KR102648226B1 - Method for providing specific content to predict and reduce other persons' negative emotions and computing device using the same - Google Patents

Abstract

In the method of providing specific content to predict and reduce the negative emotions of the other person, (a) a computing device, as a preset condition for the specific content, (i-1) at least one of the user's appearances in the specific content. Determining specific object data for at least one specific object as an object for interaction with one other party and (i-2) specific state data for the specific object, and (ii-1) as a variable setting condition for the specific content Specific expression vector data corresponding to the other person's facial expression - The specific facial expression vector data has as components a first to nth facial expression component for representing each of the first to nth counterpart emotion categories - and ( ii-2) Specific behavior vector data corresponding to the other party's behavior - The specific behavior vector data includes a first to nth behavior component for representing each of the first to the nth counterpart's emotion category. - Determines at what timing of the specific content to expose at least some of the components, the specific object data and the specific state data as the preset conditions, the specific facial expression vector data as the variable setting conditions, and the specific providing the specific content with reference to at least some of the behavior vector data; (b) When the specific expression vector data and the specific action vector data are provided, the computing device causes the user to select a preset of the first to nth expression components included in the specific expression vector data. A process for predicting the presence or absence of an i-th other person's emotion category corresponding to at least one ith facial expression component corresponding to an expression threshold or higher, and the first to n-th behavior components included in the specific behavior vector data. performing a process to support predicting the presence or absence of a jth opponent's emotion category corresponding to at least one jth behavior component corresponding to a preset behavior threshold or higher; and (c) in a state in which the presence or absence of the i-th counterpart emotion category corresponding to the i-th facial expression component and the presence or absence of the j-th counterpart emotion category corresponding to the j behavior component are predicted by the user, For adjusting the size of the emotional level of at least some of the i-th emotional level of the other party corresponding to the i-th other-party emotion category and the j-th emotional level of the other party corresponding to the j-th counterpart emotional category by the user. When a specific user response action is input, the computing device generates an i-th partner adjusted emotion level, which is an emotional level adjusted to correspond to each of the i-th partner emotion level and the j-th partner emotion level with reference to the specific user response behavior. and exposing the jth counterpart's adjusted emotional level. A method and a computing device using the same are disclosed.

Description

A method of providing specific content to predict and reduce the other person's negative emotions and a computing device using the same {METHOD FOR PROVIDING SPECIFIC CONTENT TO PREDICT AND REDUCE OTHER PERSONS' NEGATIVE EMOTIONS AND COMPUTING DEVICE USING THE SAME}

The present invention relates to a method of providing specific content to predict and reduce the other person's negative emotions, and a computing device using the same.

Social cognitive ability is a part of cognitive ability related to social interaction and refers to the ability of an individual to interact with and understand other people in a social environment. Social cognitive abilities include conversation, communication, emotional recognition, situational awareness, problem solving, and cooperation, and require the ability to understand and respond appropriately to social rules, behavioral patterns, and cultural differences.

However, as the average lifespan increases, the number of elderly people with problems not only in physical ability but also in cognitive function is increasing, and problems that elderly people with poor social cognitive ability may experience include social isolation and communication difficulties. At this time, social isolation can cause loneliness and depression due to a decrease in social connectivity, and communication difficulties have the problem of limiting social activities due to difficulties in conversation, expressing emotions, and understanding situations.

Therefore, an improvement plan to solve the above problems is required.

The present invention aims to solve all of the above-mentioned problems.

In addition, the present invention refers to at least some of specific object data for a specific object as a preset condition, specific state data, specific facial expression vector data for the other person as a variable setting condition, and specific action vector data, and provides a specific data to the user. The purpose is to provide content.

In addition, the present invention allows a computing device to cause a user to use an i-th facial expression component corresponding to at least one i-th facial expression component corresponding to a preset expression threshold or higher among the first to n-th facial expression components included in specific facial expression vector data. The jth corresponding to at least one jth behavior component corresponding to a preset behavior threshold or higher among the first to nth behavior components included in the process and specific behavior vector data that support predicting the existence of the other person's emotional category. Another purpose is to perform a process that supports predicting the existence of the other person's emotion category.

In addition, the present invention provides that, when a specific user response action is input by a user, the computing device provides a first emotional level adjusted to correspond to each of the ith opponent's emotional level and the jth opponent's emotional level with reference to the specific user responsive action. Another purpose is to expose the i counterpart's adjusted emotion level and the j's counterpart's adjusted emotion level.

In order to achieve the purpose of the present invention as described above and realize the characteristic effects of the present invention described later, the characteristic configuration of the present invention is as follows.

According to one aspect of the present invention, in the method of providing specific content to predict and reduce the negative emotions of the other person, (a) a computing device sets the specific content as a preset condition for (i-1) the user. determines specific object data for at least one specific object as an object for interaction with at least one other party appearing in the specific content and (i-2) specific state data for the specific object, and determines the variable of the specific content As a setting condition, (ii-1) specific expression vector data corresponding to the other person's facial expression - the specific facial expression vector data is a first to nth facial expression component for representing each of the first to nth counterpart's emotion categories. Having a component as an element - and (ii-2) specific behavior vector data corresponding to the other party's behavior - the specific behavior vector data is a first for representing each of the first counterpart's emotion category to the nth counterpart's emotion category. Having a behavior component to an n-th behavior component as components - determining at what timing of the specific content to expose at least some of the specific object data and the specific state data as the preset conditions, and the variable setting conditions as the specific object data and the specific state data providing the specific content with reference to at least some of the specific facial expression vector data and the specific behavior vector data; (b) When the specific expression vector data and the specific action vector data are provided, the computing device causes the user to select a preset of the first to nth expression components included in the specific expression vector data. A process for predicting the presence or absence of an i-th other person's emotion category corresponding to at least one ith facial expression component corresponding to an expression threshold or higher, and the first to n-th behavior components included in the specific behavior vector data. performing a process to support predicting the presence or absence of a jth opponent's emotion category corresponding to at least one jth behavior component corresponding to a preset behavior threshold or higher; and (c) in a state in which the presence or absence of the i-th counterpart emotion category corresponding to the i-th facial expression component and the presence or absence of the j-th counterpart emotion category corresponding to the j behavior component are predicted by the user, For adjusting the size of the emotional level of at least some of the i-th emotional level of the other party corresponding to the i-th other-party emotion category and the j-th emotional level of the other party corresponding to the j-th counterpart emotional category by the user. When a specific user response action is input, the computing device generates an i-th partner adjusted emotion level, which is an emotional level adjusted to correspond to each of the i-th partner emotion level and the j-th partner emotion level with reference to the specific user response behavior. and exposing the jth counterpart's adjusted emotional level.

As an example, (d) the computing device, the ith counterpart adjusted emotion level, which is an emotional level adjusted to correspond to each of the ith counterpart emotion level and the jth counterpart emotion level with reference to the specific user response behavior, and In a state in which the jth counterpart's adjusted emotion level is exposed, user emotion vector data due to the specific user response action - the user emotion vector data is the first user emotion level for each of the first user emotion category to the nth user emotion category Indicates to the nth user emotion level - when it is detected that the mth user emotion level for the mth user emotion category, which is at least part of the mth user emotion level, has changed more than a preset change threshold, the mth user emotion category and the corresponding change information are recorded. A method is disclosed, characterized in that it further comprises the step:

As an example, when it is detected that the mth user emotion category has changed by more than the preset change threshold, the computing device selects an emotion category corresponding to the other party's positive emotion among the first counterpart emotion category and the nth counterpart emotion category. A process of increasing the size of the expression level of the a-th facial expression component for representing the a-th emotional category, which is at least part of the n Increase the size of the action level of the b action component for representing the b other person emotion category, which is at least some of the emotion categories corresponding to the other person's positive emotions, by the second size, which is the size of the preset action vector. Compensation content for lowering the mth user emotion level for the mth user emotion category that has changed more than the preset change threshold of the user to less than the preset change threshold by performing at least some of the processes exposed to the user. A method characterized by additionally being provided is disclosed.

As an example, the computing device may include: (i) information about the user and information about at least one other user having a similarity to the user greater than a preset threshold, (ii) the specific facial expression vector data and the other user corresponding thereto. other facial expression vector data that was exposed to the user, (iii) the specific action vector data and other action vector data corresponding thereto that was exposed to the other user, and (iv) the specific user corresponding action, the other action vector data that was input by the other user. At least some of the user response behavior and the corresponding specific user correct response behavior and the other user's correct response response behavior are input to the learning device to cause the learning device to: A method characterized by learning at least some of optimal specific expression vector data, optimal other expression vector data, optimal specific action vector data, and optimal other action vector data to increase the accuracy of at least some of the corresponding actions. This begins.

As an example, in step (a), the computing device, as the variable setting condition of the specific content, (ii-3) specific language vector data corresponding to the other party's utterance - the specific language vector data is the first Having as components a first language component to an n-th language component for representing each of the other person's emotion categories and the n-th other person's emotional category - additionally determines at which timing of the specific content to be exposed, and the specific content as the variable setting condition The specific content is provided by additionally referring to language vector data, and in step (b), when the specific language vector data is additionally provided, the computing device causes the user to include the specific content included in the specific language vector data. Additional processes are performed to support predicting the presence or absence of a k-th counterpart emotion category corresponding to at least one k-th language component corresponding to a preset language threshold or more among the first language component to the n-th language component, wherein the ( In step c), in a state where the presence or absence of the k-th counterpart emotion category corresponding to the k-th language component is additionally predicted by the user, the presence or absence of the k-th counterpart emotion category corresponding to the k-th language component is additionally predicted by the user. When the specific user response action for additionally adjusting the size of the k opponent's emotional level is input, the computing device generates a kth emotional level adjusted to correspond to the kth opponent's emotional level with reference to the specific user response action. A method characterized by additionally exposing the other party's adjusted emotional level is disclosed.

As an example, (e) with reference to the specific user response behavior, the i-th opponent's adjusted emotion level and the j-th opponent's adjustment, which are emotional levels adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level. When it is confirmed that the emotion level satisfies the ith opponent's stable emotion level and the jth opponent's stable emotion level, respectively, which are preset stable levels, the computing device determines the specific object data and the specific object data as the preset conditions of the specific content. By performing at least some of the processes of adjusting state data and adjusting the specific expression vector data and the specific action vector data as the variable setting conditions and the ith opponent's emotion level and the jth opponent's emotion level, A method is disclosed that further includes the step of providing coordinated content that adjusts specific content.

As an example, with reference to the specific user response behavior, the i-th opponent's adjusted emotion level and the j-th opponent's adjusted emotion level are emotional levels adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level. If it is confirmed that the ith counterpart's stable emotion level and the jth counterpart's stable emotion level, which are respectively preset stability levels, are satisfied, the computing device provides the user with the correct response behavior for the specific user response behavior as a plurality of solutions. A method is disclosed, wherein the adjusted content is provided by adjusting at least some of the preset conditions and the variable setting conditions to exist.

As an example, a method is disclosed wherein the computing device selects and displays the i-th counterpart emotion category and the j-th counterpart emotion category from emotion categories representing negative emotions of the other party.

Additionally, according to another aspect of the present invention, a computing device that provides specific content for predicting and reducing negative emotions of another person, comprising: at least one memory storing instructions; and at least one processor configured to execute the instructions, wherein the processor determines (I) as a preset condition for the specific content, (i-1) the user with at least one other party appearing in the specific content; Determining specific object data for at least one specific object as an object for interaction and (i-2) specific state data for the specific object, and (ii-1) the other party's facial expression as a variable setting condition for the specific content specific expression vector data corresponding to - the specific expression vector data has as components a first to nth expression component for representing each of the first to nth counterpart emotion categories - and (ii-2) Specific behavior vector data corresponding to the other party's behavior - The specific behavior vector data has as components a first behavior component to an nth behavior component for representing each of the first to nth counterpart emotion categories - determining at what timing of the specific content to expose at least some of the specific content, the specific object data and the specific state data as the preset conditions, the specific facial expression vector data as the variable setting conditions, and the specific action vector data. A process of providing the specific content, at least in part with reference to the content; (II) When the specific expression vector data and the specific action vector data are provided, the user is asked to make one of the first to nth expression components included in the specific expression vector data correspond to a preset expression threshold or higher. a process for predicting the presence or absence of the ith other person's emotion category corresponding to at least one ith facial expression component, and a preset behavior threshold among the first to nth behavior components included in the specific behavior vector data. A process for performing a process that supports predicting the presence or absence of the jth opponent's emotion category corresponding to at least one jth behavior component corresponding to the above; and (III) in a state in which the presence or absence of the i-th counterpart emotion category corresponding to the i-th facial expression component and the presence or absence of the j-th counterpart emotion category corresponding to the j behavior component are predicted by the user, For adjusting the size of the emotional level of at least some of the i-th emotional level of the other party corresponding to the i-th other-party emotion category and the j-th emotional level of the other party corresponding to the j-th counterpart emotional category by the user. When a specific user response action is input, the i-th opponent adjustment emotion level and the j-th opponent adjustment are emotional levels adjusted to correspond to each of the i-th opponent emotion level and the j-th opponent emotion level with reference to the specific user response action. A computing device is disclosed, characterized in that it performs a process of exposing an emotional level.

