WO2024090713A1

WO2024090713A1 - User psychology management system through empathic psychology-based chatbot service

Info

Publication number: WO2024090713A1
Application number: PCT/KR2023/009746
Authority: WO
Inventors: 송환구
Original assignee: 주식회사 닥터송
Priority date: 2022-10-28
Filing date: 2023-07-10
Publication date: 2024-05-02
Also published as: KR102507811B1

Abstract

In order to provide a technology which provides, for a user's psychological enhancement, an empathic psychology-based chatbot service and manages users' states through characters according to the chatbot service, and enables substantial and targeted psychological counseling and enhancement through an interaction with a psychological counselor, a user psychology management system through an empathic psychology-based chatbot service according to an embodiment of the present invention comprises: an empathic sentence provision unit for generating and providing a correspondence sentence with respect to a user's input sentence, wherein the correspondence sentence, which includes a response keyword matched to an input keyword corresponding to a keyword included in the user's input sentence and is generated using an automatic sentence generation algorithm of generating a sentence associated with an empathic psychology-based topic, is generated and provided; a character management unit for outputting, on a user interface on which the correspondence sentence is provided, a character indicating the user's psychological state according to the user's input sentence, wherein a character having an appearance transformed according to user state information including pre-input state user information and the input keyword included in the input sentence is output so that the user's psychological state can be recognized through the user interface; and a psychological enhancement induction unit for providing a counselor account with the user state information and providing management content for the user's psychological state enhancement on the user interface on the basis of request information input from the counselor account.

Description

User psychological management system through chatbot service based on empathetic psychology

The present invention relates to a chatbot service that can conduct non-face-to-face psychological counseling with users based on empathic psychology and technology to manage the psychological state of users using it. Specifically, it is customized to the psychological state of the user and responds to the user's reaction. Technology that provides a chatbot containing sentences and content that resonates with the user, providing psychological treatment effects through forming a consensus with users, while enabling interaction with a psychological counselor and intuitively recognizing the user's condition. It's about.

Recently, artificial intelligence chatbots have been used in a wide variety of fields, including answering user inquiries, automatic ordering, and translation. In particular, thanks to the development of natural language analysis algorithms, artificial intelligence chatbots and artificial intelligence systems that can communicate with users while providing the same effect as talking to a natural person are being released.

This is especially used very efficiently in voice recognition and conversation systems such as Siri, Google Assistant, and Bixby. Users can enter their needs by voice and do what they want or get the information they want in a natural conversation format. there is.

Meanwhile, it is pointed out that these artificial intelligence chatbots have great potential for use in terms of using professional services, such as user health management, that users could normally use by visiting experts such as counselors and medical staff. Previously, counseling was possible by visiting a hospital or counseling center run by psychological counselors, psychiatrists, etc. or through a landline phone call. Accordingly, it was impossible for users who did not have money or time to receive such psychological counseling. This is because when implemented as an intelligent chatbot, the burden of time and cost consumption is greatly reduced.

In response to these needs, technologies for coaching and counseling on the health status of users using artificial intelligence are being released. For example, in Korean Patent No. 10-2421643, a process was added to analyze users' interest in psychotherapy scenarios by linking apps with robots such as counseling systems and chatbots, and to motivate users through this. Thus, it provides technology to increase the effectiveness of psychotherapy by keeping users interested in psychotherapy scenarios.

As mentioned above, these technologies, such as artificial intelligence chatbots for psychotherapy, have the advantage of greatly reducing the burden of time and cost consumption for psychotherapy, but despite utilizing natural language processing technology, they cannot accurately understand the user's psychology. There is a shortcoming in providing technology that can provide practical help in relieving depression and stress by analyzing and using this to provide users with the best psychological treatment effect.

In addition, since it is limited to a non-face-to-face service that can simply improve the psychological state, continuity with expert counseling is not practically guaranteed, so there is no synergy between the psychological improvement effect of using the service and the effect of expert counseling, so the service There is a problem that reliability is greatly reduced.

Accordingly, the present invention provides a chatbot service according to a psychotherapy method that can induce positive psychology by sympathizing with the status and reactions of users based on empathic psychology, especially to optimize depression and stress, and provide a chatbot service for each user's status. The purpose is to provide a customized treatment method and provide technology to maximize the effect of treatment by creating a positive atmosphere.

In addition, the present invention shares the state of psychological improvement using a chatbot with a professional counselor and provides various additional content that can be used in the counselor's consultation, thereby creating a synergy between the psychological improvement effect using a chatbot and the psychological treatment effect through professional counseling. Another purpose is to maximize the effectiveness of psychological treatment by encouraging the participation of professional counselors as well as chatbot services, while providing technology that stimulates the market.

In order to achieve the above purpose, the user psychology management system through a chatbot service based on empathic psychology according to an embodiment of the present invention generates and provides a corresponding sentence for the user's input sentence, and provides a corresponding sentence included in the user's input sentence. An empathic sentence providing unit that generates and provides corresponding sentences that include response keywords matched to the input keywords and are generated using an automatic sentence generation algorithm that generates sentences related to topics based on empathic psychology; A character representing the user's psychological state according to the user's input sentence is output on the user interface where the corresponding sentence is provided, according to the user state information including the input keyword included in the input sentence and user information in the previously input state. a character management unit that outputs a character whose appearance is transformed and allows the user's psychological state to be recognized through a user interface; and a psychological improvement inducing unit that provides the user state information to the counselor's account and provides management content for improving the user's psychological state on the user interface based on the request information input from the counselor's account.

