. Is generated (step S16). And the display production | generation part 17 transmits the produced | generated display screen to the comment display terminal 7 (step S17), and complete | finishes a process.

  As described above, according to the action promotion system 101 of the first embodiment, comments on topics based on the results of sentence analysis of biometric information, voice data, image data, action information, and comments of the input user's voice and images. Analyzing the emotion of the input user at the time of input, calculating the emotion value, selecting and sorting comments with large emotion values, the behavior related to the topic of the browsing user who viewed the comment It can be expected to promote.

(Embodiment 2)
FIG. 7 is a diagram illustrating a configuration example of the action promoting system according to the second embodiment. In the behavior promotion system 102 of the second embodiment, an emotion visualization unit 18 is added instead of the information selection unit 16 of the behavior promotion system 101. The emotion visualization unit 18 generates an image and an icon that match the emotion value of each comment, and the display generation unit 17 reads the comment from the comment storage unit 12 and comments using the image and icon generated by the emotion visualization unit 18. Generate a display screen that displays

  FIG. 8 is a diagram for explaining a first example of emotion visualization according to the second embodiment. In the example of FIG. 8, the display screen generated by the display generation unit 17 when the emotion visualization unit 18 uses a method of changing the font size to be displayed according to the size of the emotion value for each comment is shown. The emotion visualization unit 18 increases the font size for a comment having a larger emotion value. In the example of FIG. 8, emotion values are larger in the order of user B> user A> user D> user E> user C.

  FIG. 9 is a diagram for explaining a second example of emotion visualization according to the second embodiment. In the example of FIG. 9, the emotion visualization unit 18 changes the font color to red as the emotion value of the comment increases in the positive direction, and to blue as the comment value increases in the negative direction. By expressing the excitement and enthusiasm of the input user relative to each other by color, it can be expected that the viewing user can intuitively understand the excitement and enthusiasm of the input user.

  FIG. 10 is a diagram for explaining a third example of emotion visualization according to the second embodiment. When the comment input is not a free description but multiple choice, the emotion visualization unit 18 generates an image or icon that reflects only the emotion value, and the display generation unit 17 uses the image or icon to display the display screen. May be generated.

  For example, as shown in FIG. 10, if the topic “What do you think about XX?” Is an alternative of “great” and “chunky”, the emotion change analysis unit 15 votes The emotion visualizing unit 18 calculates the average value of the emotion values of all the input users for each option, and generates an image or icon that expresses the average value. In the example of FIG. 10, the magnitude of the emotion value is represented by a flame image. The display generation unit 17 generates a voting result display screen using the image and reflects the emotion value of the input user. In FIG. 10, although the number of votes of “great” is lower than the number of votes of “frustration”, the emotion value at the time of voting, that is, excitement and enthusiasm, shows that “great” is larger.

  In particular, in social psychology, it is known that those who recognize their opinions as minorities tend to be in sync with the majority, fearing criticism and isolation from the majority. Such minority's anxiety is mainly influenced by the number of opinions different from oneself, but not only by displaying the number of opinions as shown in FIG. 10, but by expressing emotional factors other than the number of opinions, Even if you are a minority, you can create a situation that makes it easy to send out opinions.

Further, in the majority voting scene as shown in FIG. 10, a counting method using not only the number of votes but also emotion values may be performed. That is, in FIG. 10, when only the number of votes is viewed, it is counted as 5 votes in favor and 30 votes against, but it may be counted by weighting emotion values to each vote. For example, if the total of the emotional value of approval is “50” and the total of the opposite emotional value is calculated as “5”, 5 × 50 = 250 votes for the approval by applying the emotional value to the number of votes. Is corrected as 30 × 5 = 150 votes.

  Such total correction is an index for the information selection unit 16 to determine the selection and display order when the number of votes and applause from the viewing user for each comment is given to each comment in the first embodiment. It may be used. In this case, the information selection unit 16 selects comments having a large value obtained by multiplying the emotion value calculated by the emotion change analysis unit 15 by the number of votes or the number of applauses, and rearranges them in descending order.