As an example, the processor may: (IV) set the i-th opponent's adjusted emotion level, which is an emotion level adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level, with reference to the specific user response behavior; In a state where the jth opponent's adjusted emotion level is exposed, the user emotion vector data is generated due to the specific user response action - the user emotion vector data is the first user emotion level for each of the first user emotion category to the nth user emotion category. Indicates the nth user emotion level - a process of recording the mth user emotion category and corresponding change information when the mth user emotion level for at least a part of the mth user emotion category is detected to have changed by more than a preset change threshold. A computing device is disclosed, characterized in that it further performs.

As an example, when the processor detects that the mth user emotion category has changed by more than the preset change threshold, one of the emotion categories corresponding to the other party's positive emotion among the first counterpart emotion category to the nth counterpart emotion category A process of increasing the size of the expression level of the a-th facial expression component for representing at least a part of the a-th emotional category by a first size, which is the size of a preset facial expression vector, and exposing it to the user, and the first through the n-th emotional categories. The user increases the size of the action level of the b action component for representing the b other person's emotion category, which is at least part of the emotion category corresponding to the other person's positive emotion among the other person's emotion categories, by the second size, which is the size of the preset action vector. Additional compensation content for lowering the mth user's emotion level for the mth user's emotion category that has changed more than the preset variation threshold of the user to less than the preset variation threshold by performing at least some of the processes of exposing to the user A computing device characterized in that it performs a process provided is disclosed.

As an example, the processor may be configured to: (i) information about the user and information about at least one other user having a similarity level of more than a preset threshold to the user, (ii) the specific facial expression vector data and the other user corresponding thereto. other facial expression vector data that was exposed to the other user, (iii) the specific action vector data and the corresponding other action vector data that was exposed to the other user, and (iv) the specific user's corresponding action, the other user that was entered by the other user. At least some of the corresponding behavior and the corresponding specific user correct response behavior and the other user's correct response behavior are input to the learning device to cause the learning device to: Characterized by performing a process of learning at least some of optimal specific expression vector data, optimal other expression vector data, optimal specific action vector data, and optimal other action vector data in order to increase the accuracy of at least some of the actions. A computing device is disclosed.

As an example, in the process (I), the processor, as the variable setting condition of the specific content, (ii-3) specific language vector data corresponding to the utterance of the other party - the specific language vector data is provided to the first counterpart Having as components a first language component to an n-th language component for representing each of the emotional categories and the n-th emotional category of the other party - additionally determining at which timing of the specific content to be exposed, and the specific language as the variable setting condition The specific content is provided by additionally referring to vector data, and in the process (II), when the specific language vector data is additionally provided, the user is provided with the first language component included in the specific language vector data. A process is additionally performed to support predicting the presence or absence of a k-th opponent's emotion category corresponding to at least one k-th language component corresponding to a preset language threshold or more among the n-th language components, and in the process (III), In a state in which the existence of the k-th counterpart emotion category corresponding to the k-th language component is additionally predicted by the user, the size of the k-th counterpart emotion level of the other party corresponding to the k-th counterpart emotion category by the user When the specific user response action for additional adjustment is input, a process of additionally exposing the kth opponent's adjusted emotion level, which is an emotional level adjusted in correspondence to the kth opponent's emotion level, is performed with reference to the specific user response action. Disclosed is a computing device characterized in that:

As an example, the processor may: (V) adjust the i-th opponent's adjusted emotion level, which is an emotion level adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level with reference to the specific user response behavior; When it is confirmed that the jth counterpart's adjusted emotion level satisfies the ith counterpart's stable emotion level and the jth counterpart's stable emotion level, which are preset stable levels, respectively, the specific object data and the specific state as the preset conditions of the specific content By performing at least some of the process of adjusting data and the process of adjusting the specific expression vector data and the specific action vector data as the variable setting conditions and the ith opponent's emotion level and the jth opponent's emotion level, A computing device is disclosed, characterized in that it further performs a process of providing coordinated content and providing coordinated content.

As an example, the processor adjusts the i-th opponent's adjusted emotion level and the j-th opponent, which are emotional levels adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level with reference to the specific user response behavior. If it is confirmed that the adjusted emotion level satisfies the i-th counterpart's stable emotion level and the j-th counterpart's stable emotion level, which are respectively preset stable levels, the correct response behavior for the specific user response behavior exists as a plurality of solutions for the user. A computing device is disclosed, characterized in that it performs a process of providing the adjusted content by adjusting at least some of the preset conditions and the variable setting conditions so as to do so.

As an example, a computing device is disclosed, wherein the processor performs a process of selecting and exposing the i-th counterpart emotion category and the j-th counterpart emotion category from emotion categories for indicating negative emotions of the other party. .

The present invention refers to at least some of specific object data for a specific object as a preset condition, specific state data, specific facial expression vector data for the other person as a variable setting condition, and specific action vector data, and provides specific content to the user. It has an effect.

In addition, the present invention allows a computing device to cause a user to use an i-th facial expression component corresponding to at least one i-th facial expression component corresponding to a preset expression threshold or higher among the first to n-th facial expression components included in specific facial expression vector data. The jth corresponding to at least one jth behavior component corresponding to a preset behavior threshold or higher among the first to nth behavior components included in the process and specific behavior vector data that support predicting the existence of the other person's emotional category. It has the effect of performing a process that supports predicting the existence of the other person's emotional category.

In addition, the present invention provides that, when a specific user response action is input by a user, the computing device provides a first emotion level adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level with reference to the specific user response action. It has the effect of exposing the i other party's adjusted emotional level and the j other party's adjusted emotional level.

The following drawings attached for use in explaining embodiments of the present invention are only some of the embodiments of the present invention, and to those skilled in the art (hereinafter “those skilled in the art”), the invention Other drawings can be obtained based on these drawings without further work being done.
Figure 1 schematically illustrates a computing device that provides specific content for evaluating a user's social cognitive ability, according to the present invention.
Figure 2 schematically shows a process for providing specific content for evaluating a user's social cognitive ability, according to the first embodiment of the present invention.
Figure 3 schematically shows a database in which predicted values obtained from the user and preset correct values are recorded in correspondence with the setting conditions of specific content provided to the user, according to the first embodiment of the present invention.
Figure 4 schematically shows a process for providing specific content to predict and reduce the other person's negative emotions, according to a second embodiment of the present invention.
5 shows a specific user response behavior, an emotional state toward the other party that changes in response to a specific user response behavior, and feelings toward the user, with reference to the setting conditions of specific content provided to the user, according to a second embodiment of the present invention. This is a schematic illustration of the database that records the status.
Figure 6 schematically shows a process for providing specific content to reduce negative emotions of a user, according to a third embodiment of the present invention.
7 shows a specific user action, a specific opponent's action corresponding to a specific user action, and a user that changes according to the specific opponent's corresponding action with reference to the setting conditions of specific content provided to the user, according to the third embodiment of the present invention. This is a schematic illustration of a database that records changes in emotional states.
Figures 8a and 8b schematically show some of the first content, which is content for predicting and reducing the other person's negative emotions among specific content, according to the present invention.
9A and 9B schematically show some of the second content, which is content for reducing the user's negative emotions, among specific content, according to the present invention.

The detailed description of the present invention described below refers to the accompanying drawings, which show by way of example specific embodiments in which the present invention may be practiced to make clear the objectives, technical solutions and advantages of the present invention. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention.

Additionally, throughout the description and claims, the word “comprise” and variations thereof are not intended to exclude other technical features, attachments, components or steps. Other objects, advantages and features of the invention will appear to those skilled in the art, partly from this description and partly from practice of the invention. The examples and drawings below are provided by way of example and are not intended to limit the invention.

Moreover, the present invention encompasses all possible combinations of the embodiments shown herein. It should be understood that the various embodiments of the invention are different from one another but are not necessarily mutually exclusive. For example, specific shapes, structures and characteristics described herein with respect to one embodiment may be implemented in other embodiments without departing from the spirit and scope of the invention. Additionally, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the invention. Accordingly, the detailed description that follows is not intended to be taken in a limiting sense, and the scope of the invention is limited only by the appended claims, together with all equivalents to what those claims assert, if properly described. Similar reference numbers in the drawings refer to identical or similar functions across various aspects.

Hereinafter, in order to enable those skilled in the art to easily practice the present invention, preferred embodiments of the present invention will be described in detail with reference to the attached drawings.

Figure 1 schematically illustrates a computing device that provides specific content for evaluating a user's social cognitive ability, according to the present invention.

Referring to FIG. 1, a computing device 100 is shown for providing specific content for the first to third embodiments of the present invention, and the computing device 100 includes a memory 101 and a processor 102. may include. At this time, the computing device 100 may additionally include a display 103, but it may also be configured separately rather than included in the computing device 100.

The memory 101 of the computing device 100 that provides specific content for evaluating the user's social cognitive ability may store instructions to be performed by the processor 102. Specifically, the instructions are computing instructions for providing content. Code generated for the purpose of causing device 100 to function in a particular manner may be stored in computer-usable or computer-readable memory capable of being directed to a computer or other programmable data processing equipment. Instructions may perform processes to execute functions described in the specification of the present invention.

In the present invention, content will be explained assuming content implemented in virtual reality, but it is of course not limited thereto.

In addition, the processor 102 of the computing device 100 for providing content has a hardware configuration such as a Micro Processing Unit (MPU) or Central Processing Unit (CPU), cache memory, and data bus. It can be included. Additionally, the computing device 100 may further include an operating system and a software component of an application that performs a specific purpose.

Additionally, the computing device 100 to provide specific content may be linked to the database 110. Here, the database 110 includes a flash memory type, hard disk type, multimedia card micro type, card type memory (for example, SD or XD memory), At least one of RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (ReadOnly Memory), EEPROM (Electrically Erasable Programmable ReadOnly Memory), PROM (Programmable ReadOnly Memory), magnetic memory, magnetic disk, and optical disk. It may include one type of storage medium, but is not limited to this and may include all media capable of storing data. In addition, the database 110 may be installed separately from the computing device 100 for providing content, or alternatively, may be installed inside the computing device 100 for providing content to transmit data or record received data. Alternatively, as shown, it may be implemented separately into two or more parts, which may vary depending on the implementation conditions of the invention.

Additionally, the display 103 may be separate from the memory 101 and processor 102 in the computing device 100 as a separate device.

Additionally, the computing device 100 may be a user terminal such as an HMD, or may be a server for providing a virtual reality service. In the latter case, the server may be linked to a user terminal (not shown) such as a separate HMD.

Figure 2 schematically shows a process for providing specific content for evaluating a user's social cognitive ability, according to the first embodiment of the present invention.

First, the first embodiment of providing the specific contents of FIG. 2 will be described focusing on the first content, which is one of the specific contents. At this time, the second content, which is one of the specific contents, may also be applied to the first embodiment, but since the second content is described in detail in the third embodiment, the description will be omitted.

Referring to FIG. 2, first, in order to evaluate the user's social cognitive ability, the computing device 100 determines (i-1) at least one user 820 that appears in the specific content as a preset condition for the specific content. Determine specific object data for at least one specific object as an object for interaction with the other party and (i-2) specific state data for the specific object, and (ii-1) determine the specific state data for the specific object as a variable setting condition for the specific content, and (ii-1) determine the facial expression of the other party. Corresponding specific expression vector data (specific expression vector data has as components a first expression component to an nth expression component for representing each of the first to nth emotion categories) and (ii-2) the other person's behavior At least part of the corresponding specific behavior vector data (the specific behavior vector data has as components a first behavior component to an n-th behavior component for representing each of the first emotion category to the n-th emotion category) is stored in some part of the specific content. Determine whether to expose at the timing, and provide specific content (S201) with reference to at least some of specific object data and specific state data as preset conditions, specific expression vector data as variable setting conditions, and specific action vector data. You can.

At this time, referring to FIG. 8A, in the first content to predict and reduce the negative emotions of the other person 810 through the mission of giving up a seat to the other person 810, who is a couple, on a bench in the park, the user 820 It represents a situation where you see and hear the facial expressions, actions, and language of the other person in front of you. The user 820 can determine what the other party wants by looking at the conditions set within the specific content and the facial expression, behavior, and language of the other party 810 and perform appropriate response actions. Additionally, the user 820 in FIGS. 8A and 8B may refer to a specific user in the first embodiment. Meanwhile, in FIGS. 8A and 8B, it is assumed that the user 820 is visible from a third-person perspective, but in order to make it more realistic, the user 820 is not visible at all or only a part of the user's body is visible from a first-person perspective. may be visible, and this may also be applied to FIGS. 9A and 9B, which will be described later.

Here, the preset condition is determined by referring to information about the user, such as the user's age, gender, degree of social cognitive decline, and previous history of using specific content, before providing specific content to the user. It may mean specific object data for at least one specific object as an object for interaction with at least one other party and specific state data for the specific object.

Here, it is assumed that the other party is not an actual person as an object of interaction with the user, but a virtual person appearing in specific content, but is not limited to this.

For example, in the first embodiment, the specific object refers to a three-person bench on which a user is sitting within specific content, the specific object data refers to the number of benches that are a specific object, and the specific state data refers to the occupancy rate of the bench that is a specific object. However, examples of preset conditions are not limited to this, and various modifications may be possible.

In addition, the variable setting condition is a condition that can be adjusted with reference to the user's reaction while providing specific content to the user, and is specific expression vector data corresponding to the other person's facial expression and specific action vector data corresponding to the other person's behavior. You can.