The sympathetic sentence providing unit, as user information in a pre-entered state among the user state information, user input information for menus for determining the user's psychological state provided through the user interface, and a sensor of the user terminal Provides an initial sentence generated using the automatic sentence generation algorithm based on the response keyword matched with the initial user state keyword, which is a keyword representing the user's psychological state identified based on the user's facial image and voice information input through first step; A second step of providing a plurality of response-type sentences according to the user's input sentence for the initial sentence; And when the number of times the user's input sentence is received reaches a preset threshold, a third step of providing a closing sentence based on a response keyword matching the keyword corresponding to the user status information. Through a process including, It is possible to provide the initial sentence, response sentence and closing sentence as the corresponding sentence.

In the first step, the sympathetic sentence providing unit is a menu output through a user interface, and provides a first keyword input through a keyword selection menu from which at least one of a plurality of keywords can be selected, and an input menu capable of character input. A second keyword judged to be related to the text input through a correlation analysis technique, an image matching at least one keyword is output, and an image selected through an image selection menu that allows selection of some of the output images is selected. A third keyword, which is a keyword, a word included in the user's voice input through a voice input means, a fourth keyword, which is a keyword derived by analyzing voice information that is at least one of the tone, frequency, and speaking speed of the voice information, and a video recording means. It is possible to derive a user status keyword using at least one of the fifth keywords, which are keywords derived as a result of analyzing the user's facial image captured through an emotion analysis algorithm.

The sympathetic sentence providing unit includes a first process that manages association information between input keywords and response keywords based on matching information input from a plurality of preset expert accounts, and a response included in a sentence already provided by the sympathetic sentence providing unit. By using keywords and the influence figures of the response keywords on the user's psychological state derived using the time-specific change information of the user state information, the user can be adjusted according to the degree of improvement in the user's psychological state included in the user state information. It is possible to manage the matching relationship between the input keyword and the response keyword by at least one of the second processes that manage the correlation value between the input keyword and the response keyword included in the status information.

The sympathetic sentence providing unit is capable of generating corresponding sentences using an automatic sentence generation algorithm that can reflect LSTM-based contextual meaning in word vector expressions.

The character management unit changes one of the character's expressions, movements, colors, and accessories according to the user status information, and changes the character's expressions, movements, color types, and accessories according to the type of psychological state according to the user's status information. It is possible to vary the type, and depending on the degree of psychological condition, the size of the character's movements, color saturation or brightness, and the size of the accessories.

After the sympathetic sentence is provided on the user interface by the sympathetic sentence providing unit, the character management unit receives the user's corresponding input sentence, during the time when the sympathetic sentence and the input sentence that caused the input sentence are output, or The goal is to output a character with a modified appearance according to the sympathetic sentence that caused the input sentence and the user status information according to the input keyword identified by the input sentence in the area where the input sentence is output. possible.

The character management unit is capable of outputting different characters according to the type of user's psychological state.

The psychological improvement induction unit is capable of providing information on measurement values and change patterns of user status information during a preset period to the counselor's account.

The psychological improvement inducing unit includes, as the management content, at least first mission content suggesting the number of uses of the sympathetic sentence provision unit, second mission content suggesting use of a service related to improving psychological state online, and the counselor account; It is possible to provide a menu that can use content including any one of the third mission content that suggests psychological state improvement activities through interaction as the management content on the user interface.

According to the present invention, the user's status is comprehensively analyzed through the keywords selected by the users, sentences entered, images or voices selected, etc., and sentences based on empathic psychology with matching keywords are provided to the users. In particular, while providing a number of sentences and closing comments to users in stages, very positive sentences are selected to users based on empathic psychology or generated and provided through a natural language processing algorithm.

In this case, compared to simple basic psychotherapy techniques, the effect of relieving depression and stress is greatly increased, and optimally targeted psychotherapy effects can be provided to these users, making psychotherapy of focused users possible. It works.

In addition, because it allows very positive emotions to be accurately reflected in the creation and provision of sentences, it has the effect of making it possible to provide focused sentences based on empathic psychology.

In addition, in matching and providing keywords, it is provided based on users' reactions and psychological improvement status, so it is greatly helpful in improving the psychological status of users. When creating sentences, the sentences are designed to have a sentence structure that can accurately improve the users' status. This has the effect of maximizing the effectiveness of psychotherapy.

Meanwhile, according to the present invention, users can intuitively check their psychological state as it changes while using sympathetic sentences through changes in the appearance of the character, so they can feel new fun in using the service and clearly understand their state. This has the effect of improving reliability and usage rate by intuitively understanding the effect of using the service.

In addition, according to the present invention, changes in the psychological state that users have while using the service are provided to the professional counselor's account, and various additional contents are provided based on the requested information received from the counselor's account. Because the synergy between the effect of improving the psychological state through use and the effect of improving the psychological state by a professional counselor is maximized and provides a new target market for professional counselors, the chatbot service of the present invention as well as the participation of professional counselors is encouraged. , It has the effect of increasing the effectiveness of psychological treatment and at the same time inducing market revitalization.

Figure 1 is a block diagram of a user psychology management system through a chatbot service based on empathic psychology according to an embodiment of the present invention.

Figure 2 is a diagram for explaining the function of a user input analysis unit according to the implementation of an embodiment of the present invention.

Figure 3 is a diagram for explaining an example of voice input being analyzed according to the implementation of an embodiment of the present invention.

Figure 4 is a diagram for explaining the function of a keyword derivation unit according to the implementation of an embodiment of the present invention.