  In addition to “excitement / sedation”, an uncomfortable expression or a pleasant expression may be detected by analyzing the expression of the image data input user, and the “comfortable / uncomfortable” state may be estimated. Further, the state of “pleasant / uncomfortable” may be estimated from the voice data. In this case, each value of the “pleasant” state and the “unpleasant” state is calculated, and the emotion value is calculated in consideration of the magnitude of the value. At this time, the time change analysis as described above is also used to calculate the values of the “comfort” state and the “discomfort” state.

  In addition to the examples shown in FIGS. 8 to 10, emotion visualization changes typefaces (fonts), character decorations, frames, backgrounds, icons displayed with comments, emphasized characters (exclamation marks, emoticons, etc.), etc. Or may be added.

  FIG. 11 is a diagram illustrating an example of emotion value definition. When estimating the “pleasant / uncomfortable” state in addition to the “excited / sedated” state, the emotional visualization is defined by defining four categories of empathy, interest, disgust, and indifference as shown in FIG. The unit 18 may change the image or icon to be generated. For example, the font colors of the four categories are set to different colors, or the display of the comments in which indifference is detected is suppressed. In addition, in a voting scene as shown in FIG. 10, a comment in which indifference is detected may be expressed by an ice picture instead of a flame.

  FIG. 12 is a flowchart illustrating an example of an action promoting operation according to the second embodiment. First, the comment acquisition unit 11 acquires an input user's comment from the comment input terminal 6 (step S21). The comment acquisition unit 11 stores the acquired comment of the input user in the comment storage unit 12 in association with the acquisition time. The user information acquisition unit 13 acquires the input user's biometric information, audio data, and image data from the comment input terminal 6 (step S22).

  The user information acquisition unit 13 stores the acquired biometric information, voice data, and image data of the input user in the user information storage unit 14 in association with the acquired time. The emotion change analysis unit 15 calculates an emotion value based on the input user's biometric information, voice data, and image data corresponding to the comment acquired by the comment acquisition unit 11 (step S23). The emotion visualization unit 18 generates an image or an icon based on the emotion value calculated by the emotion change analysis unit 15 (step S24).

  The display generation unit 17 determines whether or not a comment display instruction has been acquired from the comment display terminal 7 (step S25). If a comment display instruction has not been acquired (step S25; NO), the process ends. When the display instruction of the comment is acquired (step S25; YES), the display generation unit 17 reads the comment from the comment storage unit 12, and displays the comment using the image or icon generated by the emotion visualization unit 18. Is generated (step S26). And the display production | generation part 17 transmits the produced | generated display screen to the comment display terminal 7 (step S27), and complete | finishes a process.

  As described above, according to the action promotion system 102 of the second embodiment, by changing the display of the comment according to the emotion value calculated for each comment, the comment of the input user is appealed to the intuition of the viewing user. Can do.

(Embodiment 3)
FIG. 13 is a diagram illustrating a configuration example of an action promoting system according to the third embodiment. The behavior promotion system 103 according to the third embodiment includes both the information selection unit 16 of the behavior promotion system 101 and the emotion visualization unit 18 of the behavior promotion system 102.

  FIG. 14 is a diagram for explaining an example of comment selection and rearrangement and emotion visualization according to the third embodiment. Similar to the example of FIG. 8, the example of FIG. 14 assumes a movie critique site where a topic is a movie and an input user who views the movie individually writes comments. Further, it is assumed that there is a limit that only 5 comments can be displayed for one movie on the movie criticism site. The information selection unit 16 selects five comments in descending order of the emotion value calculated by the emotion change analysis unit 15 and determines the display order in descending order. The emotion visualization unit 18 employs a method of changing the font size to be displayed according to the size of the emotion value for each comment, and the display generation unit 17 generates a display screen with a larger font size for a comment with a larger emotion value. In the example of FIG. 14, the emotion values increase in the order of user B> user A> user D> user E> user C.