At this time, the specific facial expression vector data has first to n-th facial expression components as components to represent each of the first to n-th emotional categories, and in the first embodiment, the first to n-th emotional categories are For convenience of explanation, I will explain the basic human emotions of joy, anger, sadness, joy, love, hate, and greed as emotional categories. For example, the first emotional category can be explained by assuming that the first emotional category is an emotion that represents joy, the second emotional category is an emotion that represents anger, and the seventh emotional category is an emotion that represents greed. In addition, specific action vector data may be able to correspond to the first to seventh emotional categories with seven basic emotions, similar to specific facial expression vectors. In addition, the first to seventh facial expression components for representing each of the first to seventh emotional categories may be the other person's facial expressions exposed to the user to represent the emotions of each emotional category, and the first emotion The first to seventh action components for representing each of the categories to the seventh emotion category may be the other party's actions exposed to the user to represent the emotions of each emotion category. Here, for the first embodiment, emotional categories for joy, anger, sorrow, joy, love, hate, and greed are illustrative. Although it has been explained as follows, the second and third embodiments that follow will also be explained assuming the same.

Meanwhile, referring to FIG. 3, specific object data and specific state data for a specific object as preset conditions, specific expression vector data for the other person as variable set conditions, and specific action vector data are stored in the database 110. You can see that it has been done. Here, for convenience of explanation, the values stored in the database 110 are assumed to store information about a specific user, User 1 (hereinafter referred to as a specific user), but the data stored in the database 110 is limited to this. Of course, it will be possible to store data about many people.

In addition, it can be seen that the specific object data and specific state data stored in the database 110 are object type: bench for 3 & number of objects: 1, seat occupancy: 33% & occupied location: center seat. Of course, it can also be assumed that the number of benches is greater, and in this case, the number of options a specific user can select increases, so the difficulty of specific content may increase.

In addition, the values of the first to seventh expression components for representing each of the first to seventh emotion categories, which are components of specific expression vector data that are variable setting conditions, are (0.1, 0.3, 0.6, 0.2, 0.2, It can be seen that it is 0.4, 0.4). In addition, the values of the first to seventh action components to represent each of the first to seventh emotion categories, which are components of specific action vector data, which is another variable setting condition, are (0.2, 0.3, 0.5, 0.1, It can be seen that it is 0.1, 0.7, 0.5). In other words, it can be said that data on the intensity with which the facial expressions and actions of the other person appearing in the first content provided to a specific user will be exposed is recorded in vector form.

Returning to the main topic, the computing device 100 determines at which timing of specific content to expose at least some of specific object data and specific state data as preset conditions, specific facial expression vector data as variable setting conditions, and specific action vector data. You can. At this time, at least some of the preset conditions and variable setting conditions may be exposed at any time when the user interacts with the user.

Next, referring again to FIG. 2, when specific expression vector data and specific action vector data are provided, the computing device 100 causes the user to select the first to nth expression components included in the specific expression vector data. A process that supports predicting the presence or absence of the ith emotion category corresponding to at least one ith facial expression component corresponding to a preset first threshold or more, and the first to nth action components included in the specific action vector data. A process that supports predicting the presence or absence of the jth emotion category corresponding to at least one jth behavior component that is greater than or equal to a preset second threshold may be performed (S202).

Here, i and j may be specified as the same value and exposed as facial expressions and actions representing the same emotion, or may be specified as different values and may be exposed in combination with facial expressions and actions representing different emotions, and the ith facial expression component and The jth behavioral component may mean not one component but multiple components.

Referring again to FIG. 3, the computing device 100 sets the third facial expression component representing the emotion of sadness among the first to seventh facial expression components included in the specific facial expression vector data to a preset first threshold or more to determine a specific facial expression component. It can be provided to the user. Here, the first threshold may mean a preset setting value for allowing a specific user to recognize at least some of the first to seventh facial expression components. That is, when the first threshold is set to 0.5, the values of the first to seventh expression components exposed to a specific user are (0.1, 0.3, 0.6, 0.1, 0.2, 0.4, 0.4), so the values are higher than the first threshold. It can be seen that the component with the value is the third facial expression component. This represents a situation where the couple wants to sit together on a bench where there is only one other person, but the sadness of not being able to sit side by side because the specific user is already sitting in the middle is being expressed to a specific user through facial expressions.

In addition, the computing device 100 may set the sixth behavior component representing the emotion of hatred among the first to seventh behavior components included in the specific behavior vector data to a preset second threshold or more and provide it to a specific user. there is. Here, the second threshold may mean a preset setting value for allowing a specific user to recognize at least some of the first to seventh behavior components. That is, when the second threshold is set to 0.6, the values of the first to seventh behavior components exposed to a specific user are (0.2, 0.4, 0.4, 0.1, 0.1, 0.7, 0.5), so the value is higher than the second threshold. It can be seen that the component with the value is the sixth facial expression component. This represents a situation in which the couple wants to sit together on a bench where the other person is the only one, but cannot sit side by side because the specific user is already sitting in the middle seat, and the feelings of hatred are being exposed to a specific user through actions.

In this case, the computing device 100 processes a specific user to predict the presence or absence of a third emotion category corresponding to a third facial expression component and determines the presence or absence of a sixth emotion category corresponding to a sixth behavior component. You will be able to carry out a process that supports prediction.

Next, referring again to FIG. 2, the i-th facial expression-derived emotion prediction value for the presence or absence of the i-th emotion category corresponding to the i-th facial expression component and the j-th emotion prediction value for the presence or absence of the j-th emotion category corresponding to the j-th behavioral component When the j action-derived emotion prediction value is input, the computing device 100 provides information about the i-th facial expression-derived emotion prediction value and the j-th action-derived emotion prediction value as result data on the user's social cognitive ability, respectively, and the corresponding i-th emotion prediction value. It can be recorded and managed (S203) by matching the correct answer value of the emotion derived from the facial expression and the correct answer value of the emotion derived from the jth action.

For example, referring to FIG. 3, a specific user inputs the third facial expression-derived emotion prediction value regarding the presence or absence of the third emotion category and the sixth action-derived emotion prediction value regarding the presence or absence of the sixth emotion category as YES OR NO, respectively. As a result (of course, it is omitted in FIG. 3, it may be possible to support entering YES OR NO regarding the presence of emotion categories other than the presence or absence of the third emotion category and the presence of the sixth emotion category), a specific user It can be seen that the third emotion prediction value derived from the facial expression and the sixth emotion prediction value derived from the behavior are each input as Y (YES). In addition, the computing device 100 provides information on the third expression-derived emotion prediction value and the sixth behavior-derived emotion prediction value as result data on the social cognitive ability of a specific user, respectively, and the third expression-derived emotion correct answer value and the sixth behavior-derived emotion prediction value, respectively. 6 It can be recorded and managed in the database 110 by matching it with the correct behavior-derived emotion value.

Meanwhile, in step S202, when specific expression vector data and specific action vector data are provided, the computing device 100 causes the specific user to select one of the first to nth facial expression components included in the specific facial expression vector data. A process for supporting predicting the expression level for the presence or absence of the ith emotion category corresponding to at least one ith facial expression component corresponding to a set first threshold or more as the i_1 expression level, and the first expression included in the specific action vector data A process that supports predicting the behavior level of the presence or absence of the jth emotion category corresponding to at least one jth behavior component that is higher than a preset second threshold among the behavior components to the nth behavior component as the j_1th behavior level. It can also be performed additionally.

For example, referring to Figure 3, the i_1th expression level and the j_1th action level may be prediction values that are the expression level and action level that are the same emotion category as the third expression-derived emotion prediction value and the sixth action-derived emotion prediction value, respectively, and may be prediction values that are the expression level and action level, respectively, In addition to entering YES OR NO as the 3rd expression-derived emotion prediction value and the 6th action-derived emotion prediction value, predict the level of the corresponding emotion category and additionally input 0.6 for the 3_1 expression level and 0.7 for the 6_1 action level, respectively. can do. Of course, it would be possible to support a specific user to select something like high/middle/low rather than having them enter specific numbers like this.

In addition, in step S203, the i_1th expression level regarding the presence or absence of the ith emotion category corresponding to the ith expression component and the j_1th action level regarding the presence or absence of the jth emotion category corresponding to the jth behavior component are input. Then, the computing device 100 provides information about the i_1th expression level and j_1th action level as result data about the social cognitive ability of a specific user, respectively, and the i_2th expression level and j_2th action level as the corresponding correct answer values. It can be additionally recorded and managed by matching with.

For example, referring to Figure 3, the i_2th expression level and the j_2th action level may be the correct value for the expression level and action level, which are the same emotion categories as the third expression-derived emotion correct value and the sixth behavior-derived emotion correct value, respectively. In addition to setting YES OR NO as the third facial expression-derived emotion correct answer value and the sixth behavior-derived emotion correct answer value, the computing device 100 sets 3_1, which is a predicted value that allows a specific user to predict the level of the corresponding emotion category, The 3_2 expression level and the 6_2 behavior level may be recorded to correspond to the expression level and the 6_1 behavior level, respectively.

Meanwhile, in step S201, the computing device 100 uses (ii-3) specific language vector data corresponding to the other party's utterance as a variable setting condition for specific content (specific language vector data includes the first emotion category to the nth emotion). It additionally determines at what timing of specific content to expose the first language component to the nth language component to represent each category, and provides the specific content by additionally referring to specific language vector data as a variable setting condition. can do. Here, the specific language vector data, like the specific facial expression vector data and specific action vector data mentioned above, may have language components representing emotional states of joy, anger, sadness, joy, sadness, anger, and swearing as components.

And, when specific language vector data is additionally provided, the computing device 100 causes the user to select at least one of the first to nth language components included in the specific language vector data that is equal to or higher than a preset third threshold. A process that supports predicting the presence or absence of the kth emotion category corresponding to the kth language component of can be additionally performed. For example, the computing device 100 allows the user to express at least one emotion that is higher than a preset third threshold among the seven language components: anger (k corresponds to anger, which is the second emotion among the joys, sorrows, sorrows, and desires. For example, It can be supported to additionally predict the existence of a second emotional category corresponding to a second language component expressing the emotional state (assumed).

Next, when the second language-derived emotion prediction value regarding the presence or absence of the second emotion category corresponding to the second language component is input, the computing device 100 generates the second language-derived emotion prediction value as result data about the user's social cognitive ability. Information about the predicted emotion value can be additionally recorded and managed by matching it with the corresponding emotion correct value derived from the second language.

In addition, when specific language vector data is additionally provided, the computing device 100 causes the user to select at least one of the first to nth language components included in the specific language vector data that is equal to or higher than a preset third threshold. A process to support predicting the language level for the presence or absence of the second emotion category corresponding to the second language component as the 2_1 language level may be additionally performed.

And, when the 2_1 language level regarding the presence or absence of the second emotion category corresponding to the second language component is input, the computing device 100 provides information about the 2_1 language level as result data about the user's social cognitive ability. can be additionally recorded and managed by matching it with the 2_2 language level as the corresponding correct answer value.

At this time, information about specific language vector data as variable setting conditions is omitted in FIG. 3, but this does not mean that the database 110 does not include information about specific language vector data.

And, when the i-th facial expression-derived emotion prediction value and the ith facial expression-derived emotion correct answer value for the presence or absence of the i-th emotion category corresponding to at least one ith facial expression component corresponding to the first threshold or higher are different, the computing device (100 ) increases the expression level for the presence or absence of the i-th emotion category corresponding to at least one i-th expression component corresponding to a preset first threshold or more among the first to n-th expression components by a first predetermined size. A process of providing first adjusted content in which specific content has been adjusted can be performed.

Here, the fact that the third facial expression-derived emotion prediction value for the presence or absence of the third emotion category corresponding to the third facial expression component corresponding to the first threshold (0.5) or more is different from the third facial expression-derived emotion correct value means that the specific user is 3 Entering the facial expression-derived emotion prediction value as N(NO) may mean that the third emotion category was not recognized. Then, the computing device 100 increases the expression level corresponding to the third expression component, which is a component above the preset first threshold among the components of the specific expression vector data, that is, the value of the expression indicating a sad emotion, from 0.6 to 0.7 to provide the specific user You can expose it again.

In contrast, by adjusting the expression level of other facial expression components indicating negative emotions, such as the facial expression level corresponding to the second facial expression component indicating an angry emotion, or the facial expression level corresponding to the sixth facial expression component indicating a hateful emotion, to a specific user. It may be possible to predict the value of the expression level corresponding to the third expression component, but it is not limited to this.

Next, if the jth behavior-derived emotion prediction value and the jth behavior-derived emotion correct answer value for the presence or absence of the jth emotion category corresponding to at least one jth behavior component corresponding to the second threshold are different, the computing device , A state in which the action level for the presence or absence of the jth emotion category corresponding to at least one jth action component that is higher than a preset second threshold among the first to nth action components is increased by a second predetermined amount. A process of providing second adjusted content by adjusting specific content can be performed.

Here, the fact that the sixth behavior-derived emotion predicted value for the presence or absence of the sixth emotion category corresponding to the sixth behavior component corresponding to the second threshold (0.6) or higher is different from the sixth behavior-derived emotion correct value means that the specific user 6 The expression-derived emotion prediction value was entered as N(NO), which may mean that the sixth emotion category was not recognized. Then, the computing device 100 increases the action level corresponding to the sixth action component, which is a component above the preset second threshold among the components of the specific action vector data, that is, the value of the action indicating the emotion of hate from 0.7 to 0.8, and increases the action level corresponding to the sixth action component, which is a component above the preset second threshold, from 0.7 to 0.8. You can expose it again.