Figure 5 is a diagram for explaining the detailed functions of a sympathetic sentence provision unit according to the implementation of an embodiment of the present invention.

6 and 7 are diagrams for explaining the function of the character management unit according to an embodiment of the present invention.

FIG. 8 is a diagram illustrating an example in which user status information is provided to a counselor account according to the implementation of an embodiment of the present invention.

Figure 9 is a diagram for explaining an example in which management content is output to a user interface according to the implementation of an embodiment of the present invention.

10 is an example of the internal configuration of a computing device according to an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS Various embodiments and/or aspects are now disclosed with reference to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth to facilitate a general understanding of one or more aspects. However, it will also be appreciated by those skilled in the art that this aspect(s) may be practiced without these specific details. The following description and accompanying drawings set forth in detail certain example aspects of one or more aspects. However, these aspects are illustrative and some of the various methods in the principles of the various aspects may be utilized, and the written description is intended to encompass all such aspects and their equivalents.

As used herein, “embodiment,” “example,” “aspect,” “example,” etc. may not be construed to mean that any aspect or design described is better or advantageous than other aspects or designs. .

Additionally, the terms "comprise" and/or "comprising" mean that the feature and/or element is present, but exclude the presence or addition of one or more other features, elements and/or groups thereof. It should be understood as not doing so.

Additionally, terms containing ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, a first component may be referred to as a second component, and similarly, the second component may be referred to as a first component without departing from the scope of the present invention. The term and/or includes any of a plurality of related stated items or a combination of a plurality of related stated items.

In addition, in the embodiments of the present invention, unless otherwise defined, all terms used herein, including technical or scientific terms, are as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. It has the same meaning. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless clearly defined in the embodiments of the present invention, have an ideal or excessively formal meaning. It is not interpreted as

1 is a block diagram of the configuration of a user psychological management system through a chatbot service based on empathic psychology according to an embodiment of the present invention, and FIG. 2 is a diagram illustrating the function of the user input analysis unit according to the implementation of an embodiment of the present invention. , FIG. 3 is a diagram illustrating an example in which voice input is analyzed according to the implementation of an embodiment of the present invention. Figure 4 is a diagram for explaining the function of the keyword derivation unit according to the implementation of an embodiment of the present invention, Figure 5 is a diagram for explaining the detailed function of the sympathetic sentence provision unit according to the implementation of an embodiment of the present invention, Figures 6 and 7 is a diagram for explaining the function of the character management unit according to an embodiment of the present invention, FIG. 8 is a diagram for explaining an example in which user status information is provided to a counselor account according to the implementation of an embodiment of the present invention, and FIG. 9 is a diagram for explaining the function of the character management unit according to an embodiment of the present invention. This diagram is for explaining an example in which management content is output to a user interface according to the implementation of an embodiment of the present invention. In the following description, one or more of the drawings may be referred to simultaneously for detailed description of various embodiments and components of the present invention.

To explain based on the above, the user psychological management system (10, hereinafter referred to as 'system of the present invention') through a chatbot service based on empathic psychology according to an embodiment of the present invention includes an empathetic sentence provider (11) ), a character management unit 12, and a psychological improvement induction unit 13.

The system 10 of the present invention will be described in FIG. 1 as a separate configuration linked to the user terminal 20 and the database 30. However, in various embodiments of the present invention, the system 10 of the present invention is configured to be included in the user terminal 20 and transmits and receives data with a remote database 30 and is performed according to the embodiments mentioned in the function of the present invention. Alternatively, the system 10 of the present invention is linked with the database 30 to form a server that exists remotely from the user terminal 20, and exchanges data with the user terminal 20 through the interface of the user terminal 20. It can be configured to transmit and receive.

To explain based on this, as shown in the drawings, the sympathetic sentence providing unit 11 generates and provides a corresponding sentence for the user's input sentence, and matches the input keyword, which is a keyword included in the user's input sentence. It performs the function of generating and providing response sentences created using an automatic sentence generation algorithm that includes response keywords and generates sentences related to topics based on empathic psychology.

In the present invention, the user's input sentence is a set of interfaces including a number of menus and screens output by an online website accessed to perform the function of the present invention on the user terminal 20, an application or software being executed, etc. This refers to input data in the form of text received through user input through a menu displayed on the user interface.

From the user's input sentence, keywords are derived according to the structure of the sentence, etc. through a sentence analysis algorithm. This is referred to as an input keyword, and the input keyword includes a noun included in the user's input sentence or a keyword corresponding to emotional information derived through a sentiment analysis algorithm for the input sentence.

Sentiment analysis is the process of analyzing subjective information such as opinions, emotions, evaluations, and attitudes contained in text using a computer. Sentiment analysis largely consists of two steps. The first step is to define which part of the document (sentence) contains the opinion (Opinion definition). Next, the opinions collected through the first step are summarized (opinion summerization).

The methodology of sentiment analysis can be divided into vocabulary-based sentiment analysis and machine learning-based sentiment analysis. Each method can be divided into more detailed methods, and recently, transformer-based neural network models such as BERT have been widely used.

Through this, when an input keyword is derived, the empathy sentence providing unit 11 uses an automatic sentence generation algorithm to generate a sentence related to a topic based on empathy psychology, which describes a corresponding sentence containing a response keyword that matches the user interface. It is output in the form of a response to the user's input sentence.

For example, the automatic sentence generation algorithm can be divided into a passive algorithm and an active algorithm.

The passive algorithm filters and provides sentences containing response keywords from a sentence bank format that is created and stored in advance by an expert account, or includes response keywords from a number of sentences that already exist online, and provides input keywords. It includes a process of generating matching sentences as corresponding sentences.

