JP5921507B2 - Dialog tendency scoring device, method and program - Google Patents

Description

  The present invention relates to a technique for calculating an index representing a tendency between interlocutors.

  Non-Patent Document 1 describes a technique for estimating an empathy interpretation that is a dialogue state between two parties, such as empathy and counterfeit, and displaying the estimated empathy interpretation.

Shiro Kumano, Kazuhiro Otsuka, Amami Mikami, Junji Yamato, “Estimation model and evaluation of empathy / antisense based on facial expression and gaze for multi-person dialogue”, IEICE Technical Report, Human Communication Fundamentals Study Group, HCS 111 (214), pp. 33-38, 2011.

  However, in Non-Patent Document 1, an index that represents a tendency between interlocutors other than the sympathetic interpretation has not been calculated.

  It is an object of the present invention to provide a dialog tendency scoring apparatus, method, and program for calculating an index that represents a tendency between interlocutors other than the sympathy interpretation using the sympathy interpretation.

ある人物iについての個人明確度を以下の式により定義されるA_iとし、

ある人物iについての個人人物選択度を以下の式により定義されるS_iとし、

ある人物iについての個人活発度を以下の式により定義されるE_iとして、

求まった指標P(e_t ^(i,j)=e)を用いて、個人バイアス、個人明確度、個人人物選択度及び個人活発度の中の２個の指標のそれぞれをある人物について計算する個人指標計算部と、計算された人物についての２個の指標に基づいて、その人物を所定のタイプに分類する個人分類部と、人物についての２個の指標を軸とした表に個人分類部による分類結果を表示する表示部と、を備えている。 The dialogue tendency scoring device according to one aspect of the present invention obtains an index P (e _t ^{(i, j)} = e) representing a probability that the empathy state between the person i and the person j at time t is e. A state estimation device, and C ₁ , C ₂ , C ₃ ′, C ₄ ′ , X ₂ , X ₃ are predetermined constants, a ₂ , a ₃ are predetermined real constants, and w _e (e = 1, 2, ..., N _e ), w _e '(e = 1,2, ..., N _e ) are given constants, N is the total number of people, and T is the sympathetic state to consider for calculating each index. a predetermined time length of the data, as the total number of empathy possible states of N _e, and M _i is defined an individual bias for certain persons i by the following equation,

Let the individual clarity for a person i be A _i defined by the following formula:

Let the individual person selectivity for a person i be S _i defined by the following formula:

The individual activity level for a person i is defined as E _i defined as

An individual who calculates each of the two indicators among individual bias, individual clarity, individual person selectivity and individual activity using the obtained index P (e _t ^{(i, j)} = e) An index calculation unit, a personal classification unit that classifies the person into a predetermined type based on the two indexes for the calculated person, and a table centered on the two indexes for the person. And a display unit for displaying the classification result.

ある人物iについての個人明確度を以下の式により定義されるA_iとし、

ある人物iについての個人人物選択度を以下の式により定義されるS_iとし、

ある人物iについての個人活発度を以下の式により定義されるE_iとして、

求まった指標P(e_t ^(i,j)=e)を用いて、個人バイアス、個人明確度、個人人物選択度及び個人活発度の中の２個の指標のそれぞれをある人物について計算する個人指標計算部と、C₅,C ₆ ,C₇,C₈,X₇を所定の定数として、あるグループについてのグループバイアスを以下の式により定義されるMとして、

あるグループについてのグループ明確度を以下の式により定義されるAとして、

あるグループについてのグループ人物選択度を以下の式により定義されるSとして、

あるグループについてのグループ活発度を以下の式により定義されるEとして、

個人指標計算部で計算された２個の指標に基づいて、グループバイアス、グループ明確度、グループ人物選択度及びグループ活発度の中の２個の指標のそれぞれをあるグループについて計算するグループ指標計算部と、計算されたグループについての２個の指標に基づいて、そのグループを所定のタイプに分類するグループ分類部と、グループについての２個の指標を軸とした表にグループ分類部による分類結果を表示する表示部と、を備えている。 The dialogue tendency scoring device according to one aspect of the present invention obtains an index P (e _t ^{(i, j)} = e) representing a probability that the empathy state between the person i and the person j at time t is e. A state estimation device, and C ₁ , C ₂ , C ₃ ′, C ₄ ′ , X ₂ , X ₃ are predetermined constants, a ₂ , a ₃ are predetermined real constants, and w _e (e = 1, 2, ..., N _e ), w _e '(e = 1,2, ..., N _e ) are given constants, N is the total number of people, and T is the sympathetic state to consider for calculating each index. a predetermined time length of the data, as the total number of empathy possible states of N _e, and M _i is defined an individual bias for certain persons i by the following equation,

Let the individual clarity for a person i be A _i defined by the following formula:

Let the individual person selectivity for a person i be S _i defined by the following formula:

The individual activity level for a person i is defined as E _i defined as

An individual who calculates each of the two indicators among individual bias, individual clarity, individual person selectivity and individual activity using the obtained index P (e _t ^{(i, j)} = e) The index calculator and C ₅ , C ₆ , C ₇ , C ₈ , X ₇ as predetermined constants, and the group bias for a group as M defined by the following equation:

The group clarity for a group is defined as A defined by the following formula:

The group person selectivity for a group is defined as S defined by the following formula:

The group activity for a group is defined as E defined by the following formula:

A group index calculation unit that calculates each of the two indexes among group bias, group clarity, group person selectivity, and group activity based on the two indexes calculated by the individual index calculation unit. Based on the two indices for the calculated group, a group classification section for classifying the group into a predetermined type, and a classification result by the group classification section in a table centered on the two indices for the group A display unit for displaying.

  Other than the sympathetic interpretation, it is possible to calculate an index representing the tendency among the interlocutors.