In contrast, the behavior level of other behavior components indicating negative emotions, such as the behavior level corresponding to the second behavior component indicating angry emotions or the behavior level corresponding to the third behavior component indicating sad emotions, is adjusted to a certain size to a specific user. It may be possible to predict the value of the action level corresponding to the sixth action component, but it is not limited to this.

In addition, it is detected that the i_1th expression level as the predicted value and the i_2th expression level as the corresponding correct answer value match, and the j_1th behavior level as the predicted value and the j_2th behavioral level as the corresponding correct answer value are detected to match. When it is detected that the u-th user-corresponding action, which is one of the first user-corresponding action to the m-th user-corresponding action, has been selected by a specific user with reference to the i_1-th facial expression level and the j_1-th behavioral level, the computing device 100 First, the u-th user response action, which is the result data on the social cognitive ability of a specific user, can be additionally recorded and managed by matching it with the i_1th facial expression level and j_1th behavior level.

In other words, when it is detected that the 3_1 expression level, which is the value predicted by a specific user, and the 3_2 expression level, which is the correct answer, match, and the 6_1 behavior level, which is the value predicted by a specific user, and the 6_2 behavior level, which is the correct answer value, match, , a specific user can perform a u-th user response action, which is a specific response action, to the other person with reference to the 3_1st expression level and 6_1th action level selected by the user. Here, the specific response action may be an action performed by a specific user regarding which action to select from the first user response action to the mth user response action to reduce the other person's negative emotions.

At this time, as examples of the first to mth user response actions, 1. Move the seat one space to the side. 2. Invite the man to sit on his right. 3. Pretend not to know. 4. Comfort the woman sincerely. etc., where a specific user may select option 1 as the first user response action as an appropriate response action. Then, the computing device 100 may additionally record and manage the first user response action as result data on the social cognitive ability of a specific user by matching it with the 3_1 facial expression level and the 6_1 behavioral level.

Meanwhile, in the following, the computing device 100 includes a learning device (not shown), and an example of a learning process using the learning device will be described.

First, the computing device 100 exposes (i) information about the user and information about at least one other user having a similarity to the user greater than a preset threshold, (ii) specific facial expression vector data and corresponding information to other users. other facial expression vector data, (iii) specific behavior vector data and corresponding other behavior vector data exposed to other users, and (iv) i-th facial expression regarding the presence or absence of the i-th emotion category corresponding to the i-th facial expression component. Emotion predicted value derived from the jth action, emotion predicted value derived from the jth action regarding the presence or absence of the jth emotion category corresponding to the jth action component, emotion predicted value derived from another facial expression entered by another user, emotion predicted value derived from another action and the ith corresponding thereto. At least some of the emotion correct value derived from the facial expression, the emotion correct value derived from the jth action, the emotion correct value derived from another facial expression, and the emotion correct value derived from the other behavior can be input to the learning device. At this time, (i) information about the user may include, but is not limited to, age, gender, mbti, heart rate, respiratory rate, etc., and information about other users includes information about the user among those who have previously used specific content. It may refer to a person who has a similarity between the information and the predicted value entered by the user above a preset threshold. In this way, in (ii), (iii), and (iv), vector data about the facial expression or behavior exposed to the user, vector data about the facial expression or behavior exposed to other users, or input by the user or other users Predicted values for the other person's expression and behavior, and corresponding correct answer values for the other person's expression and behavior, etc. can be input into the learning device.

The computing device 100 inputs such data into the learning device, and causes the learning device to calculate the emotion prediction value derived from the ith facial expression of the user and other users, the emotion prediction value derived from the jth behavior, the emotion prediction value derived from other facial expressions, and other behaviors. In order to increase the accuracy of at least some of the derived emotion prediction values, at least some of optimal specific expression vector data, optimal other facial expression vector data, optimal specific behavior vector data, and optimal other behavior vector data may be learned.

Here, the computing device 100 causes the learning device to learn at least some of the optimal specific expression vector data, optimal other facial expression vector data, optimal specific behavior vector data, and optimal other behavior vector data, so that the user and other users can interact with each other. You can increase the probability of understanding the other person's emotions by looking at their facial expressions or actions. Here, the concept of increasing the probability of identifying emotions may literally mean that the probability of identifying the other person's emotions is high, but it also includes the level of improvement in the ability to understand emotions through watching this content compared to before. You may.

In addition, the computing device 100, similarly, includes (i) information about the user and information about at least one other user having a similarity to the user greater than a preset threshold, (ii) specific expression vector data and other users corresponding thereto. Other facial expression vector data that has been exposed to other users, (iii) specific behavior vector data and corresponding other behavior vector data that has been exposed to other users, and (iv) information on the existence of the i-th emotion category corresponding to the i-th facial expression component. i expression-derived emotion prediction value, j-th action-derived emotion prediction value regarding the presence or absence of the j-th emotion category corresponding to the j-th action component, other facial expression-derived emotion prediction value entered by another user, other behavior-derived emotion prediction value and its corresponding At least some of the emotion correct value derived from the ith facial expression, the emotion correct value derived from the j action, the emotion correct value derived from another facial expression, and the emotion correct value derived from another behavior can be input to the learning device. At this time, (i) information about the user may include, but is not limited to, age, gender, mbti, heart rate, respiratory rate, etc., and information about other users includes information about the user among those who have previously used specific content. It may refer to a person who has a similarity between the information and the predicted value entered by the user above a preset threshold. In this way, in (ii), (iii), and (iv), vector data about the facial expression or behavior exposed to the user, vector data about the facial expression or behavior exposed to other users, or input by the user or other users Predicted values for the other person's expression and behavior, and corresponding correct answer values for the other person's expression and behavior, etc. can be input into the learning device.

The computing device 100 inputs such data into the learning device, and causes the learning device to calculate the emotion prediction value derived from the ith facial expression of the user and other users, the emotion prediction value derived from the jth behavior, the emotion prediction value derived from other facial expressions, and other behaviors. A first optimal value of the expression level for the presence or absence of the ith emotion category corresponding to the ith expression component, and the existence of the jth emotion category corresponding to the jth behavior component, to increase the accuracy of at least some of the derived emotion prediction values. It is possible to learn at least some of the second optimal values of the action level.

Here, the first optimal value and the second optimal value can be learned to enable a specific user to better understand the other person's emotions, which is the optimal value for the level of components of a specific emotional category in facial expressions and actions that express the other person's emotions. It can mean value.

Then, the computing device 100 causes the learning device to determine the first optimal value of the expression level for the presence or absence of the ith emotion category corresponding to the ith expression component and the jth emotion category corresponding to the jth behavior component. By learning at least some of the second optimal values of the action level for presence or absence, the emotion predicted value derived from the ith facial expression of the user and other users, the emotion predicted value derived from the jth behavior, the emotion predicted value derived from another facial expression, and the emotion predicted value derived from the other behavior. It will be possible to increase the accuracy of at least some of them.

At this time, referring to FIG. 8B, finally, the third expression-derived emotion prediction value for the presence or absence of the third emotion category corresponding to the third expression component of the other party 810, which is the prediction value input by the user 820, the other party ( 810), the sixth behavior-derived emotion prediction value for the presence or absence of the sixth emotion category corresponding to the sixth behavior component is the correct answer value for the presence or absence of the third emotion category, respectively, the third facial expression-derived emotion correct value, and the sixth emotion. It was confirmed to be consistent with the 6th action-derived emotion correct value, which is the correct value for the existence of the category, and the 3_1 facial expression level, which is the predicted value of the facial expression level for the presence of the 3rd emotional category, and the 6th emotional category. The 6_1 behavior level, which is the predicted value of the behavior level for It is confirmed that the levels match, and finally, the user 820 completes the mission of the specific content and gives up his or her seat to the other person 810, who is a couple, and sits together.

Figure 4 schematically shows a process for providing specific content to predict and reduce the other person's negative emotions, according to a second embodiment of the present invention.

At this time, the second embodiment is based on providing content where the correct answer to which the user must respond (select), for example, when the user is sitting on a bench, is to move away from the seat in order to reduce the other person's negative emotions. It will be explained assuming that, but it is not limited to this.

Referring to FIG. 4, first, in order for the user to predict and reduce the other person's negative emotions, the computing device 100 sets (i-1) at least one item in which the user appears in the specific content as a preset condition for the specific content. Determining specific object data for at least one specific object as an object for interaction with the other party and (i-2) specific state data for the specific object, and (ii-1) the other party's facial expression as a variable setting condition for the specific content specific expression vector data corresponding to - the specific expression vector data has as components a first to nth expression component for representing each of the first to nth counterpart emotion categories - and (ii-2) the other party specific behavior vector data corresponding to the behavior - the specific behavior vector data has as components a first behavior component to an nth behavior component for representing each of the first partner's emotion category to the nth partner's emotion category - specifying at least some of the Determine at what timing the content will be exposed, and provide specific content with reference to at least some of the specific object data and specific state data as preset conditions, specific expression vector data as variable setting conditions, and specific action vector data ( S401) You can.

Again, referring to FIG. 8A, in the first content to predict and reduce the negative emotions of the other person 810 through a mission to give up a seat to the other person 810, a couple, on a bench in the park, the user 820 It represents a situation where you see the facial expressions and actions of the other person in front of you. The user 820 can determine what the other party wants by looking at the conditions set within the specific content and the facial expression and behavior of the other party 810 and perform appropriate response actions. Additionally, the user 820 in FIGS. 8A and 8B may refer to a specific user in the second embodiment.

Here, the preset conditions are determined by referring to information about a specific user, such as the user's age, gender, degree of social cognitive decline, and previous history of using specific content, before providing specific content to a specific user. It may mean specific object data for at least one specific object as an object for interaction with at least one other party that appears and specific state data for the specific object.

Here, it is assumed that the other party is not an actual person as an object of interaction with a specific user, but a virtual person appearing in specific content, but is not limited to this.

For example, in the second embodiment, the specific object refers to a three-person bench on which a specific user is sitting within specific content, the specific object data refers to the number of benches that are a specific object, and the specific state data refers to the occupancy rate of the bench that is a specific object. However, examples of preset conditions are not limited to this and various modifications may be possible.

And, the variable setting condition is a condition that can be adjusted with reference to a specific user's reaction while providing specific content to a specific user, and includes specific expression vector data corresponding to the other person's facial expression and specific action vector corresponding to the other person's behavior. It could be data.

At this time, the specific facial expression vector data has as components a first facial expression component to an n-th facial expression component for representing each of the first to n-th emotional categories, and in the second embodiment, the first to n-th emotional categories. For convenience of explanation, the other person's emotion category is an emotion category for the basic human emotions of joy, anger, sadness, joy, love, hate, and greed. Let me explain. For example, it can be explained by assuming that the first emotion category is an emotion representing joy, the second emotion category is an emotion representing anger, and the seventh emotion category is an emotion representing greed. In addition, the specific action vector data may be able to correspond to the first through seventh emotional categories of the other person to seven basic emotions, similar to the specific facial expression vector data. In addition, the first to seventh facial expression components for representing each of the first to seventh partner emotion categories may be the other person's facial expressions exposed to the user to represent the emotions of each of the other partner's emotional categories, The first to seventh behavior components for representing the first to seventh counterpart emotion categories, respectively, may be the other party's actions exposed to a specific user to express the emotions of each other party's emotion category.

Meanwhile, referring to FIG. 5, specific object data and specific state data for a specific object as preset conditions, specific facial expression vector data for the other person as variable setting conditions, and specific action vector data are stored in the database 110. You can see that it has been done. Here, for convenience of explanation, the values stored in the database 110 are assumed to store information about a specific user, User 1 (hereinafter referred to as a specific user), but the data stored in the database 110 is limited to this. Of course, it will be possible to store data about many people.

In addition, it can be seen that the specific object data and specific state data stored in the database 110 are object type: bench for 3 people & number of objects: 1, seat occupancy rate: 33%, and occupied position: center seat. Of course, it can also be assumed that the number of benches is greater, and in this case, the number of options a specific user can select increases, so the difficulty of specific content may increase.

In addition, the values of the first to seventh expression components for representing each of the first to seventh opponent emotion categories, which are components of the specific expression vector data that is a variable setting condition, are (0.1, 0.3, 0.6, 0.2, It can be seen that it is 0.2, 0.4, 0.4). In addition, the values of the first to seventh action components to represent each of the first to seventh opponent emotion categories, which are components of specific action vector data, which is another variable setting condition, are (0.2, 0.3, 0.5, It can be seen that it is 0.1, 0.1, 0.7, 0.5). In other words, it can be said that data on the intensity with which the facial expressions and actions of the other person appearing in the first content provided to a specific user will be exposed is recorded in vector form.

Returning to the main topic, the computing device 100 determines at which timing of specific content to expose at least some of specific object data and specific state data as preset conditions, specific facial expression vector data as variable setting conditions, and specific action vector data. You can. At this time, at least some of the preset conditions and variable setting conditions may be exposed at any time when interacting with a specific user.

Next, referring again to FIG. 4, when specific expression vector data and specific action vector data are provided, the computing device 100 causes the user to select the first to nth expression components included in the specific expression vector data. A process that supports predicting the presence or absence of the ith other person's emotion category corresponding to at least one ith facial expression component corresponding to a preset facial expression threshold or more, and the first to nth behavioral components included in the specific behavior vector data. A process that supports predicting the presence or absence of the jth other party's emotion category corresponding to at least one jth behavior component that is greater than or equal to a preset behavior threshold may be performed (S402).