For example, among a plurality of sentences stored online, sentences related to improvement by user's psychological state are crawled, and then sentences containing a response keyword are selected and provided, but among the crawled plurality of sentences, the response keyword is The process of selecting the included sentences as candidate sentences is carried out.

Afterwards, a sentence sentiment analysis algorithm based on empathy psychology was applied to the positive and negative state information of the sentences derived through the natural language processing algorithm for the candidate sentences, and among the candidate sentences, sentences judged to induce positive sympathy for the user's state were selected. This refers to the process of selecting as the final corresponding sentence.

With the development of neuroscience, research results on cognitive psychology or the psychology of empathy are continuously being published. Joachim Bauer approaches it from a psychiatric perspective, and through the discovery of 'mirror neurons', it is possible to create emotional resonance with others. Empathy psychology provides a clue to understand the phenomenon of being able to use intuitive knowledge to understand the intentions of people around you, and to share the joy and pain of others.

This empathy psychology also includes the topic of positive psychology, which is a field of psychology that has recently been attracting attention that studies the strengths and advantages that make individuals and society prosper. Research is being conducted to improve general life rather than just treating mental illness. In other words, positive psychology is more about medical conditions or clinical processes than medical conditions or clinical processes, in the context that a life that puts life for oneself and life for others in the same direction contributes to the original life of a human being. It is a field of psychology that places greater emphasis on the development of an individual's naturally occurring abilities.

Positive psychologists have suggested many ways in which individual happiness can be promoted. Social connections with partners, family, friends and wider networks through work, community or social organizations are particularly important, while physical exercise and the practice of meditation can also contribute to happiness. In this context, happiness can increase even when financial income stagnates or even declines due to no further economic gains.

Deriving keywords based on this empathic psychology is a process carried out for the purpose of empathizing with the subject's psychology and continuously providing positive stimulation. In other words, for example, if a user state keyword called 'collapse' is derived, keywords such as 'satisfaction', 'achievement', and 'rewarding', which are keywords that are judged to have an improvement effect, are derived as response keywords.

The active algorithm refers to an algorithm that automatically generates sentences related to the topic from the beginning based on response keywords, in addition to the passive sentence selection algorithm described above.

For example, the empathic sentence provider 11 determines the priority of improving the psychological state of the semantic information for each response keyword according to the arrangement order for each keyword derived through keyword analysis processing of a large number of existing sentence information online. performs the function of This means the same process as the process of the same name for determining the output order of sentences in the above-described embodiment.

Afterwards, a plurality of sentences containing response keywords are generated according to a determined order through a natural language processing algorithm, and the composition frame of the sentence and the emotional condition information of the sentence set by the expert account based on empathy psychology are applied to the natural language processing algorithm. A process of generating multiple sentences is performed through a sentence generation algorithm.

The sentence generation algorithm is a natural language processing algorithm and may include, for example, a sentence generation algorithm through an LSTM-based language learning model. At this time, an LSTM-based sentence generation algorithm that can automatically generate sentences by considering the topic of the document can be used through a model that can reflect the contextual meaning of the word in the word vector expression. At this time, considering the topic of the document, various topics related to empathy and positive psychology can be set.

For example, the Many-to-one model is used to reflect the context information of words, and document vectors can be additionally learned and used with words to reflect the topic of the document. As described above, sentences are generated through an LSTM-based language learning model. At this time, word vectors are influenced by document vectors, making it possible to express words according to topics.

Afterwards, a process is performed to control the plurality of sentences generated to be output through the output means of the user terminal 20 as corresponding sentences according to the sentence order determined by the above process.

Meanwhile, the above passive algorithm and active algorithm can be used independently or in combination with each other. When used in combination, the active algorithm is applied preferentially so that the sentences provided by the active algorithm are stored in a space included in the online space, which is the standard for sentence selection in one embodiment, and the sentences stored in the online space are stored in a space included in the online space. When the number reaches a preset threshold number (for example, 10,000), a passive algorithm is applied, and other embodiments are continuously provided, but are first provided to the counselor terminal 40, and after feedback from the expert account, online. By being stored in space, it can be processed into a sentence that can be used in one embodiment.

According to various embodiments of the present invention, the user's status is comprehensively analyzed through the keywords selected by the users, sentences entered, images or voices selected, etc., and sentences based on empathic psychology with keywords matching the users are provided to the users. I do it. In particular, while providing a number of sentences and closing comments to users in stages, very positive sentences are selected to users based on empathic psychology or generated and provided through a natural language processing algorithm.

Meanwhile, in the present invention, corresponding sentences can be provided in a step-by-step structure for psychological treatment. For example, when starting psychological treatment, it may include an initial sentence to open the minds of users, a response sentence based on the users' full-fledged input sentences, and a closing sentence to finally complete the psychological improvement.

At this time, as shown in FIG. 5, the sympathetic sentence providing unit 11 determines the user's psychological state provided through the user interface as the user information (D1) in a pre-entered state among the above-described user state information. Based on the response keyword matched with the initial user state keyword, which is a keyword representing the user's psychological state identified based on the user's input information for the menus and the user's facial image and voice information input through the sensor of the user terminal. The first step (S10) of providing an initial sentence generated using the automatic sentence generation algorithm (SG) is first performed to provide an initial sentence (S11).

At this time, as user information in a pre-entered state in the user status information, the user's input information, facial image, voice information, etc. can be identified using FIGS. 2 and 3.

