Method and apparatus for providing artificial intelligence-based personalized dialogues for aesthetic medical procedure consultation

Abstract

A customer-customized conversation provision server according to one embodiment may perform the following operations: acquiring basic information and behavioral data of a customer using a beauty medical procedure consultation platform; generating customized procedure information that the hospital can provide from the customer's basic information and behavioral data based on a large-scale language model linked to a database of a hospital providing beauty medical procedures, and providing this information to the customer; controlling the large-scale language model to dynamically generate a follow-up message to maintain the customer's interest or induce additional consultation based on interaction data including the customer's question regarding the customized procedure information and the answer of the large-scale language model, at a time when a preset time has elapsed since the customer's last response occurred; and transmitting the follow-up message to the customer through the platform and multiple channels linked to the platform.

Description

Technology Field

[0001] The present invention relates to an artificial intelligence-based customer consultation technology, and more specifically, to a technology that utilizes basic customer information and behavioral data through a large-scale language model to provide customized conversations and provides follow-up messages to maintain customer interest or induce additional consultation. Background Technology

[0002] The field of aesthetic medicine is one that provides services to improve or maintain external beauty and health by utilizing medical technologies such as skin care, cosmetic procedures, and plastic surgery. In this field, personalized services that reflect customers' individual needs and preferences are essential, and to achieve this, increasingly sophisticated and advanced technologies are required.

[0003] In particular, data-driven digital technologies are gaining attention to meet complex customer needs and provide a high-level consultation experience. For instance, there is a growing need for systems that comprehensively analyze diverse information, such as customer skin condition, health records, and beauty goals, and utilize this data in consultations. Such systems must reflect real-time changing data and provide dynamic, personalized consultations based on customer behavior and interests.

[0004] However, existing consultation systems in the field of aesthetic medicine primarily operate by passively responding to customer requests or following predefined scenarios. Consequently, these systems have limitations in that they lack the functionality to effectively handle complex customer needs, facilitate follow-up consultations, and maintain long-term customer interest. This acts as a significant obstacle to building customer trust and establishing long-term relationships.

[0005] Therefore, there is a need for a new type of consultation system that provides dynamic, customized consultation based on diverse information collected from customers. This invention presents a technical solution to meet this need, aims to overcome the limitations of existing consultation systems, and establish a new standard for consultation on cosmetic medical procedures. Prior art literature

[0006] Korean Patent Publication No. 10-2014589 (August 20, 2019) The problem to be solved

[0007] The present invention aims to provide a personalized consultation experience for cosmetic medical procedures by effectively utilizing customers' basic information and behavioral data. To this end, the present invention proposes a technology capable of analyzing customer questions and behavioral data based on a large-scale language model linked to a hospital database, and dynamically generating customized consultations and follow-up messages that reflect the customer's reactions and emotional state. This technology seeks to maximize the quality and effectiveness of the consultation by maintaining customer interest, inducing additional consultations, and providing customers with more reliable information.

[0008] Furthermore, the present invention aims to contribute to continuously maintaining and strengthening interactions with customers by providing follow-up messages through multiple channels to retain customer interest or induce further consultation. Through this, the invention seeks to implement a system capable of responding to customer needs in real time and providing a more satisfactory consultation experience.

[0009] Furthermore, the present invention aims to improve the efficiency of the system by learning various customer interaction data and continuously optimizing the timing, channel, and content of consultation content and follow-up messages.

[0010] Meanwhile, the technical problems of the present invention are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0011] A method performed by a customer-customized conversation provision server operated by a processor according to one embodiment may include: acquiring basic information and behavioral data of a customer using a beauty medical procedure consultation platform; generating customized procedure information that can be provided by the hospital from the customer's basic information and behavioral data based on a large-scale language model linked to a database of a hospital providing beauty medical procedures, and providing it to the customer; controlling the large-scale language model to dynamically generate a follow-up message to maintain the customer's interest or induce additional consultation based on interaction data including the customer's question regarding the customized procedure information and the answer of the large-scale language model at a time when a preset time has elapsed since the customer's last response occurred; and transmitting the follow-up message to the customer through the platform and multiple channels linked to the platform.

[0012] Additionally, the operation of dynamically generating the subsequent message may include, when there is no response from the customer from the time the customer's last conversation occurred until a preset time has elapsed, classifying the nature of the customer's last conversation using the large-scale language model, and controlling the large-scale language model to generate a subsequent message including customized phrases that reflect the customer's interests, emotional state, or concerns based on the interaction data.

[0013] Additionally, the operation of dynamically generating the subsequent message may include: an operation of controlling the large-scale language model to generate a subsequent message containing content that maintains customer interest or induces additional consultation when the classified last conversation is of the nature of indicating the end of a conversation; an operation of controlling the large-scale language model to generate a subsequent message containing additional explanation regarding the procedure when the classified last conversation is of the nature of a question regarding the procedure; an operation of controlling the large-scale language model to generate a subsequent message containing additional explanation regarding the price or discount benefits when the classified last conversation is of the nature of a question regarding the price; an operation of controlling the large-scale language model to generate a subsequent message containing additional information regarding the medical staff when the classified last conversation is of the nature of a question regarding the medical staff; and an operation of controlling the large-scale language model to generate a subsequent message containing additional information regarding the reservation option when the classified last conversation is of the nature of a question regarding the reservation.