  FIG. 15 is a flowchart illustrating an example of an action promoting operation according to the third embodiment. First, the comment acquisition unit 11 acquires an input user's comment from the comment input terminal 6 (step S31). The comment acquisition unit 11 stores the acquired comment of the input user in the comment storage unit 12 in association with the acquisition time. The user information acquisition unit 13 acquires the input user's biometric information, audio data, and image data from the comment input terminal 6 (step S32).

  The user information acquisition unit 13 stores the acquired biometric information, voice data, and image data of the input user in the user information storage unit 14 in association with the acquired time. The emotion change analysis unit 15 calculates an emotion value based on the input user's biometric information, voice data, and image data corresponding to the comment acquired by the comment acquisition unit 11 (step S33). The information selection unit 16 selects a comment to be displayed based on the emotion value calculated by the emotion change analysis unit 15, and determines the display order (step S34). The information selection unit 16 causes the comment storage unit 12 to store a comment that has been selected and whose display order has been determined. The emotion visualization unit 18 generates an image or an icon based on the emotion value calculated by the emotion change analysis unit 15 (step S35).

  The display generation unit 17 determines whether or not a comment display instruction has been acquired from the comment display terminal 7 (step S36). If a comment display instruction has not been acquired (step S36; NO), the process ends. When an instruction to display a comment is acquired (step S36; YES), the display generation unit 17 reads the comment selected by the information selection unit 16 from the comment storage unit 12, and uses the image or icon generated by the emotion visualization unit 18. Then, a display screen for displaying comments in the order determined by the information selection unit 16 is generated (step S37). And the display production | generation part 17 transmits the produced | generated display screen to the comment display terminal 7 (step S38), and complete | finishes a process.

  As described above, according to the action promoting system 103 of the third embodiment, the emotion of the user who has input a comment regarding the topic based on the voice analysis of the input user, the image, the biological information, the action information, and the comment analysis result of the comment. By analyzing comments more precisely, selecting and rearranging comments based on the analysis results, and changing the display of the comment according to the emotion value calculated for each comment, the user viewing the comment with the emotion of the input user Can be expected to promote actions related to the topic.

  FIG. 16 is a diagram illustrating an example of a hardware configuration of the behavior promoting device according to the embodiment of the present invention. As shown in FIG. 16, the behavior promoting device 1 includes a control unit 21, a main storage unit 22, an external storage unit 23, an operation unit 24, a display unit 25, and a transmission / reception unit 26. The main storage unit 22, the external storage unit 23, the operation unit 24, the display unit 25, and the transmission / reception unit 26 are all connected to the control unit 21 via the internal bus 20.

  The control unit 21 includes a CPU (Central Processing Unit) and the like, and in accordance with a control program 29 stored in the external storage unit 23, the comment acquisition unit 11, the user information acquisition unit 13, and the emotion change analysis unit 15 of the behavior promoting device 1 Each process of the information selection unit 16, the display generation unit 17, and the emotion visualization unit 18 is executed.

  The main storage unit 22 is composed of a RAM (Random-Access Memory) or the like, loads a control program 29 stored in the external storage unit 23, and is used as a work area of the control unit 21.

  The external storage unit 23 includes a nonvolatile memory such as a flash memory, a hard disk, a DVD-RAM (Digital Versatile Disc Random Access Memory), a DVD-RW (Digital Versatile Disc ReWritable), and controls the processing of the action promoting device 1. A program to be executed by the unit 21 is stored in advance, and data stored by the program is supplied to the control unit 21 in accordance with an instruction from the control unit 21, and the data supplied from the control unit 21 is stored. The comment storage unit 12 and the user information storage unit 14 are configured in the external storage unit 23.

  The operation unit 24 includes a pointing device such as a keyboard and mouse, and an interface device that connects the keyboard and pointing device to the internal bus 20. For example, when setting a threshold value, an instruction is supplied to the control unit 21 via the operation unit 24.

  The display unit 25 includes a CRT (Cathode Ray Tube) or an LCD (Liquid Crystal Display), and displays an operation screen when setting a threshold value.