Here, i and j may be specified as the same value and exposed as facial expressions and actions representing the same emotion, or may be specified as different values and may be exposed in combination with facial expressions and actions representing different emotions, and the ith facial expression component and The jth behavioral component may mean not one component but multiple components. Additionally, the computing device 100 may select and display the i-th counterpart's emotion category and the j-th counterpart's emotion category from the emotion categories for indicating the other party's negative emotions.

Again, referring to FIG. 5, the computing device 100 sets the third facial expression component representing the emotion of sadness among the first to seventh facial expression components included in the specific facial expression vector data to a specific expression threshold or higher. It can be provided to the user. Here, the facial expression threshold may mean a preset setting value for allowing a specific user to recognize at least some specific facial expression components among the first to seventh facial expression components. That is, when the expression threshold is set to 0.5, the values of the first to seventh expression components exposed to a specific user are (0.1, 0.3, 0.6, 0.1, 0.2, 0.4, 0.4), so the values above the expression threshold are It can be seen that the branch component is the third facial expression component. This represents a situation where the couple wants to sit together on a bench where there is only one other person, but the sadness of not being able to sit side by side because the specific user is already sitting in the middle is being expressed to a specific user through facial expressions.

In addition, the computing device 100 may set the sixth behavior component indicating an emotion of hatred among the first to seventh behavior components included in the specific behavior vector data to a value greater than or equal to a preset behavior threshold and provide the sixth behavior component to a specific user. . Here, the behavior threshold may mean a preset setting value for allowing a specific user to recognize at least some specific behavior components among the first to seventh behavior components. In other words, when the behavior threshold is set to 0.6, the values of the first to seventh behavior components exposed to a specific user are (0.2, 0.4, 0.4, 0.1, 0.1, 0.7, 0.5), so the values are above the second threshold. It can be seen that the component having is the sixth facial expression component. This represents a situation in which the couple wants to sit together on a bench where the other person is the only one, but cannot sit side by side because the specific user is already sitting in the middle seat, and the feelings of hatred are being exposed to a specific user through actions. Here, the expressions and actions exposed to a specific user are not limited to this, although the levels of the third facial expression component and the sixth behavioral component are set to values above the threshold, and are not limited to facial expressions corresponding to the emotion category that can express the other person's negative emotions. If it is a component, it can be exposed by setting the level of the facial expression component above the threshold, and similarly, if it is a behavioral component that corresponds to an emotional category that can express the other person's negative emotions, it can be exposed by setting the level of the behavioral component in question above the threshold. You can do it.

Next, the computing device 100 performs a process for supporting a specific user to predict the presence or absence of a third counterpart emotion category corresponding to the third facial expression component and the presence of a sixth counterpart emotion category corresponding to the sixth behavioral component. You will be able to perform a process that supports predicting whether or not.

Next, referring again to FIG. 4, in a state in which the presence or absence of the ith other person's emotion category corresponding to the ith facial expression component and the existence of the jth other person's emotional category corresponding to the jth behavioral component are predicted by the user, A specific user action for adjusting the magnitude of the emotional level of at least some of the i-th emotional level of the other party corresponding to the i-th counterpart emotional category and the j-th counterpart emotional level of the other party corresponding to the j-th counterpart emotional category by the user. When this is input, the computing device 100 adjusts the i-th opponent's adjustment emotion level and the j-th opponent's adjustment, which are emotional levels adjusted to correspond to the i-th opponent's emotion level and the j-th opponent's emotion level, respectively, with reference to the specific user response behavior. The emotional level can be exposed (S403).

For example, referring to Figure 5, a specific user enters the predicted value for the presence or absence of the third partner's emotion category and the predicted value for the presence or absence of the sixth partner's emotion category as YES OR NO (of course, omitted in Figure 5). However, it may be possible to support entering YES OR NO regarding the existence of other party emotion categories other than the existence of the third party emotion category and the sixth party emotion category), and a specific user may select the third party emotion category. It can be seen that the predicted value for the presence or absence of the other party's emotion category and the predicted value for the presence or absence of the sixth opponent's emotion category were each entered as Y (YES).

Next, in a state where the presence or absence of a third counterpart emotion category corresponding to the third facial expression component and the presence or absence of a sixth counterpart emotion category corresponding to the sixth behavior component are predicted by a specific user, the third counterpart emotion category is predicted by a specific user. A specific user response action may be input to adjust the size of at least some of the third emotional level of the other party corresponding to the other emotional category and the sixth emotional level of the other party corresponding to the sixth emotional category. .

For example, when a specific user recognizes the other person's feelings of sadness (the third emotion category) or hatred (the sixth emotion category) that the other person is expressing through facial expressions or actions, they try to reduce the other person's feelings of sadness or hatred. For this purpose, 'Move one space to the side and sit down for the other person' can be entered as a specific user response action. Then, the computing device 100 may change each of the third party emotion level and the sixth party emotion level into the third party adjustment emotion level and the sixth party adjustment emotion level with reference to the specific user response behavior and expose them to the user. there is.

In other words, when a specific user enters an action to move one space to the side, the other person (couple) can sit side by side, so sadness exposed through the other person's facial expression (third partner emotion category) or hate exposed through the action ( The emotional level for the sixth partner's emotional category) can be reduced, and the reduced emotional levels, the third partner's adjusted emotional level and the sixth partner's adjusted emotional level, can be exposed to the user.

At this time, referring to FIG. 5, when the third party's emotion level and the sixth party's emotion level are set to 0.6 and 0.7, respectively, and are exposed to the user, due to a specific user response action input by a specific user, the third party's emotion The third opponent's adjusted emotion level and the sixth opponent's adjusted emotion level, in which the level and the sixth opponent's emotion level are reduced to 0.3 and 0.4, respectively, may be stored in the database 110.

Next, with reference to specific user response actions, the third party's adjusted emotion level and the sixth party's adjusted emotion level, which are emotional levels adjusted to correspond to each of the third party's emotion level and the sixth party's emotion level, are exposed to the user. In, at least part of the user emotion vector data (the user emotion vector data represents the first user emotion level to the nth user emotion level for each of the first to nth user emotion categories) due to a specific user response action. When it is detected that the m-th user's emotion level for the m-user emotion category has changed by more than a preset change threshold, the computing device 100 may record the m-th user's emotion category and the change information corresponding thereto. At this time, the first user emotion category to the nth user emotion category may be the same emotion category as the first to nth opponent emotion category, but is not limited thereto.

For example, a specific user may input a specific user response behavior to reduce the other person's feelings of sadness or hate, and then the specific user response behavior may cause the specific user to show feelings of anger or sadness. At this time, the emotional state of anger or sadness of a specific user may correspond to the mth user emotion category, which is at least part of the user emotion vector data. For convenience of explanation, the mth user emotion category is referred to as the second user emotion category, which is the emotional state of anger. Let me explain.

Next, it may be detected that the second user emotion category has changed more than a preset change threshold due to a specific user response action.

At this time, referring to FIG. 5, let us assume that the user emotion vector data before a specific user response action as the emotional state for a specific user is (0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2). Then, a specific user response action is input, and the second user emotion level of the second user emotion category fluctuates above the preset fluctuation threshold (set to 0.5 in this example), so the user emotion vector data is (0.2, 0.7, 0.2, 0.2) , 0.2, 0.2, 0.2). In other words, the user succeeded in reducing the other person's feelings of sadness and hatred through specific user response actions, but the user's (specific user's) negative emotion (anger) component increased.

Returning to the main topic, when it is detected that the second user emotion category has changed by more than a preset fluctuation threshold, the computing device 100 selects one of the emotion categories corresponding to the other party's positive emotion among the first other party emotion category to the nth other party emotion category. A process of increasing the size of the expression level of the a-th expression component for representing at least a part of the a-th other emotion category by the first size, which is the size of a preset expression vector, and exposing it to a specific user, and the first other person's emotion category to the n-th other person's emotion The size of the action level of the b action component to represent the b other person's emotion category, which is at least part of the emotion category corresponding to the other person's positive emotions among the categories, is increased by the second size, which is the size of the preset action vector, and exposed to a specific user. By performing at least part of the process, compensation content for lowering the mth user's emotion level for the mth user's emotion category that has changed more than the preset change threshold of a specific user to less than the preset change threshold can be additionally provided.

In other words, in order to reduce the second user emotion level of the user's second user emotion category, the computing device sets the size of the expression level of the ath expression component representing the other person's positive emotion category to be larger and exposes it to a specific user, or The size of the behavior level of the behavior component representing the other person's positive emotion category can be set larger and exposed to a specific user. Here, the a-th facial expression component or the b-th action component may be components of an emotional category corresponding to joy, sorrow, and joy, which are positive components of joy, sorrow, sorrow, etc., but are not limited thereto. At this time, the specific user may experience feedback from the other user that further increases the positive component (e.g., positive facial expression or level of behavioral component) in response to the specific user response behavior, thereby reducing the second user's emotional level. .

Meanwhile, in step S401, the computing device 100 uses (ii-3) specific language vector data corresponding to the other party's utterance as a variable setting condition for specific content (the specific language vector data includes the first opponent emotion category to the first opponent emotion category). (having as components the first language component to the nth language component to represent each of the other person's emotional categories) is additionally determined at what timing of the specific content to be exposed, and the specific content is additionally referred to by referring to the specific language vector data as a variable setting condition. can be provided. Here, the specific language vector data, like the specific facial expression vector data and specific action vector data mentioned above, may have language components representing emotional states of joy, anger, sadness, joy, sadness, anger, and swearing as components.

And, when specific language vector data is additionally provided, the computing device 100 causes the user to select at least one language component that is higher than a preset language threshold among the first to nth language components included in the specific language vector data. A process that supports predicting the presence or absence of the kth other person's emotion category corresponding to the kth language component may be additionally performed. For example, the computing device 100 assumes, for example, that anger k, which is at least one emotion corresponding to a preset third threshold or higher among the seven language components, corresponds to anger, which is the second emotion among joy, anger, sadness, and greed, by allowing the user to do so. ) can be supported to additionally predict the existence of a second opponent's emotional category corresponding to the second language component expressing the emotional state of ).

Next, with the existence of the kth opponent's emotion category corresponding to the kth language component additionally predicted by the user, the size of the kth opponent's emotion level of the other person corresponding to the kth opponent's emotion category is additionally predicted by the user. When a specific user response behavior for adjustment is input, the computing device 100 may additionally expose the kth counterparty adjusted emotion level, which is an emotional level adjusted in response to the kth counterparty emotion level, with reference to the specific user response behavior. there is.

At this time, information about specific language vector data as variable setting conditions is omitted in FIG. 5 , but this does not mean that the database 110 does not include information about specific language vector data.

Next, with reference to a specific user response behavior, the i-th opponent's emotion level and the j-th opponent's emotion level, which are adjusted emotion levels corresponding to each of the i-th opponent's emotion level and the j-th opponent's emotion level, are respectively preset stable levels. If it is confirmed that the ith other party stable emotion level and the jth other party stable emotion level are satisfied, the computing device 100 performs a process of adjusting specific object data and specific state data as preset conditions for specific content and variable setting conditions. Adjusted content obtained by adjusting the specific content can be provided by performing at least part of the process of adjusting the specific expression vector data and the specific action vector data and the ith opponent's emotional level and the jth opponent's emotional level.

Here, adjusted content means that if a specific user succeeds in reducing the negative emotions of the other person through specific user response actions, the computing device 100 adjusts the setting conditions of the content to further increase the difficulty of the content and provide it to the specific user. can do. For example, if a specific user succeeds in predicting the existence of a third-party emotion category that expresses sadness in a specific content, in the adjusted content, the second-party emotion category that expresses anger and the third-party emotion category that expresses sadness, respectively. By setting the second and third facial expression components corresponding to above a preset facial expression threshold and exposing them together, it will be possible to enable a specific user to predict the presence or absence of a plurality of negative emotions in the other person's specific facial expression vector data. . Of course, in some cases, the second and third facial expression components are set below a preset facial expression threshold and exposed together to enable a specific user to predict the presence or absence of a plurality of negative emotions in the other person's specific facial expression vector data. It may be that the level of difficulty is high.

At this time, the computing device 100 may provide adjusted content by adjusting at least some of the preset conditions and variable setting conditions so that the correct response behavior for the specific user response behavior exists as a plurality of solutions to the specific user. Here, in specific content, there was only one action to respond to the correct answer, 'Move one space to the side for the other person and sit down', but in the adjusted content, by adjusting the preset conditions and variable setting conditions, the action to respond to the correct answer was set so that there were multiple actions. Content can be provided. For example, you can increase the difficulty of the content by increasing the number of benches or adjusting the occupancy rate of the bench. This allows you to not only 'move one space to the side and sit down for the other person', but also 'there is a seat on the other bench over there', etc. There may be actions to respond to the correct answer, but it is not limited to this.

Meanwhile, in the following, the computing device 100 includes a learning device (not shown), and an example of a learning process using the learning device will be described.

And, the computing device 100 exposes (i) information about the user and information about at least one other user having a similarity to the user greater than a preset threshold, (ii) specific expression vector data and corresponding information to other users. other facial expression vector data, (iii) specific action vector data and corresponding other action vector data exposed to other users, and (iv) specific user response actions, other user response actions entered by other users and corresponding actions thereto. At least some of the behavior corresponding to a specific user's correct answer and the behavior corresponding to the correct answer of another user can be input into the learning device.