The block diagram which shows the example of the dialog tendency scoring apparatus of 1st embodiment. The block diagram which shows the example of the dialog tendency scoring apparatus of 2nd embodiment. The block diagram which shows the modification of the dialog tendency scoring apparatus of 2nd embodiment. The block diagram which shows the example of the dialog tendency scoring apparatus of 3rd embodiment. The block diagram which shows the example of the dialog tendency scoring apparatus of 4th embodiment. The flowchart which shows the example of the dialogue tendency scoring method of 1st embodiment. The flowchart which shows the example of the dialogue tendency scoring method of 2nd embodiment. The flowchart which shows the example of the dialogue tendency scoring method of 3rd embodiment. The flowchart which shows the example of the dialogue tendency scoring method of 4th embodiment. The figure for demonstrating the example of the classification of the person based on a personal parameter | index. The figure for demonstrating the example of the classification | category of the person based on a group parameter | index. An example of a radar chart for displaying personal indicators. An example of a radar chart for displaying group indicators. The block diagram which shows the example of a dialog state estimation apparatus. The block diagram which shows the example of a parameter learning part. The figure which shows the flow of a process of a learning phase. The figure which shows the flow of a process of an estimation phase. The figure for demonstrating a time difference function. The figure for demonstrating the time difference of a dialogue person's action and empathy interpretation. The figure for demonstrating a change timing function. The figure for demonstrating the effective range of a change timing function. The figure for demonstrating the effective range of a change timing function.

  Hereinafter, each embodiment of a dialogue trend scoring apparatus and method will be described with reference to the drawings.

[First embodiment]
As shown in FIG. 1, the dialogue tendency scoring device of the first embodiment includes, for example, a personal index calculation unit 701, a personal classification unit 702, and a display unit 703.
The dialog tendency scoring method of the first embodiment is realized by the dialog tendency scoring device performing, for example, the processing of each step in FIG.

<Individual index calculator>
The personal index calculation unit 701 calculates at least two indexes among a personal bias, individual clarity, individual person selection, and individual activity for a certain person (step S1). The calculated index is provided to the personal classification unit 702.

The personal index calculation unit 701 calculates a personal index based on the sympathy interpretation obtained by the dialog state estimation apparatus 1 described later, specifically, the probability P (e _t ^{(i, j)} = e) described later.

  Hereinafter, individual bias, individual clarity, individual person selection, and individual activity will be described. In the following description, N is the total number of persons and is a predetermined positive integer. In other words, N is the number of conversation partners of the subject person i + 1. i, j is an index representing a person.

T is the time length of the data of the sympathy state to be considered for calculating each index, and is a predetermined positive integer. In the following example, each index is calculated based on the data of the sympathy state from time t = 1 to time t = T.
N _e is the total number of sympathetic states that can be taken. For example, it is assumed that three states of empathy (e = 1), neutrality (e = 2), and counterfeit (e = 3) can be taken as the empathy states. At this time, N _e = 3.
P (e _t ^{(i, j)} = e) is an index representing the probability that the empathy state between the person i and the person j at time t is e.

C₁は所定の定数である。例えばC₁は0<C₁≦1の実数である。例えば、C₁=1/((N-1)・T)とする。
w_eは、所定の定数である。例えば、w₁=1、w₂=0、w₃=-1とする。 << Personal bias >>
The individual bias with respect to a certain person is an index that indicates which of the two states in which the person is likely to be in the state of empathy / antisense / neither. For example, it is assumed that the personal bias for a certain person is an index that indicates whether the person is likely to be empathetic / antisense. In this case, individual bias M _i for person i is defined as follows, for example.

C ₁ is a predetermined constant. For example, C ₁ is a real number 0 <C ₁ ≦ 1. For example, C ₁ = 1 / ((N−1) · T).
w _e is a predetermined constant. For example, w ₁ = 1, w ₂ = 0, and w ₃ = −1.

If the sympathy interpretation is given by an intensity that is a numerical value from -1 (antisense) to 1 (sympathy) instead of the probability P (e _t ^{(i, j)} = e), the personal index calculation The unit 701 may convert the intensity into a probability, and calculate the personal bias based on the converted probability. For example, the intensity can be converted into a probability as follows.

If intensity ≧ 0, P (e _t ^{(i, j)} = 1) = intensity, P (e _t ^{(i, j)} = 2) = 1−intensity, and P (e _t ^{(i, j)} = 3) Set to 0. When the intensity is less than 0, P (e _t ^{(i, j)} = 1) = 0, P (e _t ^{(i, j)} = 2) = 1−intensity, and P (e _t ^{(i, j )} = 3) = Strength.

C₂は所定の定数である。例えばC₂は0<C₂≦1の実数である。a₂は所定の実数の定数である。 << Individual clarity >>
Individual clarity for a person is an index that represents the ease with which the person's empathy interpretation is concentrated. In other words, it is an index that indicates how easy the person is to concentrate on one interpretation with empathy and dislike. Personal clear of A _i for person i is defined as follows, for example.

C ₂ is a predetermined constant. For example, C ₂ is a real number 0 <C ₂ ≦ 1. a ₂ is a predetermined real number constant.

In order to indicate that the empathy state of the person _i is more likely to be concentrated as A _i is larger, C ₂ , a ₂ , and X ₂ are defined as follows, for example.

C ₂ , a ₂ , and X ₂ may be defined as follows, for example.

maxVar the number of elements is the maximum value of the variance of the data set of sum 1 N _e number, for example, take the following values.

<< Individual person selectivity >>
The individual person selection degree for a certain person is an index representing the ease of concentration of persons who are likely to sympathize with or dislike. In other words, it is an index that indicates whether a certain person is likely to feel empathy or disagreement with a specific person. It is assumed that N ≧ 3 when calculating the individual person selectivity. The individual person selectivity S i for the person _i is defined as follows, for example.

w_eの定義は、<<個人バイアス>>の欄で述べたものと同じである。すなわち、例えば、w₁=1、w₂=0、w₃=-1とする。
C₃’は所定の定数である。例えばC₃’は0<C₃’≦1の実数である。a₃は所定の実数の定数である。 Here, E ^{(i, j)} is defined as follows, for example.

The definition of w _e is the same as described in the << personal bias >> column. That is, for example, w ₁ = 1, w ₂ = 0, and w ₃ = −1.
C ₃ 'is a predetermined constant. For example, C ₃ ′ is a real number with 0 <C ₃ ′ ≦ 1. a ₃ is a predetermined real number constant.

In order to indicate that the larger the S _i is, the more easily the person who is likely to sympathize or dislike the person is likely to concentrate, C ₃ ′, a ₃ , and X are defined as follows, for example.

C ₃ ′, a ₃ , X may be defined as follows, for example.

C₄’は所定の定数である。例えばC₄’は0<C₄’≦1の実数である。例えばC₄’=1/(N-1)としてもよい。 << Personal activity >>
The personal activity level for a certain person is an index that represents the likelihood of the person's empathy and dislike. Personal activity level E _i for person i, for example, is defined as follows.