For example, as shown in FIG. 2, an initial user state keyword is derived using at least one keyword among the first to fifth keywords.

The first keyword is an interface for receiving user input and analyzing the user's psychological state, and is selected and entered through the keyword selection menu 110, which allows selecting one of a plurality of keywords from the menu output through the user interface. A first keyword may be received. That is, the first keyword is a menu through which one of the user status keywords can be selected through the interface. At this time, a number of candidate keywords displayed on the keyword selection menu 110 may be specified through step-by-step selection. In other words, the keyword selection menu 110 may be provided as a menu group that allows step-by-step selection to be specified from a high-level concept to a keyword corresponding to a low-level concept.

Alternatively, in the present invention, the second keyword is a keyword that is determined to be related to the text input through the input menu 100 that allows character input through a correlation analysis technique. At this time, the second keyword may also function as one of the first to fourth keywords for deriving the user state keyword, but instead of performing a plurality of steps for selecting the first keyword, a selectable keyword corresponding to the user state keyword may be used. 1 It can also be used as a keyword to derive the psychological state of the higher-level concept that filters the keyword.

At this time, the correlation analysis technique is an analysis algorithm that numerically analyzes the correlation between words and keywords included in the sentence through vector analysis between texts and derives the keyword that is closest to it. For example, the most representative algorithm for analyzing associations between words through text mining can be used in the present invention, and in addition, algorithms for analyzing associations between various keywords can be used.

Text Mining is a tool for extracting and analyzing useful information from text data in electronic documents such as web pages, blogs, electronic journals, and emails. It refers to the process of deriving new, high-quality information from text data. , Text mining refers to an area where various fields such as data mining, natural language processing, and information retrieval are converged.

In text mining techniques, the techniques used as algorithms for analyzing relationships between words of the present invention include word cloud (extraction of key keywords), cluster analysis (cluster analysis (community analysis) between similar words (or documents)), Network analysis (extraction of related words or word network analysis), topic model (topic analysis, topic analysis) / topic trend analysis, outlier analysis (anormality), spam mail detection, plagiarism analysis (calculation of document similarity), classification A process of assigning documents to predetermined topics is used.

When analysis data is crawled online or received directly from an administrator terminal, analysis data is prepared by collecting the data, and a corpus (corpus) (a set of text forming a structure in linguistics is used for statistical analysis, hypothesis verification, and language rule inspection). etc., a corpus is usually composed of electronic documents), and after processing stop words and redundant words through a pre-processing process, etc., the relationships between words are analyzed through this analysis.

In the present invention, the analysis of relationships between keywords can be processed by combining various online sites, but a plurality of online sites and spaces predetermined by the administrator terminal as influencing the psychological state of users, such as psychotherapy and video content, are mentioned above. This can be done by preparing analysis data.

For example, when the user inputs the sentence "I feel somewhat cloudy today" through the input menu 100, it is judged as various candidate keywords that are related to keywords such as "cloudy" and "mood", for example, ' Keywords such as 'joy', 'sadness', 'anger', 'surprise', etc. are displayed in the keyword selection menu 110 as candidate keywords for the first keyword, or the keyword 'collapse', which has the highest correlation among them, is displayed as a candidate keyword for the first keyword. As a second keyword, it can be selected to derive a user status keyword.

In the present invention, the second keyword can be used as a filtering criterion for the first keyword as described above, and this can also be used to select the third keyword, which will be described later. As shown in FIG. 2, the third keyword refers to a keyword matched to an image selected through the image selection menu 120, which outputs an image matching at least one keyword and selects some of the output images.

The third keyword is also selected through the same process as the first keyword, except that it is the selection of an image, so a step-by-step selection menu is provided as mentioned in the description of the first keyword, or the second keyword is provided as described above. If some keywords are proactively selected, images matching the detailed sub-keywords belonging to them may be output through the image selection menu 120.

Meanwhile, the fourth keyword refers to a keyword derived by analyzing the user's voice information 130 input through a voice input means. At this time, the fourth keyword can be generated by extracting candidate keywords from multiple user status keywords through a more detailed analysis algorithm.

For example, as shown in FIG. 3, when the user's voice information 130 is analyzed to derive the fourth keyword, the user's voice information 130 is provided through, for example, sound to text conversion. 2 The keyword is converted into the same sentence (131) as the analysis target, and accordingly, a number of keywords (K2) are derived through the second keyword analysis algorithm, that is, the word correlation analysis algorithm, or the previously stored user's standard voice frequency and input A keyword (K3) can be derived based on the result of analyzing the user's emotional state through frequency spectrum analysis derived through comparison of the input voice frequency 132 derived through frequency analysis of the voice.

At this time, emotional state analysis through voice frequency analysis refers to analysis of tone, pitch, and formant frequency of voice signals. Pitch refers to the pitch of a sound when a person hears it with the ear, or refers to one cycle of a voiced sound that represents a periodic waveform. Just as human language has a social nature, the emotions expressed in language also have a social nature, so it is possible to understand a person's emotions based on the pitch of the language.

Formant frequency refers to the frequency at which certain overtones are strengthened among partial tones, and includes the frequencies nearby, which are referred to as formants. Accordingly, changes in frequency lead to changes in energy distribution, and when emotions change, physical/physiological changes occur and formant changes occur in chain, so analysis of this can also be used to analyze emotions. .

On the other hand, since frequency analysis means analyzing changes in frequency over time, detailed analysis of speech speed, tone, sound quality, words, etc., including the pitch and formant frequency analysis in a narrow sense, is required to analyze voice information. It will be understood that it can be used in the analysis of (130).