At this time, (i) information about the user may include, but is not limited to, age, gender, mbti, heart rate, respiratory rate, etc., and information about other users includes information about the user among those who have previously used specific content. It may refer to a person who has a similarity between the information and the predicted value entered by the user above a preset threshold. In this way, in (ii), (iii), and (iv), vector data about the facial expressions or actions exposed to the user, vector data about the facial expressions or actions exposed to other users, to reduce the other person's negative emotions. Specific user response behavior, other user response behavior, corresponding specific user correct response behavior, and other user correct response response behavior inputted by the user or other users may be input to the learning device.

Then, the computing device 100 inputs such data into the learning device to enable the learning device to obtain an optimal specific facial expression vector to increase the accuracy of at least some of the specific user response behavior and other user response behavior of the user and other users. At least some of the data, optimal other facial expression vector data, optimal specific action vector data, and optimal other action vector data may be learned.

Here, the computing device 100 causes the learning device to learn at least some of the optimal specific expression vector data, optimal other facial expression vector data, optimal specific behavior vector data, and optimal other behavior vector data, so that the user and other users can interact with each other. You can increase the probability of understanding the other person's emotions by looking at their facial expressions or actions. Here, the concept of increasing the probability of identifying emotions may literally mean that the probability of identifying the other person's emotions is high, but it also includes the level of improvement in the ability to understand emotions through watching this content compared to before. You may.

At this time, referring to FIG. 8B, finally, the third opponent's emotion level corresponding to the third opponent's emotion category, which is the emotional state toward the other party due to the specific user response behavior of the user 820, and the third opponent's emotion level corresponding to the sixth opponent's emotion category The sixth partner's emotional level is adjusted to the third partner's adjusted emotional level and the sixth partner's adjusted emotional level, respectively, and finally, the user completes the mission of the specific content and gives up his/her seat to the couple's partner (810) to sit together. will be.

Figure 6 schematically shows a process for providing specific content to reduce negative emotions of a user, according to a third embodiment of the present invention.

At this time, the third embodiment of providing the specific content of FIG. 6 will be described focusing on the second content, which is one of the specific contents. At this time, the third embodiment will be assumed and explained as providing specific content in which multiple correct answers may exist in order to reduce the user's negative emotions. For example, when a user takes a specific action after first seeing the other person's actions or facial expressions, the emotions the user feels may vary depending on the person's personality. In addition, due to the user's specific actions, the other person's emotions or the other person's specific response behavior (refers to the other person's response behavior in response to the user's specific action) may also take various forms, which may determine the user's final emotional state. There may be multiple correct answers to make it stable, but it is not limited to this.

Here, the second content may be content that induces the user to feel anxious due to difficulty using a kiosk in a public place such as a cafe or restaurant, but to reduce the user's anxiety by performing appropriate actions.

Referring to FIG. 6, first, in a method of providing specific content to reduce a user's negative emotions, the computing device 100 sets the user to (i-1) the specific content as a preset condition for the specific content. Determining specific object data for at least one specific object as an object for interaction with at least one counterpart appearing and (i-2) specific state data for the specific object, and (ii-1) as a variable setting condition for specific content ) The specific opponent's initial expression vector data corresponding to the other party's expression (the specific opponent's initial expression vector data consists of the first opponent's expression component to the nth opponent's expression component to represent each of the first opponent's emotion category to the nth opponent's emotion category) ) and (ii-2) specific opponent's initial behavior vector data corresponding to the opponent's behavior (the specific opponent's initial action vector data is a first opponent's behavior component for representing each of the first opponent's emotion category to the nth opponent's emotion category) to nth opponent's behavior component as a component) determines at what timing of the specific content to be exposed, specific object data and specific state data as preset conditions, and specific opponent's initial expression vector data as variable setting conditions, And, specific content may be provided (S601) by referring to at least part of the specific counterpart's initial action vector data.

At this time, referring to FIG. 9A, it is assumed that the user 920 does not know how to use the kiosk in a public place such as a restaurant or cafe with many people. Meanwhile, the waiting line of people waiting to use the kiosk becomes longer, and the user (920) must perform the mission of controlling the user's own anxiety and politely asking questions about how to use the kiosk so as not to offend the other person (910) within the time limit. This shows part of the second content. Additionally, the user 920 in FIGS. 9A and 9B may refer to a specific user in the third embodiment.

Here, the preset condition is determined by referring to information about the user, such as the user's age, gender, degree of social cognitive decline, and previous history of using specific content, before providing specific content to the user. It may mean specific object data for at least one specific object as an object for interaction with at least one other party and specific state data for the specific object.

Here, it is assumed that the other party is not an actual person as an object of interaction with the user, but a virtual person appearing in specific content, but is not limited to this.

For example, in the third embodiment, a specific object refers to a kiosk used by a user in a public place such as a restaurant or cafe, specific object data refers to the number of kiosks that are a specific object, and specific state data refers to the use of a kiosk that is a specific object. It may mean a time limit, but examples of preset conditions are not limited to this and various variations are possible.

In addition, as an additional environment to create users' feelings of anxiety within specific content, an environment suitable for specific users can be exposed by adjusting the number of people in the place, the level of noise in the place, etc.

In addition, the variable setting condition is a condition that can be adjusted with reference to the user's reaction while providing specific content to the user, and includes the specific opponent's initial expression vector data corresponding to the other person's expression and the specific opponent's behavior corresponding to the other person's behavior. It may be initial action vector data.

At this time, the specific opponent's initial expression vector data has as components a first opponent's expression component to an n-th opponent's expression component to represent each of the first opponent's emotion category to the nth opponent's emotion category, and in the third embodiment, the first opponent's emotion is For convenience of explanation, the nth emotional category of the other person is divided into the basic human emotions of joy, anger, sadness, pleasure, love, hate, and greed. I will explain it as an emotional category. For example, it can be explained by assuming that the first emotion category is an emotion representing joy, the second emotion category is an emotion representing anger, and the seventh emotion category is an emotion representing greed. In addition, the specific opponent's first action vector data may be able to correspond to the first to seventh opponent emotion categories to seven basic emotions, similar to the specific opponent's first facial expression vector data. In addition, the first to seventh opponent's expression components for representing each of the first to seventh opponent's emotion categories may be the other's facial expressions exposed to the user to represent the emotions of each of the other's emotion categories. The first to seventh counterpart behavior components for representing each of the first to seventh counterpart emotion categories are the counterpart's behavior exposed to the user to represent the emotions of each counterpart's emotion category. You can.

At this time, the specific other person's first expression vector data and the specific other person's first action vector data may be the first expression and first action that reflect the emotional state of the other person for interaction while the user uses the kiosk, and the user may use the other person's first expression and first action. By watching the behavior and controlling your own anxiety, you will be able to ask the other person how to use the kiosk.

Meanwhile, referring to FIG. 7, specific object data and specific state data for a specific object as preset conditions, specific opponent's initial expression vector data and specific opponent's initial action vector data for the other party as variable setting conditions are stored in the database (110). ), you can see that it is stored in . Here, for convenience of explanation, the values stored in the database 110 are assumed to store information about a specific user, User 1 (hereinafter referred to as a specific user), but the data stored in the database 110 is limited to this. Of course, it will be possible to store data about multiple people.

In addition, it can be seen that the specific object data and specific status data stored in the database 110 are object type: kiosk & object number: 1, use time limit: 60 seconds, respectively.

In addition, the values of the first opponent's expression component to the seventh facial expression component to represent each of the first opponent's emotion category to the seventh opponent's emotion category, which are components of the specific opponent's initial expression vector data, which is a variable setting condition, are (0.7, 0.1, 0.3 , 0.5, 0.2, 0.1, 0.2). In addition, the values of the first to seventh opponent behavior components to represent each of the first to seventh opponent emotion categories, which are components of specific opponent behavior vector data, which is another variable setting condition, are (0.4, 0.2) , 0.1, 0.6, 0.2, 0.1, 0.2).

Returning to the main topic, the computing device 100 uses at least some of the specific object data and specific state data as preset conditions, the specific opponent's initial expression vector data as variable setting conditions, and the specific opponent's initial action vector data at a certain timing of specific content. You can decide whether to expose it to . At this time, at least some of the preset conditions and variable setting conditions may be exposed at any time when interacting with a specific user.

Next, referring again to FIG. 6, when the specific opponent's initial expression vector data and the specific opponent's initial action vector data are provided, the computing device 100 causes the user to select the first opponent included in the specific opponent's initial expression vector data. A process that supports predicting the presence or absence of the i-th opponent's emotion category corresponding to at least one i-th opponent's expression component that is higher than a preset expression threshold among the facial expression components and the n-th opponent's facial expression components and the specific opponent's initial action vector data Performing a process to support predicting the presence or absence of the jth opponent's emotion category corresponding to at least one jth opponent's behavior component that is greater than a preset behavior threshold among the included first to nth opponent's behavior components ( S602) You can.

Here, i and j may be specified as the same value and exposed as facial expressions and actions representing the same emotion, or may be specified as different values and may be exposed in combination with facial expressions and actions representing different emotions, and the ith opponent's facial expression component And the jth counterpart behavior component may mean not a single component but a plurality of components. In addition, the computing device 100 may select and display the i-th opponent's emotion category and the j-th opponent's emotion category from the emotion categories to indicate the other party's positive emotions, but in some cases, the other party's emotion category may be used to adjust the difficulty of specific content. You may be able to select and expose negative emotions from emotion categories.

Again, referring to FIG. 7, the computing device 100 sets the first opponent's expression component representing the emotion of joy among the first to seventh opponent's facial expression components included in the specific opponent's initial expression vector data to a preset facial expression threshold. You can set it above and provide it to specific users. Here, the facial expression threshold may mean a preset setting value for allowing a specific user to recognize at least some of the specific facial expression components of the first to seventh facial expression components. That is, when the facial expression threshold is set to 0.6, the values of the first to seventh facial expression components exposed to the user are (0.7, 0.1, 0.3, 0.5, 0.2, 0.1, 0.2), so they have values higher than the facial expression threshold. It can be seen that the component is the first opponent's facial expression component. This represents a situation where the other person is having a pleasant conversation with acquaintances, waiting to use the kiosk, and expressing feelings of joy to a specific user through facial expressions. At this time, the difficulty of specific content experienced by a specific user may be increased by increasing the negative components of the other person's emotions exposed to the specific user by a predetermined amount.

In addition, the computing device 100 sets the fourth behavior component representing the emotion of enjoyment among the first to seventh counterpart behavior components included in the specific counterpart's initial behavior vector data to more than a preset behavior threshold to provide the specific user with can be provided. Here, the behavior threshold may mean a preset setting value for allowing a specific user to recognize at least some specific counterpart behavior components among the first to seventh counterpart behavior components. In other words, when the behavior threshold is set to 0.5, the values of the first to seventh counterpart behavior components exposed to a specific user are (0.4, 0.2, 0.1, 0.6, 0.2, 0.1, 0.2), so the behavior threshold is higher than the behavior threshold. It can be seen that the component with the value is the fourth opponent's behavior component. This represents a situation where the other person is having a pleasant conversation with acquaintances, waiting to use the kiosk, and expressing feelings of enjoyment to a specific user through facial expressions. At this time, the difficulty of specific content experienced by a specific user may be increased by increasing the negative components of the other person's emotions exposed to the specific user by a predetermined amount.

Here, the expressions and actions exposed to a specific user are not limited to this, although the levels of the first opponent's expression component and the fourth opponent's behavior component are set to values above the threshold, and correspond to an emotion category that can express the other person's positive emotions. If it is the other person's facial expression component, it can be exposed by setting the level of the other person's facial expression component above the threshold, and similarly, if it is an opponent's behavior component that corresponds to an emotion category that can express the other person's positive emotions, the level of the other person's behavior component can be set. You may be able to expose it by setting it above the threshold.

Next, the computing device 100 performs a process for supporting a specific user to predict the presence or absence of a first counterpart emotion category corresponding to a first facial expression component and the presence of a fourth counterpart emotion category corresponding to a fourth behavioral component. You will be able to perform a process that supports predicting whether or not.

Next, referring again to FIG. 6, specific user expression vector data (specific user expression vector data is a first user expression component to represent each of the first user emotion category to the nth user emotion category) by the user. an a-th user expression component for representing the a-th user emotion category among the facial expression components), and specific user behavior vector data (the specific user behavior vector data represents each of the first to n-th user emotion categories) When a specific user action including at least a portion of the b user facial expression component for indicating the b user emotion category among the first user behavior component for bet to the n th user behavior component as a component is input, the computing device 100 ), with reference to a specific user action, a specific opponent's expression vector data having as a component a'th opponent's expression component corresponding to a'th user's expression component, and b'th opponent's action corresponding to b'th user's action component. A specific opponent response action that includes at least part of the specific opponent response action vector data that has components as components may be exposed (S603).

Referring again to FIG. 7, the specific user inputs the a-th user facial expression component to represent the a-th user emotion category as the first user facial expression component representing the emotion of joy among the components representing positive emotions, and the b-th user emotion category. The b user behavior component for representing can be input as the fourth user behavior component representing the emotion of enjoyment among the components representing positive emotions.

As an example, referring to Figure 7, although a specific user has anxious feelings, the level of the first user facial expression component (joy) is set somewhat high at 0.6, and the level of the fourth user behavior component (joy) is set somewhat high at 0.5. By setting it high and entering a specific user action, you will be able to ask the other person how to use the kiosk in a polite and positive manner.