C ₄ 'is a predetermined constant. For example, C ₄ ′ is a real number with 0 <C ₄ ′ ≦ 1. For example, C ₄ '= 1 / (N-1) may be used.

w_e’は、所定の定数である。例えば、w₁’=1、w₂’=0、w₃’=1とする。 E ′ ^{(i, j)} is the frequency at which the person i and the person j are in a state of empathy or disagreement, and is defined as follows, for example.

w _e 'is a predetermined constant. For example, w ₁ '= 1, w ₂ ' = 0, and w ₃ '= 1.

<Personal Classification Department>
The personal classifying unit 702 classifies the person into a predetermined type based on at least two indexes for the person calculated by the personal index calculating unit 701 (step S2). The classification result is provided to the display unit 703.

  For example, assume that each index is divided into a plurality of ranges. Based on the range in which each index calculated by the personal index calculation unit 701 is included, the person is classified into a predetermined type. Hereinafter, an example of the type classification based on the range in which each index is included will be described with reference to FIG.

For example, it is assumed that the personal bias is divided into three ranges of “easily sympathetic”, “balance”, and “prone to counterfeit”. Let M _c1 and M _{c2 be} predetermined real numbers that satisfy the relationship of M _c1 <M _c2 . At this time, if M _c2 ≦ M _i , the personal bias of the person i is included in the range of “easily sympathetic”, and if M _c1 ≦ M _i <M _c2 , the personal bias of the person i is “balance”. Suppose that M _i <M _c1 , the personal bias of person i is included in the range of “prone to discomfort”.

Further, it is assumed that the individual person selectivity is divided into two ranges of “high” and “low”. Let S _c1 be a predetermined real number. At this time, if S _c1 <S _i , the personal person selectivity of person i is included in the range of “high”, and if S _i ≦ S _c1 , the personal person selectivity of person i is in the range of “low”. Included in

  At this time, as illustrated in FIG. 10, (1) when the personal bias is “easily sympathetic” and the individual person selectivity is “low”, the person i is classified into the type “Happo Beauty”. (2) When the personal bias is “balance” and the individual person selectivity is “low”, the person i is classified into a type of “balance / attentiveness”, and (3) the personal bias tends to be “disgusting”. ”And the individual person selectivity is“ low ”, the person i is classified as a“ critic ”type. (4) When the individual person selectivity is“ high ”, the person i is“ preferred ”. ”.

  Here, the number of ranges to be divided and the number of types to be classified are arbitrary. In the example of FIG. 10, the person is classified into a predetermined type based on the two indicators of individual bias and individual person selectivity. However, the classification is based on the other two indicators or three or more indicators. You may make a classification based on it.

<Display section>
The display unit 703 displays the classification result obtained by the personal classification unit 702 (step S3). The display unit 703 is a display device such as a CRT or a liquid crystal display. The display unit 703 displays, for example, the table shown in FIG.

  Thus, by calculating and displaying the index based on the sympathy interpretation, it is possible to grasp the communication tendency and characteristics of each person.

[Second Embodiment]
As shown in FIG. 2, the dialogue trend scoring device of the first embodiment includes, for example, a personal index calculation unit 701, a display unit 703, a group index calculation unit 704, and a group classification unit 705.

  The dialog tendency scoring method according to the second embodiment is realized by the dialog tendency scoring device performing, for example, each step of FIG.

  Hereinafter, a description will be given centering on differences from the first embodiment. The same parts as those in the first embodiment are denoted by the same reference numerals, and redundant description is omitted.

<Individual index calculator>
Similar to the first embodiment, the personal index calculation unit 701 calculates each of at least two indexes among individual bias, individual clarity, individual person selection, and individual activity for each person included in the group. (Step S1). The calculated index is provided to the group index calculation unit 704.

<Group index calculation department>
Based on the index calculated by the personal index calculation unit 701, the group index calculation unit 704 calculates at least two of the group bias, group clarity, group person selectivity, and group activity for a certain group. (Step S4). The calculated index is provided to the group classification unit 705.

  Hereinafter, group bias, group clarity, group person selection, and group activity will be described.

C₅は所定の定数である。例えばC₅は0<C₅≦1の実数である。例えば、C₅=1/Nとする。 << Group bias >>
The group bias with respect to a certain group is an index representing whether the person included in the group is likely to be at least two of the states representing empathy / antisense / neither. For example, it is assumed that the group bias for a certain person is an index indicating whether the person is likely to be empathetic / antisense. In this case, when the group is composed of N persons, the group bias is defined as follows, for example. N is a predetermined integer of 2 or more.

C ₅ is a predetermined constant. For example, C ₅ is a real number of 0 <C ₅ ≦ 1. For example, C ₅ = 1 / N.

C₆は所定の定数である。例えばC₆は0<C₆≦1の実数である。例えば、C₆=1/Nとする。 << Group clarity >>
The group clarity for a certain group is an index representing the ease of concentration of empathy interpretation of the persons included in the group. In other words, it is an index that indicates how easy a person included in the group is to concentrate on one interpretation, such as empathy or antipathy. When a group is composed of N persons, the group clarity is defined as follows, for example.

C ₆ is a predetermined constant. For example, C ₆ is a real number 0 <C ₆ ≦ 1. For example, C ₆ = 1 / N.

<< Group person selectivity >>
The group person selectivity for a certain group is an index representing the ease of concentration of persons who are likely to sympathize or dislike the persons included in the group. When the group is composed of N persons, the group person selectivity S is defined as follows, for example.

C ₇ is a predetermined constant. For example, C ₇ is a real number 0 <C ₇ ≦ 1. In order to indicate that as the S is larger, the persons included in the group are more likely to be sympathetic or counterfeited, C ₇ and X ₇ are defined as follows, for example.

C ₇ and X ₇ may be defined as follows, for example.

C₈は所定の定数である。例えばC₈は0<C₈≦1の実数である。例えば、C₈=1/Nとする。 << Group activity >>
The group clarity for a certain group is an index that represents the likelihood of empathy and dissatisfaction of the persons included in the group. When the group is composed of N persons, the group activity is defined as follows, for example.

C ₈ is a predetermined constant. For example, C ₈ is a real number of 0 <C ₈ ≦ 1. For example, C ₈ = 1 / N.

<Group classification part>
The group classification unit 705 classifies the group into a predetermined type based on the calculated at least two indices for the group (step S5). The classification result is provided to the display unit 703.

  For example, assume that each index is divided into a plurality of ranges. Based on the range in which each index calculated by the group index calculation unit 704 is included, the group is classified into a predetermined type. Hereinafter, an example of the type classification based on the range in which each index is included will be described with reference to FIG.