  The transmission / reception unit 26 includes a network termination device or a wireless communication device connected to the network, and a serial interface or a LAN (Local Area Network) interface connected thereto. The transmission / reception unit 26 is connected to the comment input terminal 6 and the comment display terminal 7 via a network.

  The comment acquisition unit 11, the comment storage unit 12, the user information acquisition unit 13, the user information storage unit 14, the emotion change analysis unit 15, the information selection unit 16, and the display generation of the action promoting device 1 illustrated in FIG. The processing of the unit 17 and the emotion visualization unit 18 is performed by the control program 29 using the control unit 21, the main storage unit 22, the external storage unit 23, the operation unit 24, the display unit 25, and the transmission / reception unit 26 as resources. Execute.

  In addition, the hardware configuration and the flowchart described above are merely examples, and can be arbitrarily changed and modified.

  A central part that performs control processing including the control unit 21, the main storage unit 22, the external storage unit 23, the internal bus 20, and the like can be realized by using a normal computer system without using a dedicated system. . For example, a computer program for executing the above operation is stored and distributed in a computer-readable recording medium (flexible disk, CD-ROM, DVD-ROM, etc.), and the computer program is installed in the computer. Thus, the behavior promoting device 1 that executes the above-described processing may be configured. Further, the action promoting device 1 may be configured by storing the computer program in a storage device included in a server device on a communication network such as the Internet and downloading the computer program by a normal computer system.

  Further, when the functions of the action promoting device 1 are realized by sharing of an OS (operating system) and an application program, or by cooperation between the OS and the application program, only the application program portion is stored in a recording medium or a storage device. May be.

  It is also possible to superimpose a computer program on a carrier wave and distribute it via a communication network. For example, the computer program may be posted on a bulletin board (BBS: Bulletin Board System) on a communication network, and the computer program may be distributed via the network. The computer program may be started and executed in the same manner as other application programs under the control of the OS, so that the above-described processing may be executed.

  A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.

(Appendix 1)
First acquisition means for acquiring a comment input by an input user;
Second acquisition means for acquiring at least two of the input user's biological information, audio data, image data, and action information indicating the action of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis means for calculating the emotional value by judging the psychological state of time,
Generating means for generating a display screen for changing the display feature amount for each comment based on the emotion value;
A transmission means for transmitting the display screen to the terminal of the browsing user;
An action promoting device comprising:

(Appendix 2)
The analysis means determines a temporal change from a comment input time of the input user to the biometric information, the audio data, the image data, and / or the behavior information from a predetermined time on a predetermined condition to determine the emotion The action promoting device according to appendix 1, wherein a value is calculated.

(Appendix 3)
The analysis means determines a temporal change from a comment input time of the biometric information, the voice data, the image data, and / or the action information of the input user from a predetermined time on a predetermined condition, and the emotion The behavior promoting device according to appendix 1 or 2, wherein a value is calculated.

(Appendix 4)
The apparatus further comprises storage means for storing the biological information, the audio data, the image data, and / or the behavior information of the input user at normal times,
The analysis means includes the biological information, the voice data, the image data and / or the behavior information of the input user in the normal time, the biological information of the input user at the time of inputting a comment, the voice data, and the image data. And / or comparing the behavior information with each other to determine whether or not the input user is normal at the time of comment input, and for the comment that is not normal, the temporal change is determined under a predetermined condition. The behavior promotion device according to appendix 2 or 3, wherein the emotion value is calculated.

(Appendix 5)
The analysis means is configured such that the emotion value of the comment is greater than a predetermined threshold value by a predetermined period before the input user inputs the comment or the emotion value of other comments of the input user by a predetermined number. In the case, the action promoting device according to any one of appendices 1 to 4, wherein correction is performed to increase the emotion value of the comment by a predetermined value.

(Appendix 6)
The action promotion according to any one of appendices 1 to 5, wherein the generation unit generates the display screen that selects the comments in descending order of the emotion value and displays the comments in descending order of the emotion value. apparatus.