Meanwhile, the fifth keyword refers to a keyword derived as a result of analyzing the user's facial image captured through an image capturing means such as a camera of the user terminal 20 through an emotion analysis algorithm.

The emotion analysis algorithm is an algorithm that infers the user's emotions from the user's facial image included in the video data. In particular, the user's emotions are based on the positional relationship between multiple points in the user's facial area identified from the user's facial image. An algorithm can be used to infer.

For example, an algorithm for recognizing emotions using facial images using TensorFlow models and deep learning models, an algorithm for extracting facial areas using a fuzzy color filter, and extracting facial components using a virtual face model. Algorithm, facial image analysis, feature vector extraction and pattern recognition, fuzzy classifier method based on fuzzy theory suitable for emotion analysis, landmark detection in the image through Active Appearance Model and facial expression bias, and emotion analysis through movement and change of other landmarks Classification techniques can be used independently or in combination with each other.

When the first to fifth keywords are derived through the above-described process, one or more keywords representing the user's status are derived through frequency analysis or similarity and grouping analysis, and the above-mentioned user status keywords are derived. It is derived.

Meanwhile, when the initial sentence is provided, the sympathetic sentence providing unit 11 generates a response sentence (S12) generated by the automatic sentence generation algorithm (SG) described above according to the user's input sentences (S21, S22) for the initial sentence. , the second step (S20) is performed in which S13) is provided multiple times. For this, the above-mentioned active or passive algorithm is used, and at the same time, the sentence is generated through the association information between the input keyword and the response keyword included in the input sentence. The association value between the input keyword and the response keyword is as shown in Figure 4. It is managed by example.

As shown in FIG. 4, the keyword derivation unit generates user status keywords (K1, K2, K3) and response keywords (M1, M2, M3) based on matching information input from the terminal 40 used by a plurality of preset expert accounts. , M4) and the response keyword included in the sentence already provided by the sympathetic sentence provider 13, which will be described later, and the user's information derived by using the time-dependent change information of the user status keyword. By at least one of a second process that manages the correlation value between the user state keyword and the response keyword according to the degree of improvement in the psychological state of the user included in the user state keyword, using the influence value of the response keyword on the psychological state. The related keyword derivation algorithm 300 can be managed.

The first process refers to an analysis process for the correlation between user status keywords and response keywords, based on users' feedback and expert review results for keywords that are periodically performed by the administrator terminal 40.

Meanwhile, the second process, as shown in FIG. 4, is based on the results of analyzing the history information 400, that is, information on the improvement of the psychological state of users according to the use of the sentence previously provided by the sympathetic sentence provider 13. , refers to a feedback-based automated process that automatically manages relevance values according to improvement information for each user status keyword of response keywords.

In order to specifically perform the second process, the empathy sentence providing unit 11 re-acquires the user's user state information, for example, after providing the above-mentioned response sentence, and derives improvement information about the user's psychological state. can do. Improvement information is used to determine to what extent the user's psychological state has improved by using information about the correlation or error between the user state keyword after using the corresponding sentence and the user state keyword used to provide the sentence. It means information.

Thereafter, the function of managing the related keyword derivation algorithm is performed by updating the association relationship between the user status keyword and the response keyword based on the derived improvement information. In other words, the above-described second process is directly performed.

When the second step (S20) is performed, the sympathetic sentence provider 11 performs step S20, that is, when the number of times the user's input sentence is received reaches a preset threshold, the user status is re-measured at that point. The third step (S30) of providing a closing sentence (S1(n+1)) based on the response keyword matching the keyword corresponding to the information (S2n) is performed to complete the final one-time psychological counseling process.

Through this, it is a response sentence optimized for the user's condition and has the effect of automatically performing scenario treatment steps that can improve the user's psychological state.

Meanwhile, the character management unit 11 outputs a character representing the user's psychological state according to the user's input sentence on the user interface where the corresponding sentence is provided, and input keywords included in the input sentence and user information in a pre-entered state. It performs a function of allowing the user's psychological state to be recognized through the user interface by outputting a character whose appearance is modified according to the user state information included.

For example, a character output area 600 as shown in FIG. 6 may exist on the user interface, and preferably, the character output area 600 exists near the output area of the response sentence and the output area of the input sentence. This would be desirable so that the user can intuitively recognize the user's psychological state.

Through the character output area 600, a character implemented as a human face is expressed, and the character is expressed according to the current user's psychological state through facial expression 601, color 602, and accessory 603 elements. It can be implemented.

For example, the character's facial expression (601), movement, color type (602), and type of accessory (603) vary depending on the type of psychological state according to the user's state information, and the character's behavior varies depending on the severity of the psychological state. Any one of the size, color saturation or brightness, and size of the accessory can be varied.

Meanwhile, as described above, the purpose of locating the character output area 600 near the area where the input sentence and the corresponding sentence are output is to enable users to use sympathetic sentences while intuitively checking their psychological state. By being able to intuitively recognize the effect of using such sympathetic sentences, the synergy effect of improving the psychological state of users can be maximized by immediately recognizing the effect of improving the psychological state caused by the use of sympathetic sentences, Users' reliability of the service provided by the system 10 of the present invention can be significantly improved.

To this end, in one embodiment of the present invention, the character management unit 12 receives the user's corresponding input sentence after a sympathetic sentence is provided on the user interface by the sympathetic sentence providing unit, and identifies the cause of the input sentence. During the time when the sympathetic sentence and input sentence are output, or in the area where the sympathetic sentence and input sentence that caused the input sentence are output, user status information according to the input keyword identified by the sympathetic sentence and input sentence that caused the input sentence is displayed. Therefore, it is possible to output a character with a modified appearance.