Next, the computing device 100 refers to a specific user action, and generates specific opponent-corresponding expression vector data having as a component a first 'opponent expression component corresponding to the first user expression component, and a fourth user action component. A specific opponent response action that includes at least a portion of specific opponent response action vector data having the fourth 'opponent action component as an element may be exposed. Here, because a specific user asked a question to the other person while including a first user expression component and a fourth user behavior component that are positive emotions, the computing device 100 asks the other person a question corresponding to the first 'other person expression component and the fourth user behavior component. You will be able to provide answers to specific users while including components.

As an example, referring to FIG. 7, it can be seen that the levels of the first' opponent's facial expression component and the fourth' opponent's behavior component included in a specific opponent response action are 0.5 and 0.4, respectively, but the values may naturally vary. For example, in the third embodiment, the first 'counterpart expression component and the fourth 'opponent behavior component corresponding to the first user expression component (0.6) and the fourth user behavior component (0.5) are set to 0.5 and 0.4, respectively. Although it was exposed by setting it lower than the component and the fourth user behavior component, in order to provide feedback with a more positive answer to users who are more prone to anxiety, the first 'counterpart expression component' and the fourth 'counterpart behavior component' were combined with the first user expression component and the fourth 'opponent behavior component. It may be exposed by setting it higher than the fourth user behavior component.

At this time, the computing device 100 refers to at least some of the user biometric data, video data for the user, and voice data uttered by the specific user, which are acquired while the specific content is being provided, to determine the first user emotion category to the second user emotion category. At least some of the a-th user expression component for representing the a-th user emotion category among the n-th user emotion categories, and the b-th user behavior component for representing the b-th user emotion category among the first to n-th user emotion categories. can be obtained. Here, the user's biometric data may be electrocardiogram, heart rate, blood pressure, etc., the image data for the user may be data that captures the facial expressions, actions, etc. of the user using specific content, and the voice data uttered by a specific user may be a specific This may be data that detects the user's voice pitch, voice tremor, etc., but is not limited to this.

Meanwhile, in step S401, the computing device 100 uses (ii-3) the specific opponent's initial language vector data corresponding to the other party's utterance (the specific opponent's initial language vector data is the first opponent's first language vector data) as a variable setting condition for the specific content. It is additionally determined at what timing of the specific content to be exposed (having the first opponent's language component to the nth opponent's language component to represent each of the emotional category or the nth opponent's emotional category), and the specific opponent's first language as a variable setting condition. Specific content can be provided by additionally referencing vector data. Here, the specific opponent's first language vector data, like the specific opponent's first facial expression vector data and the specific opponent's first action vector data mentioned above, consists of the opponent's language components representing the emotional states of joy, anger, sadness, joy, love, oh, and swearing. You can have it.

And, when the specific counterpart's initial language vector data is additionally provided, the computing device 100 causes the user to use the first to the nth counterpart's language components included in the specific counterpart's language vector data above a preset language threshold. A process to support predicting the presence or absence of the kth counterpart's emotion category corresponding to at least one corresponding kth counterpart's language component may be additionally performed. For example, the computing device 100 causes the user to select at least one emotion, love, that corresponds to a preset language threshold or higher among the seven other language components (assuming, for example, that k corresponds to the fifth emotion, love, among the joys, sorrows, sorrows, and desires). It is possible to support additional prediction of the existence of the fifth partner's emotional category corresponding to the fifth partner's language component expressing the emotional state of the user.

Next, in a state in which the existence of the kth opponent's emotion category corresponding to the kth opponent's language component is additionally predicted by a specific user, the cth user's emotion among the first to nth user's emotion categories is determined by the specific user. When a specific user action including specific user language vector data having as a component a c-th user language component for representing a category is additionally input, the computing device 100 generates the c-th user facial expression component with reference to the specific user action. It is possible to additionally expose a specific opponent-corresponding action that includes specific opponent-corresponding language vector data that has the 'c' opponent facial expression component corresponding to as a component.

At this time, in FIG. 7, information on the specific counterpart's initial language vector data and the specific counterpart's corresponding language vector data as variable setting conditions are omitted, but this does not mean that the database 110 does not include information on this.

Next, with reference to the specific counterpart response action, the final anxiety level at which the m-th user emotion level for at least one m-th user emotion category, which is a negative emotion category among the first to n-th user emotion categories, is preset. If it is confirmed that it is the following, the computing device 100 adjusts the specific object data and specific state data as preset conditions of the specific content and the specific opponent's initial expression vector data and the specific opponent's initial action vector data as variable setting conditions. Adjusted content that adjusts specific content can be provided by performing at least some of the processes.

Here, the mth user emotion category may be an emotion category that represents the user's anxiety while using specific content, and for convenience of explanation, the mth user emotion category and the mth user emotion level are the third user emotion, which is the emotion of sadness among negative emotions. This will be explained assuming categories and third user emotion levels. In addition, the preset final anxiety level may refer to the level at which a specific user successfully reduces the emotional level of sadness by successfully asking a question with positive emotions, which is the final mission of specific content. In the present invention, the final anxiety level is I will explain by assuming that the figure is 0.2.

At this time, referring to FIG. 7, it can be seen that the third user emotion level of the user who performed the final mission decreased from 0.7 to 0.2.

Here, adjusted content means that if the user succeeds in reducing the negative emotions of a specific user through a specific user action, the computing device 100 adjusts the setting conditions of the content to further increase the difficulty of the content and provide it to the specific user. You can. For example, if it is successful in reducing the third user's emotion level of the third user's emotion category expressing sadness in a specific content, in the adjusted content, the first' opponent's expression component and the fourth' opponent's behavior component exposed to the specific user are reduced. You can reduce the size of the level by a certain amount and expose it again. In other words, the difficulty of the adjusted content may be increased by causing a specific user to experience some difficulty in reducing the user's own anxiety due to a specific counterpart's response action that includes the other's weakened positive emotional state.

In addition, the computing device 100 determines the m-th user emotion level for at least one m-th user emotion category among the first to n-th user emotion categories, with reference to a specific counterpart response action, to a preset final anxiety level. If it is confirmed that it is below the value, adjusted content may be provided by adjusting at least some of the preset conditions and variable setting conditions so that the correct action for the specific user action exists as multiple solutions for the specific user. Here, in specific content, there was only one correct action for a specific user action, 'politely ask the other person how to use the kiosk', but in adjusted content, there were multiple correct actions by adjusting the preset conditions and variable setting conditions. You can provide coordinated content by setting it to exist. For example, you can increase the difficulty of the content by increasing the number of kiosks or adjusting the usage time limit of the kiosk. In addition to 'asking the other person politely how to use the kiosk', 'Use it first, use it carefully'. There may be a correct answer action such as 'I will use it after seeing it', but it is not limited to this.

Next, referring to FIG. 9B, the user 920 inputs a specific user action with the level of the first user expression component and the fourth user action component raised, and the other party 910 inputs a specific user action in response to the specific user action. By exposing and responding to a specific opponent's action with the level of the 1' opponent's facial expression component and the 4th' opponent's behavior component raised, the emotional state toward the user 920 changes to a stable state, and finally, the specific content is provided to the opponent 910. This shows a situation where the mission of politely asking how to use the kiosk was successfully completed.

Next, the anxiety emotion severity of the user using specific content will be referred to as the kth anxiety emotion severity level, and the anxiety emotion severity level that is more severe than the kth anxiety emotion severity level will be referred to as the (k+1)th anxiety emotion severity level. At this time, the computing device 100, when the user corresponds to the kth anxiety emotion severity level, the first counterpart emotion level corresponding to the positive emotion category among the first to nth counterpart emotion categories exposed, The value of r1, which is the ratio of the other person's emotional level corresponding to the negative emotional category among the other person's emotional category or the nth other person's emotional category, is set to the first opponent exposed when the user falls under the (k+1)th anxiety emotion severity level. Greater than the value of r2, which is the ratio of the opponent's emotional level corresponding to the negative emotional category among the first and nth emotional categories of the emotional category to the emotional level of the positive emotional category among the emotional categories and the nth emotional category of the other emotional category. You can set it.

At this time, the anxiety emotion severity ranges from the first anxiety emotion severity level to the tenth anxiety emotion severity level, and as it increases from the first anxiety emotion severity level to the tenth anxiety emotion severity level, it means that it is difficult to control the anxiety emotion. You can. Therefore, when comparing the case where the anxiety emotion severity of a specific user is the third anxiety emotion severity level and the case where the anxiety emotion severity level is the fourth anxiety emotion severity level, the case of the third anxiety emotion severity level is more positive than the case of the fourth anxiety emotion severity level. The ratio of the other person's emotional level corresponding to the negative emotion category to the other person's emotional level corresponding to the emotional category may need to be set higher. Because, when the anxiety severity level is high, anxiety emotions are felt more easily, so the difficulty of specific content may need to be adjusted by exposing the positive components of the other person's emotions that appear in the specific content to a higher level to such specific users.

Meanwhile, in the following, the computing device 100 includes a learning device (not shown), and an example of a learning process using the learning device will be described.

Next, the computing device 100 includes (i) information about the user and information about at least one other user having a similarity to the user greater than a preset threshold, (ii) initial facial expression vector data of a specific opponent, and corresponding Other user's first facial expression vector data that was exposed to other users, (iii) specific user's first action vector data and corresponding other user's first action vector data that was exposed to other user, (iv) specific user's action and corresponding other user's action vector data At least some of the actions of other users that were input by the user may be input to the learning device.

At this time, information about the user may include, but is not limited to, age, gender, mbti, heart rate, respiratory rate, etc., and information about other users includes information about users among those who have previously used specific content. This may mean a person who has a similarity between the input predicted value and a preset threshold or higher.

Then, the computing device 100 inputs such data into the learning device to enable the learning device to determine the optimal initial facial expression for a specific opponent to increase the success rate of performing at least some of the specific user actions and other user actions of the user and other users. At least some of the vector data, optimal other opponent's initial facial expression vector data, optimal specific opponent's initial action vector data, and optimal other opponent's initial action vector data may be learned.

Here, the computing device 100 causes the learning device to learn at least some of the optimal specific opponent's initial expression vector data, the optimal other opponent's initial expression vector data, the optimal specific opponent's initial action vector data, and the optimal other opponent's initial action vector data. By doing so, it will be possible to increase the probability that the user and other users will see the other person's initial expression or action and input a specific user action to reduce the user's anxiety. Here, the concept of increasing the probability of entering a specific user action may literally mean that the probability of entering a specific user action to reduce the user's anxiety is high, but through viewing this content, a specific user It may also include a measure of the degree to which the probability of entering an action is improved compared to before.

And, with reference to a specific counterpart response action, computing with the m-th user emotion level for at least one m-th user emotion category that is a negative emotion category among the first to n-th user emotion categories additionally obtained. The device 100 exposes (i) information about the user and information about at least one other user having a similarity to the user equal to or greater than a preset threshold, (ii) initial facial expression vector data of a specific opponent, and corresponding information to other users. (iii) the first facial expression vector data of a specific user and the first behavioral vector data of a specific user that was exposed to other users, (iv) the facial expression vector data of a specific user, and the vector data of the specific user's facial expressions that were exposed to other users Other user expression vector data that was input, (vi) specific user behavior vector data, and other user behavior vector data that was input by the other user corresponding thereto, (v) a'th other user's expression component corresponding to the ath user's expression component. Expression vector data corresponding to a specific opponent having as a component, and facial expression vector data corresponding to another opponent having as a component a'th other opponent's expression component corresponding to the a'th other user's facial expression component that was exposed to the corresponding other user, (vi) Specific counterpart-corresponding behavior vector data having the 'b' counterpart behavior component as an element corresponding to the 'b' user behavior component, and the 'b' other-party behavior component corresponding to the 'b' other user behavior component that was exposed to other users corresponding thereto. At least a portion of the other counterpart's corresponding behavior vector data having as a component, and (vii) the additionally acquired mth user's emotion level, and the mth other user's emotion level additionally acquired from the other user corresponding thereto, can be input to the learning device. there is.

Then, the computing device 100 causes the learning device to provide optimal specific opponent initial facial expression vector data for lowering the final anxiety level of at least some of the mth user emotion level and the mth other user emotion level of the user and other users, Optimal other opponent's initial expression vector data, optimal specific opponent's initial action vector data, optimal other opponent's initial action vector data, optimal specific opponent's corresponding expression vector data, optimal other opponent's corresponding facial expression vector data, optimal specific opponent's corresponding action vector data, and optimal It is possible to learn at least some of the other opponent's corresponding action vector data.

At this time, the computing device 100 causes the learning device to provide vector data about the other person's initial expression and initial action exposed in the initial situation within the specific content and the other person's corresponding expression and corresponding action exposed in response to the specific user action. By learning at least some of the vector data, it will be possible to increase the probability of lowering the final anxiety level of the user and other users. Here, the concept of increasing the probability of lowering the final anxiety level may literally mean that the probability of lowering the user's anxiety level is high, but the ability to lower the final anxiety level through viewing this content is greater than before. It may also include a measure of the degree of improvement.

Additionally, the embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention, or may be known and usable by those skilled in the computer software field. Examples of computer-readable recording media include hard disks, magnetic media such as floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floptical disks. , and hardware devices specifically configured to store and perform program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include not only machine language code such as that created by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the invention and vice versa.

In the above, the present invention has been described with specific details such as specific components and limited embodiments and drawings, but this is only provided to facilitate a more general understanding of the present invention, and the present invention is not limited to the above embodiments. , a person skilled in the art to which the present invention pertains can make various modifications and variations from this description.