For example, it is assumed that the group bias is divided into three ranges of “easily sympathetic”, “balance”, and “prone to antisense”. Let M _c1 ′ and M _c2 ′ be predetermined real numbers that satisfy the relationship of M _c1 ′ <M _c2 ′. At this time, if M _c2 ′ ≦ M, the group bias is included in the range of “easily sympathetic”, and if M _c1 ′ ≦ M <M _c2 ′, the group bias is included in the range of “balance”. If M <M _c1 ′, the group bias is included in the range of “prone to counterfeit”.

Further, it is assumed that the group person selectivity is divided into two ranges of “high” and “low”. Let S _c1 'be a predetermined real number. At this time, if S _c1 ′ <S _i , the group person selectivity is included in the range of “high”, and if S _i ≦ S _c1 ′, the group person selectivity is included in the range of “low”. To do.

  At this time, as illustrated in FIG. 11, (1) when the group bias is “easily sympathetic” and the group person selectivity is “low”, the group is classified into a type of “collaborative dialogue”. (2) When the group bias is “balance” and the group person selectivity is “low”, the group is classified into a type of “all-participatory dialogue”, and (3) the group bias is If the group person selectivity is “low”, the group is classified into a type of “competitive dialogue”. (4) If the group person selectivity is “high”, the group Is classified into a type of “leader-type dialogue”.

  Here, the number of ranges to be divided and the number of types to be classified are arbitrary. Further, in the example of FIG. 11, the person is classified into a predetermined type based on the two indicators of group bias and group person selectivity. However, the classification is based on the other two indicators or three or more indicators. You may make a classification based on it.

<Display section>
The display unit 703 displays the classification result by the group classification unit 705 (step S3). The display unit 703 displays, for example, the table shown in FIG.

  Note that the dialogue tendency scoring device of the second embodiment may further include a personal classification unit 702 as shown in FIG.

  In this case, as in the first embodiment, the personal classification unit 702 classifies the person into a predetermined type based on at least two indices for the person calculated by the personal index calculation unit 701 (step S2). The classification result is provided to the display unit 703.

  As a result, the display unit 703 displays not only the classification result by the group classification unit 705 but also the classification result by the personal classification unit 702.

  Thus, by calculating and displaying the index based on the empathy interpretation, it is possible to grasp the communication tendency and characteristics of each person and / or the entire group.

[Third embodiment]
The dialogue tendency scoring device of the third embodiment includes, for example, a personal index calculation unit 701 and a display unit 703 as shown in FIG.

  The dialog tendency scoring method of the third embodiment is realized by the dialog tendency scoring device performing, for example, the processing of each step in FIG.

  Hereinafter, a description will be given centering on differences from the first embodiment. The same parts as those in the first embodiment are denoted by the same reference numerals, and redundant description is omitted.

<Individual index calculator>
As in the first embodiment, the personal index calculation unit 701 calculates each of at least two indexes among a personal bias, individual clarity, individual person selection, and personal activity for a certain person (step S1). . The calculated index is provided to the display unit 703.

<Display section>
The display unit 703 displays at least two indicators about the person calculated by the individual indicator calculator 701 as a radar chart (step S3). For example, the display unit 703 displays each index as a radar chart as illustrated in FIG.

  Thus, by calculating and displaying the index based on the sympathy interpretation, it is possible to grasp the communication tendency and characteristics of each person.

[Fourth embodiment]
As shown in FIG. 5, the dialog tendency scoring device of the fourth embodiment includes, for example, a personal index calculation unit 701, a group index calculation unit 704, and a display unit 703.

  The dialog tendency scoring method of the fourth embodiment is realized by the dialog tendency scoring device performing, for example, the processing of each step in FIG.

  Hereinafter, a description will be given centering on differences from the second embodiment. The same parts as those in the second embodiment are denoted by the same reference numerals, and redundant description is omitted.

<Individual index calculator>
Similar to the second embodiment, the personal index calculation unit 701 calculates each of at least two indexes among individual bias, individual clarity, individual person selection, and individual activity for each person included in the group. (Step S1). The calculated index is provided to the group index calculation unit 704.

<Group index calculation department>
Similar to the second embodiment, the group index calculation unit 704 is based on the index calculated by the individual index calculation unit 701 and includes at least two of group bias, group clarity, group person selection, and group activity. Each index is calculated for a certain group (step S4). The calculated index is provided to the display unit 703.

<Display section>
The display unit 703 displays at least two indexes for the group calculated by the group index calculation unit 704 as a radar chart. For example, the display unit 703 displays each index as a radar chart as illustrated in FIG.

  Note that, as indicated by a broken line in FIG. 5, the index calculated by the personal index calculation unit 701 may be provided to the display unit 703.

  In this case, the display unit 703 not only displays each index calculated by the group index calculation unit 704 as a radar chart, but also displays each index calculated by the individual index calculation unit 701 as a radar chart.

  Thus, by calculating and displaying the index based on the sympathy interpretation, it is possible to grasp the communication tendency and characteristics of each person and / or the entire group.

[Modification]
The above processes are not only executed in time series in the order described, but may be executed in parallel or individually as required by the processing capability of the apparatus that executes the processes or as necessary.

  Further, when each process in the dialog tendency scoring device is realized by a computer, the processing contents of the functions that the device should have are described by a program. Then, by executing this program on a computer, each process is realized on the computer.

  The program describing the processing contents can be recorded on a computer-readable recording medium. As the computer-readable recording medium, for example, any recording medium such as a magnetic recording device, an optical disk, a magneto-optical recording medium, and a semiconductor memory may be used.

  Each processing means may be configured by executing a predetermined program on a computer, or at least a part of these processing contents may be realized by hardware.

  Needless to say, other modifications are possible without departing from the spirit of the present invention.

[Conversation state estimation device]
Hereinafter, an example of a dialog state estimation apparatus that seeks empathy interpretation will be described. The sympathy interpretation is obtained, for example, by the dialogue state estimation apparatus 1 described in FIG.

<Configuration>
With reference to FIG. 14, the structural example of the dialog state estimation apparatus 1 of this embodiment is demonstrated. The dialog state estimation device 1 includes an input unit 10, an action recognition unit 20, a sympathy interpretation imparting unit 30, a parameter learning unit 40, a learning video storage unit 70, and a model parameter storage unit 74. The learning video storage unit 70 can be configured by, for example, a main storage device such as a RAM (Random Access Memory), an auxiliary storage device including a semiconductor memory element such as a hard disk, an optical disk, or a flash memory. The model parameter storage unit 74 may be configured in the same manner as the learning video storage unit 70, or may be configured by middleware such as a relational database or a key value store.