In addition, as shown in FIG. 7, characters 611 with completely different appearances are output for each type of psychological state 610, so that users can more clearly understand what psychological state they are in. can do.

Meanwhile, the psychological improvement induction unit 13 provides the user state information identified as described above to the counselor account, and provides management content for improving the user's psychological state on the user interface based on the request information entered from the counselor account. performs its function.

For example, when using a plurality of corresponding sentences for each scenario provided by the empathy sentence providing unit 11, the psychological improvement inducing unit 13 provides detailed information about the corresponding corresponding sentences and input sentences to the counselor terminal 40. It can be provided to enable the counselor to analyze the psychological state of the user account in detail.

Additionally, understanding the degree of change in psychological state over time is more important than understanding the psychological state at a specific point in time. To this end, the psychological improvement induction unit 13 is capable of providing information on measurement values and change patterns of user status information during a preset period to the counselor account. For example, as shown in FIG. 8, the screen 620 displayed on the counselor terminal 40 includes user status information 621 indicating the type and degree of the psychological state of the user U1, and user status information 621. A change trend 622 of information over time may be provided. At this time, a plurality of user state information 621 and change trends 622 may be provided for each psychological state type, as shown in FIG. 8.

Meanwhile, the psychological improvement induction unit 13 not only provides the user's psychological state to the counselor's account as described above, but also provides interaction with the counselor in order to more effectively improve the user's psychology through synergy with providing sympathetic sentences. Content related to coaching can be provided as feedback to the user account.

For example, as shown in FIG. 9, the psychological improvement induction unit 13 may provide management content to the user account through a user interface. Through the management content management screen 630, users can check and use management content using request information entered by the counselor account, that is, selection information about management content to improve the user's psychology.

Management content includes, for example, first mission content (M1) suggesting the number of uses of the sympathetic sentence provision unit, second mission content (M2) suggesting use of services related to improving psychological status online, and the counselor account and It may include any one of the third mission contents (M3) that propose activities to improve psychological state through interaction.

Users can check the type (631) and main content (632) of the management content through the screen (630), and confirm the use and specific use of the management content that has already been performed through the menu (633) to check whether or not each management content is used. can be performed.

FIG. 10 illustrates an example of the internal configuration of a computing device according to an embodiment of the present invention. In the following description, descriptions of unnecessary embodiments that overlap with the description of FIGS. 1 to 9 described above will be omitted. Do this.

As shown in FIG. 10, the computing device 10000 includes at least one processor 11100, a memory 11200, a peripheral interface 11300, and an input/output subsystem ( It may include at least an I/O subsystem (11400), a power circuit (11500), and a communication circuit (11600). At this time, the computing device 10000 may correspond to a user terminal (A) connected to a tactile interface device or the computing device (B) described above.

The memory 11200 may include, for example, high-speed random access memory, magnetic disk, SRAM, DRAM, ROM, flash memory, or non-volatile memory. there is. The memory 11200 may include software modules, instruction sets, or various other data necessary for the operation of the computing device 10000.

At this time, access to the memory 11200 from other components such as the processor 11100 or the peripheral device interface 11300 may be controlled by the processor 11100.

The peripheral interface 11300 may couple input and/or output peripherals of the computing device 10000 to the processor 11100 and the memory 11200. The processor 11100 may execute a software module or set of instructions stored in the memory 11200 to perform various functions for the computing device 10000 and process data.

The input/output subsystem 11400 can couple various input/output peripheral devices to the peripheral interface 11300. For example, the input/output subsystem 11400 may include a controller for coupling peripheral devices such as a monitor, keyboard, mouse, printer, or, if necessary, a touch screen or sensor to the peripheral device interface 11300. According to another aspect, input/output peripheral devices may be coupled to the peripheral interface 11300 without going through the input/output subsystem 11400.

Power circuit 11500 may supply power to all or some of the terminal's components. For example, power circuit 11500 may include a power management system, one or more power sources such as batteries or alternating current (AC), a charging system, a power failure detection circuit, a power converter or inverter, a power status indicator, or a power source. It may contain arbitrary other components for creation, management, and distribution.

The communication circuit 11600 may enable communication with another computing device using at least one external port.

Alternatively, as described above, if necessary, the communication circuit 11600 may include an RF circuit to transmit and receive RF signals, also known as electromagnetic signals, to enable communication with other computing devices.

This embodiment of FIG. 10 is only an example of the computing device 10000, and the computing device 11000 omits some components shown in FIG. 10, further includes additional components not shown in FIG. 10, or 2. It may have a configuration or arrangement that combines more than one component. For example, a computing device for a communication terminal in a mobile environment may further include a touch screen or a sensor in addition to the components shown in FIG. 10, and may be configured to use various communication methods (WiFi, 3G, LTE) in the communication circuit 1160. , Bluetooth, NFC, Zigbee, etc.) may also include a circuit for RF communication. Components that may be included in the computing device 10000 may be implemented as hardware, software, or a combination of both hardware and software, including one or more signal processing or application-specific integrated circuits.

Methods according to embodiments of the present invention may be implemented in the form of program instructions that can be executed through various computing devices and recorded on a computer-readable medium. In particular, the program according to this embodiment may be composed of a PC-based program or a mobile terminal-specific application. The application to which the present invention is applied can be installed on a user terminal through a file provided by a file distribution system. As an example, the file distribution system may include a file transmission unit (not shown) that transmits the file according to a request from the user terminal.