(Appendix 7)
The generation unit changes the display size, color, typeface, character decoration, frame, background, icon to be displayed along with the highlighted character, and the highlighted character according to the size of the emotion value. The action promoting device according to any one of appendices 1 to 6, wherein a display screen is generated.

(Appendix 8)
An extraction means for extracting a phrase representing the emotion of the input user of the comment from the sentence of the comment;
The analysis means is configured to determine the emotion based on a combination of at least two of the biological information of the input user corresponding to the comment, the voice data, the image data, the behavior information, and a phrase representing the emotion of the input user. 8. The behavior promoting device according to any one of appendices 1 to 7, wherein a value is calculated.

(Appendix 9)
An action promoting method executed in a system in which an input user inputs a comment via a communication network and a browsing user browses the comment,
A first acquisition step of acquiring the comment input by the input user;
A second acquisition step of acquiring at least two of the input user's biological information, voice data, image data, and behavior information indicating the behavior of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis step to determine the psychological state of the time and calculate the emotion value;
Based on the emotion value, a generation step for generating a display screen to be displayed by changing the display feature amount for each comment;
A transmission step of transmitting the display screen to the terminal of the browsing user;
An action promoting method characterized by comprising:

(Appendix 10)
In the analyzing step, a temporal change from a comment input time of the input user to the biometric information, the audio data, the image data, and / or the action information until a predetermined time is determined based on a predetermined condition, and the emotion The behavior promotion method according to appendix 9, wherein a value is calculated.

(Appendix 11)
In the analyzing step, a time change from a comment input time of the biometric information, the voice data, the image data, and / or the behavior information of the input user to a predetermined time is determined based on a predetermined condition and the emotion The action promoting method according to appendix 9 or 10, wherein a value is calculated.

(Appendix 12)
A storage step of storing the biological information, the voice data, the image data, and / or the behavior information of the input user in a normal time;
In the analysis step, the input user's normal biological information, the voice data, the image data and / or the behavior information, and the input user's biological information, voice data, and image data when inputting a comment And / or comparing the behavior information with each other to determine whether or not the input user is normal at the time of comment input, and for the comment that is not normal, the temporal change is determined under a predetermined condition. The behavior promotion method according to appendix 10 or 11, wherein the emotion value is calculated.

(Appendix 13)
In the analysis step, the emotion value of the comment is greater than a predetermined threshold value by a predetermined period before the input user inputs the comment or the emotion value of other comments of the input user by a predetermined number. In the case, the action promotion method according to any one of appendices 9 to 12, wherein correction is performed to increase the emotion value of the comment by a predetermined value.

(Appendix 14)
14. The action promotion according to any one of appendices 9 to 13, wherein, in the generation step, the comments are selected in descending order of the emotion value, and the display screen is displayed in order of the emotion value in descending order. Method.

(Appendix 15)
In the generation step, the comment display size, color, typeface, character decoration, frame, background, at least one of icons to be displayed and highlighted characters are changed according to the emotion value. 15. The action promoting method according to any one of appendices 9 to 14, wherein a display screen is generated.

(Appendix 16)
An extraction step of extracting a phrase representing the emotion of the input user of the comment from the comment text;
In the analyzing step, the emotion is based on a combination of at least two of the biological information of the input user corresponding to the comment, the voice data, the image data, the behavior information, and a phrase representing the emotion of the input user. 16. The behavior promotion method according to any one of appendices 9 to 15, wherein a value is calculated.

(Appendix 17)
On the computer,
A first acquisition step of acquiring a comment input by an input user;
A second acquisition step of acquiring at least two of the input user's biological information, voice data, image data, and behavior information indicating the behavior of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis step to determine the psychological state of the time and calculate the emotion value;
Based on the emotion value, a generation step for generating a display screen to be displayed by changing the display feature amount for each comment;
A transmission step of transmitting the display screen to a browsing user's terminal;
A program characterized by having executed.