  With reference to FIG. 15, the structural example of the parameter learning part 40 of this embodiment is demonstrated. The parameter learning unit 40 includes a prior distribution learning unit 42, a timing model learning unit 44, and a static model learning unit 46.

<Learning phase>
With reference to FIG. 11, the operation example in the learning phase of the dialog state estimation apparatus 1 will be described.

  A learning video is input to the input unit 10 (step S11). The learning video is a video that captures a situation where a plurality of persons interact, and at least the head of the dialog must be captured. The video for learning can be taken by preparing one camera for each conversation person and multiplexing the pictures taken by multiple cameras, or by using a fisheye lens, etc. It may be a video of shooting. The input learning video is stored in the learning video storage unit 70.

  The action recognition unit 20 receives the learning video stored in the learning video storage unit 70 and detects facial expressions, gaze, head gestures, presence / absence of speech, etc. as the actions of each conversation person captured in the learning video. And the time series of the action of the dialogue person generated as a result is outputted (Step S21). In this embodiment, four action channels including facial expression, line of sight, head gesture, and presence / absence of speech are recognized. An action channel is a form of action. Facial expressions are the main pathway for expressing emotions. In this embodiment, six states of no expression / smile / smile / bitter smile / thinking / other are the facial expression recognition targets. The line of sight represents at least one of, for example, who is trying to convey emotions and / or observing the actions of others. In this embodiment, the line-of-sight recognition target is a person who is looking at one of the other persons and who is / is not looking at that person. That is, the number of states is the number of interlocutors. Here, the dialogue person refers to all who participate in the dialogue including the subject who is measuring the line of sight. As a method for recognizing a facial expression and a line of sight, a method described in “JP 2012-185727 A (Reference Document 1)” or Non-Patent Document 1 may be used.

  Head gestures are often expressed as an expression of attitude to the opinions of others. In this embodiment, four states of none / whit / neck / tilt / a combination thereof are recognized as head gesture recognition targets. Any known method can be used as a method for recognizing a head gesture. For example, described in "Ejiri Yasushi, Kobayashi Tetsunori," Recognition of Head Gestures During Dialogue ", IEICE Technical Report, PRMU2002-61, pp.31-36, Jul.2002. (Reference 2) This method may be used. The presence or absence of utterance is a major indicator of the conversation role of the speaker / listener. In this embodiment, two states of utterance / silence are recognized as utterance presence / absence recognition targets. As a method for recognizing the presence or absence of utterance, it may be determined that the utterance is made when the audio power in the video is detected and a predetermined threshold is exceeded. Alternatively, the presence or absence of an utterance may be detected by detecting the movement of the conversation person's mouth in the video. Each action may be recognized by a single device, or may be recognized by using a separate device for each action. For example, in the case of facial expression recognition, “Patent No. 4942197 (Reference 3)” may be used as an example of an action recognition device. The action recognition unit 20 may perform manual labeling in the same manner as the empathy interpretation giving unit 30 and output the result.

  For facial expressions and head gestures, “strength” may be estimated and output. The intensity of the facial expression can be obtained from the probability that the facial expression is the target. Further, the strength of the head gesture can be obtained from the ratio of the value of the amplitude of the acquired motion to the maximum value of the amplitude (the maximum value of the scooping angle if it is whispered).

  The empathy interpretation giving unit 30 outputs a learning sympathy interpretation time series in which a plurality of external observers label the sympathy interpretation based on the learning video stored in the learning video storage unit 70 (step S30). The learning sympathy interpretation time series is a time series in which learning videos are presented to a plurality of external observers, and the external observers manually label the sympathetic interpretations between the two conversations at each time. In this embodiment, the three states of empathy / disapproval / neither are targeted as the conversation state between the two parties. The state of dialogue between the two is deeply related to the pressure of entrainment (feeling that many others have to follow the same opinion different from their own) and build consensus building and relationships The basic element above. In addition, these states interpreted by an external observer are collectively referred to as empathy interpretation. That is, the dialogue state interpretation in this embodiment is a sympathy interpretation.

  The learning action time series output from the action recognition unit 20 and the learning empathy interpretation time series output from the empathy interpretation assigning unit 30 are input to the parameter learning unit 40. The parameter learning unit 40 learns model parameters that relate the sympathy interpretation of the external observer and the behavior of the dialog person. Model parameters include a timing model that represents the likelihood of empathy interpretation based on prior distribution of empathy interpretation among the interlocutors, the time difference between the behaviors of the interlocutors and the consistency of the behavior between the interlocutors, and the behavior between the interlocutors. And a static model representing the likelihood of sympathy interpretation based on the co-occurrence of.

  The prior distribution learning unit 42 included in the parameter learning unit 40 learns the prior distribution using the learning sympathy interpretation time series (step S42). The timing model learning unit 44 included in the parameter learning unit 40 learns a timing model using the learning action time series and the learning sympathy interpretation time series (step S44). The static model learning unit 46 included in the parameter learning unit 40 learns a static model using the learning action time series and the learning sympathy interpretation time series (step S46). The obtained model parameters are stored in the model parameter storage unit 74.

<< Overview of model >>
The model of this embodiment will be described in detail. In this embodiment, it is assumed that the empathy interpretation given by the external observer is independent for each combination of two dialogues. Therefore, in the following, it is assumed that there are only two participants. Note that when there are three or more interlocutors, it is only necessary to learn by focusing only on the combination of the two interrogators.

In this embodiment, the posterior probability distribution P (e _t | B) of the sympathetic interpretation e of the external observer at each time t given the time series B of the conversation person's behavior is _expressed using a naive Bayes model. Model and estimate. The naive Bayes model assumes that the stochastic dependence between the dependent variables (here, empathy interpretation) and each explanatory variable (here, the actions of each interactor) is independent among the explanatory variables. The Naive Bayes model is an excellent model that has been confirmed to show high estimation performance in many fields despite being simple. There are two advantages of using the naive Bayes model in this invention. One is that it is easy to avoid over-learning because it does not model all co-occurrence between behavioral channels (for example, a state in which all of facial expressions, gaze, head gestures, and utterances occur simultaneously). This is particularly effective when there are few learning samples for the target variable space. The other is that it is easy to add or delete action channels as observation information.