The device described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), It may be implemented using one or more general-purpose or special-purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device may include multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used by any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. It can be embodied permanently or temporarily. Software may be distributed over networked computing devices and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (magneto-optical media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described with limited examples and drawings as described above, various modifications and variations can be made from the above description by those skilled in the art. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent. Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims described below.

Claims

A user psychological management system through a chatbot service based on empathic psychology implemented with a computing device including one or more processors and one or more memories that store commands executable by the processor,

It generates and provides corresponding sentences for the user's input sentences, includes response keywords that match the input keywords, which are keywords included in the user's input sentences, and uses an automatic sentence generation algorithm that generates sentences related to topics based on empathic psychology. A sympathetic sentence provision unit that generates and provides a corresponding sentence generated using;

A character representing the user's psychological state according to the user's input sentence is output on the user interface where the corresponding sentence is provided, according to the user state information including the input keyword included in the input sentence and user information in the previously input state. a character management unit that outputs a character whose appearance is transformed and allows the user's psychological state to be recognized through a user interface; and

Empathy psychology, comprising: a psychological improvement induction unit that provides the user status information to a counselor account and provides management content for improving the user's psychological status on a user interface based on the request information input from the counselor account; User psychological management system through chatbot-based service.
According to paragraph 1,

The sympathy sentence provision section,

Among the user state information, the user information in a pre-entered state includes the user's input information on menus for determining the user's psychological state provided through the user interface and the user's face input through the sensor of the user terminal. A first step of providing an initial sentence generated using the automatic sentence generation algorithm based on a response keyword matched to an initial user state keyword, which is a keyword representing the user's psychological state identified based on video and audio information;

A second step of providing a plurality of response-type sentences according to the user's input sentence for the initial sentence; and

When the number of times the user's input sentence has been received reaches a preset threshold, a third step of providing a closing sentence based on a response keyword matching the keyword corresponding to the user status information, through a process including, A user psychological management system through a chatbot service based on empathic psychology, characterized by providing an initial sentence, a response sentence, and a closing sentence as the corresponding sentences.
According to paragraph 2,

The sympathy sentence provision section,

In the first step,

It is a menu output through a user interface, and is related to the first keyword entered through a keyword selection menu where at least one of a plurality of keywords can be selected, and the text entered through an input menu that allows character input through a correlation analysis technique. A second keyword determined to be a keyword, a third keyword that is a keyword matched to an image selected through an image selection menu that outputs an image matching at least one keyword and selects some of the output images, and is input through a voice input means. An emotion analysis algorithm is used on the fourth keyword, which is a keyword derived by analyzing voice information, which is at least one of the words included in the user's voice, the tone of voice information, frequency, and speaking speed, and the user's facial image captured through video capture means. A user psychological management system through a chatbot service based on empathic psychology, characterized by deriving user status keywords using at least one of the fifth keywords, which are keywords derived as a result of analysis.
According to paragraph 1,

The sympathetic sentence provision section,

A first process for managing association information between input keywords and response keywords based on matching information input from a plurality of preset expert accounts, response keywords included in sentences already provided by the sympathetic sentence provider, and the user status Input included in the user state information according to the degree of improvement in the user's psychological state included in the user state information, using the influence figure of the response keyword on the user's psychological state derived using information on changes over time. A user psychological management system through a chatbot service based on empathy psychology, characterized in that the matching relationship between the input keyword and the response keyword is managed by at least one of the second processes that manage the correlation value between the keyword and the response keyword.
According to paragraph 1,

The sympathy sentence provision section,

A user psychology management system through a chatbot service based on empathic psychology, characterized by generating corresponding sentences using an automatic sentence generation algorithm that can reflect LSTM-based contextual meaning in word vector expressions.
According to paragraph 1,

The character management department,

Depending on the user status information, any one of the character's facial expressions, movements, colors, and accessories are modified,

The character's facial expressions, movements, color types, and types of accessories vary depending on the type of psychological state according to the user's state information, and depending on the severity of the psychological state, the size of the character's movements, color saturation or brightness, and the size of the accessories are different. A user psychological management system through a chatbot service based on empathic psychology, characterized by making one thing different.
According to paragraph 1,

The character management department,

After the sympathetic sentence is provided on the user interface by the sympathetic sentence providing unit, upon receiving the user's corresponding input sentence, the sympathetic sentence that caused the input sentence and the time the input sentence was output, or the cause of the input sentence Empathy psychology, characterized in that a character with a modified appearance is output in the area where the sympathetic sentence and the input sentence are output according to the user status information according to the sympathetic sentence that caused the input sentence and the input keyword identified by the input sentence. User psychological management system through chatbot-based service.
According to paragraph 1,

The character management department,

A user psychological management system through a chatbot service based on empathic psychology, characterized by outputting different characters according to the type of user's psychological state.
According to paragraph 1,

The psychological improvement induction department,

A user psychological management system through a chatbot service based on empathic psychology, characterized in that it provides information on measurement values and change patterns of user status information during a preset cycle to the counselor account.
According to paragraph 1,

The psychological improvement induction department,

As the management content,

At least first mission content suggesting the number of uses of the sympathetic sentence provider, second mission content suggesting use of a service related to improving psychological state online, and suggesting activities to improve psychological state through interaction with the counselor account. A user psychological management system through a chatbot service based on empathic psychology, characterized in that a menu that can use content including any one of the third mission content is provided to the user interface as the management content.