Claims (16)

In a method for predicting and reducing the negative emotions of others appearing in specific content provided to users,
(a) The computing device, as a preset condition for the specific content, (i-1) specific object data for at least one specific object as an object for the user to interact with at least one other party appearing in the specific content, and (i-2) determining specific state data for the specific object, and (ii-1) specific expression vector data corresponding to the facial expression of the other party as a variable setting condition for the specific content - the specific facial expression vector data is the first Contains as components a first facial expression component to an n-th facial expression component for representing each of the opponent's emotion categories and the nth opponent's emotion category - and (ii-2) specific behavior vector data corresponding to the opponent's behavior - the specific behavior vector Data has as components a first behavioral component to an n-th behavioral component for representing each of the first emotional category to the nth emotional category; determining at what timing of the specific content to expose at least a portion of the data; providing the specific content with reference to at least some of the specific object data and the specific state data as the preset conditions, the specific expression vector data as the variable setting conditions, and the specific action vector data;
(b) When the specific expression vector data and the specific action vector data are provided, the computing device causes the user to select a preset of the first to nth expression components included in the specific expression vector data. A process for predicting the presence or absence of an i-th other person's emotion category corresponding to at least one ith facial expression component corresponding to an expression threshold or higher, and the first to n-th behavior components included in the specific behavior vector data. performing a process to support predicting the presence or absence of a jth opponent's emotion category corresponding to at least one jth behavior component corresponding to a preset behavior threshold or higher; and
(c) In a state in which the user predicts the existence of the ith other person's emotion category corresponding to the ith facial expression component and the existence of the jth other person's emotional category corresponding to the jth behavioral component, the user A specific method for adjusting the size of the emotional level of at least some of the i-th emotional level of the other party corresponding to the ith emotional category and the j-th emotional level of the other party corresponding to the j-th emotional category When a user response action is input, the computing device, with reference to the specific user response action, sets an i-th partner adjusted emotion level, which is an emotional level adjusted to correspond to each of the i-th partner emotion level and the j-th partner emotion level, and jth step of exposing the other party's adjusted emotional level;
How to include . According to paragraph 1,
(d) the computing device adjusts the ith counterpart's adjusted emotional level and the jth emotional level, which are emotional levels adjusted to correspond to each of the ith counterpart's emotional level and the j'th counterpart's emotional level with reference to the specific user response behavior. In a state where the other party's adjusted emotion level is exposed, user emotion vector data due to the specific user response action - the user emotion vector data is the first to nth user emotion levels for each of the first to nth user emotion categories Indicates user emotion level - when it is detected that the mth user emotion level for at least a part of the mth user emotion category has changed by more than a preset change threshold, a step of recording the mth user emotion category and change information corresponding thereto is further performed. A method characterized by comprising: According to paragraph 2,
When it is detected that the mth user emotion category has changed by more than the preset change threshold, the computing device selects at least some of the emotion categories corresponding to the positive emotion of the other person among the first other person's emotion category and the nth other person's emotion category. A process of increasing the size of the expression level of the a-th expression component for representing the a-th other person's emotion category by a first size, which is the size of a preset expression vector, and exposing it to the user, and the first other person's emotion category to the n-th other person's emotion. Increase the size of the action level of the b action component for representing the b other person emotion category, which is at least part of the emotion category corresponding to the positive emotion of the other person among the categories, by the second size, which is the size of the preset action vector, and expose it to the user. Additionally providing compensation content for lowering the m-th user emotion level for the m-th user emotion category that has changed more than the pre-set change threshold of the user to less than the pre-set change threshold by performing at least some of the processes. A method characterized by: According to paragraph 1,
The computing device exposes (i) information about the user and information about at least one other user having a similarity to the user greater than a preset threshold, (ii) the specific facial expression vector data and corresponding information to the other user. other facial expression vector data, (iii) the specific action vector data and the corresponding other action vector data exposed to the other user, and (iv) the specific user corresponding action, the other user corresponding action input by the other user. and inputting at least some of the specific user's correct answer response behavior and the other user's correct response response behavior corresponding thereto into the learning device, thereby causing the learning device to select among the specific user response behavior and the other user response behavior of the user and the other user. Characterized by learning at least some of optimal specific expression vector data, optimal other expression vector data, optimal specific action vector data, and optimal other action vector data in order to increase at least some of the accuracy above a preset threshold. How to. According to paragraph 1,
In step (a) above,
The computing device, as the variable setting condition of the specific content, (ii-3) specific language vector data corresponding to the other party's utterance - the specific language vector data includes the first counterpart emotion category to the nth counterpart emotion category Having as components a first language component to an n-th language component to represent each - additionally determines at which timing of the specific content to be exposed, and additionally refers to the specific language vector data as the variable setting condition to display the specific content. provide,
In step (b) above,
When the specific language vector data is additionally provided, the computing device causes the user to use at least one of the first language component to the n-th language component included in the specific language vector data that is equal to or higher than a preset language threshold. Additional processes are performed to support predicting the existence of the k-th emotional category corresponding to the k-th language component of
In step (c) above,
In a state in which the existence of the k-th counterpart emotion category corresponding to the k-th language component is additionally predicted by the user, the k-th counterpart emotion level of the other party corresponding to the k-th counterpart emotion category is predicted by the user. When the specific user response action for additionally adjusting the size is input, the computing device determines the kth opponent adjusted emotion level, which is an emotional level adjusted corresponding to the kth opponent emotion level, with reference to the specific user response action. A method characterized by additional exposure. According to paragraph 1,
(e) With reference to the specific user response behavior, the i-th opponent's adjusted emotion level and the j-th opponent's adjusted emotion level are emotional levels adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level. If it is confirmed that the ith opponent's stable emotion level and the jth opponent's stable emotion level, which are respectively preset stability levels, are satisfied, the computing device collects the specific object data and the specific state data as the preset conditions of the specific content. By performing at least some of the process of adjusting and the process of adjusting the specific expression vector data and the specific action vector data as the variable setting conditions and the ith opponent's emotion level and the jth opponent's emotion level, the specific content The method further comprising providing coordinated content. According to clause 6,
With reference to the specific user response behavior, the i-th opponent's emotion level and the j-th opponent's emotion level are preset, which are emotion levels adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level. If it is confirmed that the ith counterpart stable emotion level and the jth counterpart stable emotion level, which are stable levels, are satisfied, the computing device tells the user that the correct response behavior for the specific user response behavior exists as a plurality of solutions. A method characterized in that the adjusted content is provided by adjusting at least some of the preset conditions and the variable setting conditions. According to paragraph 1,
A method characterized in that the computing device selects and displays the i-th counterpart emotion category and the j-th counterpart emotion category from emotion categories representing negative emotions of the other party. In a computing device for predicting and reducing the negative emotions of others appearing in specific content provided to the user,
at least one memory storing instructions; and
At least one processor configured to execute the instructions,
The processor may include (I) as a preset condition for the specific content, (i-1) specific object data for at least one specific object as an object for the user to interact with at least one other party appearing in the specific content, and (i-2) determining specific state data for the specific object, and (ii-1) specific expression vector data corresponding to the facial expression of the other party as a variable setting condition for the specific content - the specific facial expression vector data is the first Contains as components a first facial expression component to an n-th facial expression component for representing each of the opponent's emotion categories and the nth opponent's emotion category - and (ii-2) specific behavior vector data corresponding to the opponent's behavior - the specific behavior vector Data has as components a first behavioral component to an n-th behavioral component for representing each of the first emotional category to the nth emotional category; determining at what timing of the specific content to expose at least a portion of the data; A process of providing the specific content with reference to at least some of the specific object data and the specific state data as the preset conditions, the specific facial expression vector data, and the specific action vector data as the variable setting conditions; (II) When the specific expression vector data and the specific action vector data are provided, the user is asked to make one of the first to nth expression components included in the specific expression vector data correspond to a preset expression threshold or higher. a process for predicting the presence or absence of the ith other person's emotion category corresponding to at least one ith facial expression component, and a preset behavior threshold among the first to nth behavior components included in the specific behavior vector data. A process for performing a process that supports predicting the presence or absence of the jth opponent's emotion category corresponding to at least one jth behavior component corresponding to the above; and (III) in a state in which the presence or absence of the i-th counterpart emotion category corresponding to the i-th facial expression component and the presence or absence of the j-th counterpart emotion category corresponding to the j behavior component are predicted by the user, For adjusting the size of the emotional level of at least some of the i-th emotional level of the other party corresponding to the i-th other-party emotion category and the j-th emotional level of the other party corresponding to the j-th counterpart emotional category by the user. When a specific user response action is input, the i-th opponent adjustment emotion level and the j-th opponent adjustment are emotional levels adjusted to correspond to each of the i-th opponent emotion level and the j-th opponent emotion level with reference to the specific user response action. A computing device characterized in that it performs a process of exposing emotional levels. According to clause 9,
The processor,
(IV) With reference to the specific user response behavior, the i-th opponent's adjusted emotion level and the j-th opponent's adjusted emotion level, which are emotional levels adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level. In the exposed state, user emotion vector data due to the specific user response action - the user emotion vector data represents the first user emotion level to the nth user emotion level for each of the first user emotion category to the nth user emotion category. - When it is detected that the m-th user emotion level for at least a part of the m-th user emotion category has changed by more than a preset change threshold, further performing a process of recording the m-th user emotion category and corresponding change information. A computing device that does. According to clause 10,
The processor,
When it is detected that the mth user's emotion category has changed by more than the preset change threshold, the ath other person's emotion category is at least some of the emotion categories corresponding to the positive emotion of the other person among the first to the nth other person's emotion categories. A process of increasing the size of the expression level of the a-th expression component to indicate a first size, which is a preset expression vector size, and exposing it to the user; At least part of the process of increasing the size of the action level of the action component for representing the emotion category of the other person, which is at least part of the emotion category corresponding to a positive emotion, by the second size, which is the size of the preset action vector, and exposing it to the user. Performing a process of additionally providing compensation content for lowering the mth user emotion level for the mth user emotion category that has changed more than the preset variation threshold of the user to less than the preset variation threshold by performing the process of A computing device characterized by: According to clause 9,
The processor is configured to: (i) information about the user and information about at least one other user having similarity to the user greater than a preset threshold, (ii) the specific facial expression vector data and the information corresponding thereto that has been exposed to the other user. Other facial expression vector data, (iii) the specific action vector data and corresponding other action vector data exposed to the other user, and (iv) the specific user corresponding action, the other user corresponding action input by the other user, and At least some of the specific user's correct answer response behavior and the other user's correct response response behavior corresponding thereto are input to the learning device, thereby causing the learning device to: at least one of the specific user response behavior and the other user response behavior of the user and the other user Performing a process of learning at least some of the optimal specific expression vector data, optimal other expression vector data, optimal specific action vector data, and optimal other action vector data in order to increase some of the accuracy above a preset threshold. A computing device characterized in that. According to clause 9,
In process (I) above,
The processor,
As the variable setting condition of the specific content, (ii-3) specific language vector data corresponding to the other party's utterance - the specific language vector data is for representing each of the first counterpart emotion category to the nth counterpart emotion category. Having a first language component to an nth language component as components - additionally determines at what timing the specific content will be exposed, and provides the specific content by additionally referring to the specific language vector data as the variable setting condition,
In process (II) above,
When the specific language vector data is additionally provided, the user is prompted to select at least one k-th language component corresponding to a preset language threshold or higher among the first language component to the n-th language component included in the specific language vector data. Additional processes are performed to support predicting the existence of the kth counterpart's emotion category corresponding to
In process (III) above,
In a state in which the existence of the k-th counterpart emotion category corresponding to the k-th language component is additionally predicted by the user, the k-th counterpart emotion level of the other party corresponding to the k-th counterpart emotion category is predicted by the user. When the specific user response action for additionally adjusting the size is input, a process of additionally exposing the kth opponent's adjusted emotion level, which is an emotional level adjusted in correspondence to the kth opponent's emotion level, with reference to the specific user response action. A computing device characterized in that: According to clause 9,
The processor,
(V) With reference to the specific user response behavior, the i-th opponent's adjusted emotion level and the j-th opponent's adjusted emotion level are emotional levels adjusted corresponding to each of the i-th opponent's emotion level and the j-th opponent's emotion level. A process of adjusting the specific object data and the specific state data as the preset conditions of the specific content when it is confirmed that the ith opponent's stable emotion level and the jth opponent's stable emotion level, respectively, which are preset stability levels, are satisfied; Adjusted content obtained by adjusting the specific content by performing at least part of the process of adjusting the specific expression vector data and the specific action vector data as variable setting conditions, the ith opponent's emotion level, and the jth opponent's emotion level. A computing device characterized in that it further performs the process provided. According to clause 14,
The processor,
With reference to the specific user response behavior, the i-th opponent's emotion level and the j-th opponent's emotion level are preset, which are emotion levels adjusted to correspond to each of the i-th opponent's emotion level and the j-th opponent's emotion level. If it is confirmed that the ith counterpart stable emotion level and the jth counterpart stable emotion level, which are stable levels, are satisfied, the preset condition and the above so that the correct response behavior for the specific user response behavior exists as multiple solutions to the user. A computing device characterized in that it performs a process of providing the adjusted content by adjusting at least some of the variable setting conditions. According to clause 9,
The processor,
A computing device characterized in that it performs a process of selecting and exposing the i-th counterpart's emotion category and the j-th counterpart's emotion category from emotion categories representing negative emotions of the other party.