ここで、P(dt_t ^b|c_t ^b,e_t)はタイミングモデルであり、時刻tの周辺で行動チャネルbについて二者間の行動が時間差dt_t ^bで一致性c_t ^bであるときに外部観察者の共感解釈がeとなる尤度を表す。一致性cとは、二者間で行動が一致しているか否かを表す二値状態のことであり、対話二者の行動のカテゴリが同じか否かで判断する。P(b_t,e_t)は静的モデルであり、時刻tのその瞬間において行動チャネルbが対話二者間でどう共起しているのかをモデル化している。これら二つのモデルについては以下で順に説明する。P(e_t)は共感解釈eの事前分布であり、行動を考えないときに各共感解釈eがどれくらいの確率で生成されるかを表す。 In the naive Bayes model in this embodiment, the posterior probability distribution P (e _t | B) is defined as shown in Equation (1).

Here, P (dt _t ^b | c _t ^b , e _t ) is a timing model, and when the behavior between the two parties is behavior coherence c _t ^b with time difference dt _t ^b around time t Represents the likelihood that the external observer's sympathy interpretation is e. The coincidence c is a binary state indicating whether or not the behaviors of the two parties are the same, and is determined based on whether or not the categories of the behaviors of the two parties are the same. P (b _t , e _t ) is a static model that models how the action channel b co-occurs between the two parties at the instant of time t. These two models will be described in turn below. P (e _t ) is a prior distribution of the sympathy interpretation e, and represents the probability that each sympathy interpretation e is generated when no action is considered.

<< Timing model >>
The timing model for the action channel b in this embodiment is defined as shown in Equation (2).

As is clear from equation (2), this timing model is based on the time difference function P (d ~) representing the likelihood of the empathy interpretation e when the time difference between the actions of the two conversations is dt and the coincidence is c. t _t ^b | c _t ^b , e _t ) and a change timing function π _t representing the timing at which the sympathetic interpretation e changes in the vicinity of the interaction. d to t _t ^b are bin numbers when the time series of the external observer's empathy interpretation is converted into a histogram. For example, the bin size is 200 milliseconds.

  In this embodiment, for each behavior channel, a timing model between two parties is constructed within the behavior channel, but a model between behavior channels may be constructed. For example, the relationship between the time difference dt between the facial expression and the head gesture, the consistency c, and the empathy interpretation e can be modeled. However, in this case, when determining the consistency c, it is necessary to introduce a new category group that can determine the consistency c even between different behavior channels such as positive / neutral / negative, for example. is there. These categories may be recognized when the action channel is detected from the video, or once they are recognized by different category groups for each action channel, and their labels are affirmed if the facial expression is small. You may reclassify later as positive / neutral / negative.

<< Time difference function >>
The time difference function P (d ~ t _t ^b | c _t ^b , e _t ) indicates that the sympathetic interpretation e is based on the coincidence c indicating whether or not the actions between the two parties are matched in the action channel b and the time difference dt. The likelihood of which type is likely to be represented. In this embodiment, bin numbers d to t _t ^b when the time series of the sympathy interpretation of the external observer are converted into a histogram are used. For example, the bin size is 200 milliseconds.

FIG. 18 shows an example of the time difference function of this embodiment. Time difference function ^{_{P (d ~ t t b |}} c t b, e t) determines the likelihood of sympathetic interpretation e by the bin number d ~ t _t ^b Consistency c and time difference of behavior of the interlocutor. FIG. 18A is an example of a time difference function when the actions between the interlocutors match, and FIG. 18B is an example of the time difference function when the actions between the interlocutors do not match. For example, if the behaviors of the interlocutors match, and the time difference between the giver's action expression and the receiver's reaction expression is 500 milliseconds, the likelihood that the empathy interpretation e is "sympathy" The likelihood of about 0.3, “Neither” is about 0.2, and the likelihood of “antisense” is about 0.5. The time difference function calculates the time series of empathy interpretations labeled by external observers in units of time difference bins, and normalizes the sum of likelihood in all time difference bins to be 1 for each category of empathy interpretation e. Ask.

<< Change timing function >>
The change timing function π represents at which timing the empathy interpretation e changes. Viewed another way, the change timing function π determines to what extent the time difference function contributes to the estimation of the empathy interpretation e in equation (1) over which range.

ここで、t_aは対象の相互作用における与え手の行動表出開始の時刻を表す。また、時刻t'は与え手の行動表出開始の時刻をt'=0とし、受け手の反応表出開始時刻をt'=1としたときの相互作用中での相対時間を表し、t'=(t-t_a)/dtとして計算される。 In this embodiment, the change timing function is modeled as shown in Equation (3).

Here, t _a represents the time behavior expression initiation hand given in the interaction of interest. In addition, time t ′ represents the relative time during the interaction when the action expression start time of the giver is t ′ = 0 and the reaction expression start time of the receiver is t ′ = 1. Calculated as = (tt _a ) / dt.

π = 0 represents that the timing model P (dt _t ^b | c _t ^b , e _t ) does not contribute at all in the posterior probability distribution P (e _t | B) represented by the equation (1). π = 1 represents that the timing model P (dt _t ^b | c _t ^b , e _t ) contributes completely in the posterior probability distribution P (e _t | B).

  The condition dt> L represents that the response expression of the receiver is too late with respect to the action expression of the giver. For example, in this embodiment, the threshold value L is 2 seconds. This is a value based on the research results that the expression of the listener's facial expression for the vocabulary important phrase of the speaker occurs in the range of about 500 to 2,500 milliseconds, and it is generally within this range in any action channel. Based on assumptions. For details on the above research results, see “GR Jonsdottir, J. Gratch, E. Fast, and KR Thorisson,“ Fluid semantic back-channel feedback in dialogue: Challenges & progress ”, International Conference Intelligent Virtual Agents (IVA), pp. 154-160, 2007. (Reference 5).

The condition tt _a > W means that the time elapses from the expression of the facial expression of the giving hand that was most recently expressed before time t. When interaction is performed by expressing actions between the two parties, the sympathy interpretation of the external observer is affected by the timing for a certain period of time, but if the next interaction does not occur for a while after that, It is modeled that the effect disappears. The threshold value W may be any value as long as it is a positive value, and there is no problem even if the threshold value W is infinite when interaction between the two parties is continuously generated as in a two-party dialogue. However, in a situation where one person is mainly speaking in a dialogue with a large number of people and the interaction is not always frequent, such as an interaction between two listeners, a threshold is set. It is also possible that W is too long. In this embodiment, the threshold value W is empirically set to 4 seconds. This is based on the experimental result that the estimation accuracy is the highest when the threshold W is set to around 4 seconds.