1 action promotion device
6 Comment input terminal
7 Comment display terminal
11 Comment acquisition department
12 Comment storage
13 User information acquisition unit
14 User information storage
15 Emotion change analysis part
16 Information selection part
17 Display generator
18 Emotion Visualization Department
21 Control unit
22 Main memory
23 External storage
24 Operation unit
25 Display section
26 Transceiver
29 Control program
61 Input section
62 Detector
63 Display
71 Input section
72 Display Units 101, 102, 103 Action Promotion System

Claims (10)

First acquisition means for acquiring a comment input by an input user;
Second acquisition means for acquiring at least two of the input user's biological information, audio data, image data, and action information indicating the action of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis means for calculating the emotional value by judging the psychological state of time,
Generating means for generating a display screen for changing the display feature amount for each comment based on the emotion value;
A transmission means for transmitting the display screen to the terminal of the browsing user;
An action promoting device comprising:   The analysis means determines a temporal change from a comment input time of the input user to the biometric information, the audio data, the image data, and / or the behavior information from a predetermined time on a predetermined condition to determine the emotion The behavior promoting device according to claim 1, wherein a value is calculated.   The analysis means determines a temporal change from a comment input time of the biometric information, the voice data, the image data, and / or the action information of the input user from a predetermined time on a predetermined condition, and the emotion The behavior promoting device according to claim 1, wherein a value is calculated. The apparatus further comprises storage means for storing the biological information, the audio data, the image data, and / or the behavior information of the input user at normal times,
The analysis means includes the biological information, the voice data, the image data and / or the behavior information of the input user in the normal time, the biological information of the input user at the time of inputting a comment, the voice data, and the image data. And / or comparing the behavior information with each other to determine whether or not the input user is normal at the time of comment input, and for the comment that is not normal, the temporal change is determined under a predetermined condition. The behavior promoting device according to claim 2, wherein the emotion value is calculated.   The analysis means is configured such that the emotion value of the comment is greater than a predetermined threshold value by a predetermined period before the input user inputs the comment or the emotion value of other comments of the input user by a predetermined number. 5. The behavior promoting device according to claim 1, wherein correction is performed to increase the emotion value of the comment by a predetermined value. 6.   The behavior according to claim 1, wherein the generation unit generates the display screen that selects the comments in descending order of the emotion value and displays the comments in descending order of the emotion value. Promotion device.   The generation unit changes the display size, color, typeface, character decoration, frame, background, icon to be displayed along with the highlighted character, and the highlighted character according to the size of the emotion value. The action promoting device according to claim 1, wherein a display screen is generated. An extraction means for extracting a phrase representing the emotion of the input user of the comment from the sentence of the comment;
The analysis means is configured to determine the emotion based on a combination of at least two of the biological information of the input user corresponding to the comment, the voice data, the image data, the behavior information, and a phrase representing the emotion of the input user. The action promoting device according to claim 1, wherein a value is calculated. An action promoting method executed in a system in which an input user inputs a comment via a communication network and a browsing user browses the comment,
A first acquisition step of acquiring the comment input by the input user;
A second acquisition step of acquiring at least two of the input user's biological information, voice data, image data, and behavior information indicating the behavior of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis step to determine the psychological state of the time and calculate the emotion value;
Based on the emotion value, a generation step for generating a display screen to be displayed by changing the display feature amount for each comment;
A transmission step of transmitting the display screen to the terminal of the browsing user;
An action promoting method characterized by comprising: On the computer,
A first acquisition step of acquiring a comment input by an input user;
A second acquisition step of acquiring at least two of the input user's biological information, voice data, image data, and behavior information indicating the behavior of the input user;
The input user's comment input under a predetermined condition based on a combination of at least two of the biological information, the voice data, the image data, and the behavior information of the input user during a period including the time when the comment is input An analysis step to determine the psychological state of the time and calculate the emotion value;
Based on the emotion value, a generation step for generating a display screen to be displayed by changing the display feature amount for each comment;
A transmission step of transmitting the display screen to a browsing user's terminal;
A program characterized by having executed.

Images