  FIG. 19 shows an example of sympathy interpretation, behavioral expression of the giver, and response expression of the receiver. The filled pattern in FIG. 19 represents a difference in category of behavior or empathy interpretation. For the values of α and β, for example, α = 0.2 and β = 0.8 are set. These values are determined so that the change timing function π of Equation (3) approximates the cumulative probability most.

  FIG. 20 shows an example of the change timing function π. The points plotted on the graph are actually the empathetic interpretation labels given by nine external observers to the dialogue data of four dialogue groups of four women (16 people in total). This represents the cumulative probability of the change at time t ′. It can be seen that this change timing function can be approximated well. However, α and β are not limited to these values, and α + β = 1, 0 ≦ α ≦ 1, and 0 ≦ β ≦ 1 are satisfied. As a simple setting, “α = 0, β = 1” may be used.

  21 and 22 are diagrams schematically showing an example of the effective range of the change timing function. A black fill indicates a state in which no action is detected, and a white fill and diagonal hatching indicate a category of action. The vertical hatching of the sympathy interpretation indicates empathy, and the horizontal hatching indicates counteraction. FIG. 21A shows an effective range in the case where actions between the interlocutors match. Since the behavior of the giver and the response of the recipient match, “sympathy” continues only during the threshold W. FIG. 21B shows an effective range in the case where the behaviors between the interlocutors are inconsistent. Since the behavior of the giver and the response of the recipient are inconsistent, “disgust” continues only during the threshold W. FIG. 21C shows a situation where the response expression of the receiver is too late with respect to the action expression of the giver, that is, the change timing function is out of the effective range because dt> L. In this case, the “neither” state continues throughout. FIG. 22 shows the effective range when the two dialogues alternately express their actions. The basic concept is the same as in FIGS. 21A to 21C.

<< Static model >>
The static model P (b _t | e _t ) shows the likelihood that the sympathetic interpretation e is generated when a specific action co-occurs between two parties for the action channel b at time t. Modeled.

  The modeling method has been proposed in Patent Document 1 and Non-Patent Document 1 for facial expression and line of sight, and therefore, it is sufficient to follow the description in these documents. What is necessary is just to combine the model of the co-occurrence with the state of the expression for every state. Here, the gaze state between the two may be, for example, three states of mutual gaze / one-side gaze / mutual gaze.

The static model for head gestures is represented by P (g | e). Here, g represents a combination state of head gestures between two parties. When the number of states of the head gestures of interest and N _g, the number of states of combinations of head gestures between two parties becomes N _g × N _g. Arbitrary types and numbers may be targeted as categories, but if the number is too large, overlearning tends to occur when the number of learning samples is small. In that case, the categories prepared first may be further grouped by clustering. For example, Sequential Backward Selection (SBS) is one of the methods. For example, when targeting the category of head gesture, the estimation using only the head gesture, that is, the posterior probability P (e | B): = P (e) P (g '| e) The two categories selected for the best estimation accuracy are integrated into one. It is sufficient to reduce the number of categories one by one by repeating this until just before the estimation accuracy deteriorates. Here, g ′ is a combined state of the head gesture between the two after grouping. As for the presence or absence of utterance, the co-occurrence between two parties is modeled in the same way as the head gesture.

<< Model learning method >>
In this embodiment, since any model is described as a discrete state, in the learning phase, the frequency of how many times the discrete state appears in the learning sample is taken, and finally the frequency is normalized (probabilized). )do it.

  At this time, as a policy to prepare the model, if the group of conversations captured in the learning video used to learn the model parameters is the same as the group of conversations captured in the estimation video for which the conversation state is to be estimated. For example, the parameters are learned independently for each of the two conversations, and the parameters learned from the data of the two conversations may be used for estimation of the two conversations. On the other hand, if the group of interrogators captured in the video for learning differs from the group of interrogators captured in the video for estimation, one model is learned without distinguishing between the two conversations, It is only necessary to make an estimation about two parties who want to estimate using the model.

<Estimation phase>
With reference to FIG. 17, the operation example in the estimation phase of the dialog state estimation apparatus 1 will be described.

  The estimation video is input to the input unit 10 (step S12). The estimation video is a video that captures a situation in which a plurality of persons interact, and at least the conversation person's head must be captured. The estimation video is an unknown video different from the learning video. The estimation video capturing method is the same as the learning video capturing method in the learning phase described above. The input estimation video is stored in the estimation video storage unit 72.

  The action recognition unit 20 receives the estimation video stored in the estimation video storage unit 72, and detects facial expressions, gaze, head gestures, presence / absence of speech, etc. as the actions of each person captured in the estimation video Then, the time series B of the behavior of the dialogue person generated as a result is output (step S22). Since the behavior recognition method is the same as the behavior recognition method in the learning phase described above, description thereof is omitted here.

The estimation action time series B output from the action recognition unit 20 is input to the posterior probability estimation unit 50. The posterior probability estimation unit 50 uses the model parameters stored in the model parameter storage unit 74 to determine the posterior probability distribution P (e _t | B) of the sympathy interpretation between the conversationers from the estimation action time series B to the time t. Is estimated (step S50). The posterior probability estimation unit 50 includes a time series B of the conversation person's behavior generated based on the estimation video, a model parameter including each parameter of the prior distribution, timing model, and static model learned by the parameter learning unit 40. As an input, the posterior probability distribution P (e _t | B) of the empathy interpretation e at time t is calculated according to the above equation (1).

The output unit 60 outputs the posterior probability distribution P (e _t | B) of the sympathy interpretation e between the interlocutors (step S60). When it is necessary to output the estimation result of empathy interpretation as one type instead of probability distribution, the type of empathy interpretation with the highest posterior probability, that is, e ~ _t = argmax _{e_t} P (e _t | B) The values e to _t may be output together.

  Further, in addition to the type of sympathy interpretation having the highest posterior probability, the strength of the sympathy interpretation having the highest posterior probability may be output. For example, in the case of empathy, the strength can be obtained from the equation strength = sympathy probability−anti-probability, and in the case of anti-sensation, the strength = probability of probability−sympathy probability. In this case, the intensity is represented by a numerical value between -1 and 1.

701 Individual index calculation unit 702 Individual classification unit 703 Display unit 704 Group index calculation unit 705 Group classification unit

Claims (7)

A dialog state estimation device for obtaining an index P (e _t ^{(i, j)} = e) representing a probability that the empathy state between the person i and the person j at time t is e,
_{C 1, C 2, C 3} ', C 4', the X _2, X ₃ and predetermined constant, and the a _2, a ₃ and a predetermined real _{constant, w e (e = 1,2,} ..., N _e ), w _e '(e = 1,2, ..., N _e ) is a predetermined constant, N is the total number of persons, and T is a predetermined time of the data of the sympathetic state to be considered for calculating each index ChoToshi, as the total number of empathy possible states of N _e,
Let M _i be the personal bias for a person i defined by

Let the individual clarity for a person i be A _i defined by the following formula:

Let the individual person selectivity for a person i be S _i defined by the following formula:

The individual activity level for a person i is defined as E _i defined as

Using the obtained index P (e _t ^{(i, j)} = e), there are two indices among the individual bias, the individual clarity, the individual person selectivity and the individual activity. A personal index calculation unit for calculating a person,
A personal classifying unit for classifying the person into a predetermined type based on the calculated two indices for the person;
A display unit for displaying the result of classification by the personal classification unit in a table centered on two indicators for the person;
Interactive trend scoring device including A dialog state estimation device for obtaining an index P (e _t ^{(i, j)} = e) representing a probability that the empathy state between the person i and the person j at time t is e,
_{C 1, C 2, C 3} ', C 4', the X _2, X ₃ and predetermined constant, and the a _2, a ₃ and a predetermined real _{constant, w e (e = 1,2,} ..., N _e ), w _e '(e = 1,2, ..., N _e ) is a predetermined constant, N is the total number of persons, and T is a predetermined time of the data of the sympathetic state to be considered for calculating each index ChoToshi, as the total number of empathy possible states of N _e,
Let M _i be the personal bias for a person i defined by

Let the individual clarity for a person i be A _i defined by the following formula:

Let the individual person selectivity for a person i be S _i defined by the following formula:

The individual activity level for a person i is defined as E _i defined as

Using the obtained index P (e _t ^{(i, j)} = e), there are two indices among the individual bias, the individual clarity, the individual person selectivity and the individual activity. A personal index calculation unit for calculating a person,
C ₅ , C ₆ , C ₇ , C ₈ , X ₇ are given constants,
Let the group bias for a group be M defined by

The group clarity for a group is defined as A defined by the following formula:

The group person selectivity for a group is defined as S defined by the following formula:

The group activity for a group is defined as E defined by the following formula:

Based on the two indicators calculated by the individual indicator calculator, each of the two indicators of the group bias, the group clarity, the group person selectivity, and the group activity is calculated for a group. A group index calculation unit,
A group classifying unit for classifying the group into a predetermined type based on the two indexes for the group calculated;
A display unit for displaying a classification result by the group classification unit in a table centered on two indicators for the group;
Interactive trend scoring device including In the dialog tendency scoring device of claim 2,
A personal classification unit for classifying the person into a predetermined type based on the two indexes for the person calculated by the personal index calculation unit;
The display unit further displays the classification result by the personal classification unit in a table centered on two indicators for the person.
Interactive trend scoring device. A dialogue state estimation step in which the dialogue state estimation device obtains an index P (e _t ^{(i, j)} = e) representing the probability that the empathy state between the person i and the person j at the time t is e;
_{C 1, C 2, C 3} ', C 4', the X _2, X ₃ and predetermined constant, and the a _2, a ₃ and a predetermined real _{constant, w e (e = 1,2,} ..., N _e ), w _e '(e = 1,2, ..., N _e ) is a predetermined constant, N is the total number of persons, and T is a predetermined time of the data of the sympathetic state to be considered for calculating each index ChoToshi, as the total number of empathy possible states of N _e,
Let M _i be the personal bias for a person i defined by

Let the individual clarity for a person i be A _i defined by the following formula:

Let the individual person selectivity for a person i be S _i defined by the following formula:

The individual activity level for a person i is defined as E _i defined as

Using the index P (e _t ^{(i, j)} = e) obtained by the individual index calculation unit, two of the individual bias, the individual clarity, the individual person selectivity, and the individual activity A personal indicator calculation step for calculating each of the indicators for a person,
A personal classifying step in which the personal classifying unit classifies the person into a predetermined type based on the two indexes for the calculated person;
A display step in which the display unit displays the classification result by the personal classification unit in a table centered on two indicators for the person;
Dialogue trend scoring method including A dialogue state estimation step in which the dialogue state estimation device obtains an index P (e _t ^{(i, j)} = e) representing the probability that the empathy state between the person i and the person j at the time t is e;
_{C 1, C 2, C 3} ', C 4', the X _2, X ₃ and predetermined constant, and the a _2, a ₃ and a predetermined real _{constant, w e (e = 1,2,} ..., N _e ), w _e '(e = 1,2, ..., N _e ) is a predetermined constant, N is the total number of persons, and T is a predetermined time of the data of the sympathetic state to be considered for calculating each index ChoToshi, as the total number of empathy possible states of N _e,
Let M _i be the personal bias for a person i defined by

Let the individual clarity for a person i be A _i defined by the following formula:

Let the individual person selectivity for a person i be S _i defined by the following formula:

The individual activity level for a person i is defined as E _i defined as

Using the index P (e _t ^{(i, j)} = e) obtained by the individual index calculation unit, two of the individual bias, the individual clarity, the individual person selectivity, and the individual activity A personal indicator calculation step for calculating each of the indicators for a person,
C ₅ , C ₆ , C ₇ , C ₈ , X ₇ are given constants,
Let the group bias for a group be M defined by

The group clarity for a group is defined as A defined by the following formula:

The group person selectivity for a group is defined as S defined by the following formula:

The group activity for a group is defined as E defined by the following formula:

Based on the two indices calculated in the individual index calculation step, the group index calculation unit calculates two indices among the group bias, the group clarity, the group person selectivity, and the group activity. A group index calculation step for calculating each for a group,
A group classification step in which a group classification unit classifies the group into a predetermined type based on the two indexes for the group calculated;
A display step in which the display unit displays the classification result by the group classification unit in a table centered on two indicators for the group;
Dialogue trend scoring method including In the dialog tendency scoring method of claim 5,
The personal classification unit further includes a personal classification step of classifying the person into a predetermined type based on the two indices for the person calculated in the personal index calculation step.
The display unit further displays the classification result of the individual classification step in a table centered on two indicators for the person;
Dialogue trend scoring method.   A program for causing a computer to function as each part of the dialogue tendency scoring device according to claim 1.

Images