[0014] Additionally, the operation of dynamically generating the subsequent message may include classifying the emotional state of the customer based on the interaction data using the large-scale language model, and controlling the large-scale language model to generate a subsequent message containing customized phrases to alleviate concerns or build trust according to the customer's emotional state.

[0015] Additionally, the operation of dynamically generating the subsequent message includes an operation of controlling the large-scale language model to generate a subsequent message containing additional information related to the behavior when there is no response from the customer until a preset time has elapsed after a preset behavior is detected from the customer, and the preset behavior may include actions regarding browsing the hospital's website, attempting to purchase services from the hospital, and attempting to schedule a consultation with the hospital. Effects of the invention

[0016] The present invention has the effect of maintaining customer interest and inducing additional consultations by providing personalized consultations on cosmetic medical procedures using customers' basic information and behavioral data.

[0017] Specifically, the present invention can significantly improve the quality and satisfaction of counseling by providing more reliable information to customers through the analysis of customer questions and interaction data based on a large-scale language model linked to a hospital database and the dynamic generation of customized messages that reflect the customer's emotional state and concerns.

[0018] In addition, the present invention enhances consultation efficiency by generating the most appropriate follow-up message based on a large-scale language model according to the customer's final response, providing it through multiple channels to diversify touchpoints with the customer, and supporting interaction in a manner preferred by the customer.

[0019] Furthermore, the present invention can enhance the flexibility and adaptability of the system by continuously learning customer behavioral data and optimizing consultation strategies. Through this, the present invention can contribute to improving the customer experience, systematizing the hospital's consultation and customer management processes, and strengthening competitiveness in the field of aesthetic medicine.

[0020] Meanwhile, the effects of the present invention are not limited to those mentioned above, and other unmentioned technical effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0021] FIG. 1 is a configuration diagram of an artificial intelligence-based customer-customized conversation provision system according to one embodiment. FIG. 2 is a configuration diagram of an artificial intelligence-based customer-customized conversation provision server according to one embodiment. FIG. 3 is a flowchart illustrating the steps of an operation performed by an artificial intelligence-based customer-customized conversation provider according to one embodiment. Specific details for implementing the invention

[0022] Detailed information regarding the purpose, technical configuration, and resulting effects of the present invention will be more clearly understood through the following detailed description based on the drawings attached to the specification of the present invention. An embodiment according to the present invention will be described in detail with reference to the attached drawings.

[0023] The embodiments disclosed herein should not be interpreted or used to limit the scope of the invention. It is obvious to those skilled in the art that the description including the embodiments herein has various applications. Accordingly, any embodiments described in the detailed description of the invention are illustrative for better explaining the invention and are not intended to limit the scope of the invention to the embodiments.

[0024] The functional blocks shown in the drawings and described below are merely examples of possible implementations. In other implementations, other functional blocks may be used without departing from the spirit and scope of the detailed description. Additionally, while one or more functional blocks of the present invention are shown as individual blocks, one or more of the functional blocks of the present invention may be a combination of various hardware and software configurations that perform the same function.

[0025] Furthermore, the expression that it includes certain components is an “open-ended” expression that merely refers to the existence of such components and should not be understood as excluding additional components.

[0026] Furthermore, when it is stated that one component is “connected” or “joined” to another component, it should be understood that while it may be directly connected or joined to that other component, there may also be other components present in between.

[0027] Hereinafter, various embodiments of the present invention are described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments and should be understood to include various modifications, equivalents, and / or alternatives of the embodiments of the present invention.

[0028] The present invention proposes a customer-customized conversation providing system (10) and a customer-customized conversation providing server (100) that implement technology to provide customized conversations by utilizing basic information and behavioral data of customers through a large-scale language model, and to provide follow-up messages to maintain customer interest or induce additional consultation.

[0029] Below, we will first examine the configuration of the artificial intelligence-based customer-customized conversation provision system (10) of the present invention.

[0030] FIG. 1 is a configuration diagram of an artificial intelligence-based customer-customized conversation provision system (10) (hereinafter referred to as 'system (10)') according to one embodiment.

[0031] Referring to FIG. 1, the system (10) may include a customer-customized conversation providing server (100) (hereinafter referred to as 'server (100)'), a customer terminal (200), and a hospital database (300).

[0032] The server (100) is a computing device that provides a beauty medical procedure consultation platform. The beauty medical procedure consultation platform (hereinafter referred to as the "platform") refers to a digital platform that connects hospitals to provide consultation regarding beauty medical procedures to customers and to provide customized procedure information based on diverse information about the customers. The server (100) can perform the role of collecting basic information and behavioral data of customers, analyzing them to generate customized procedure information for customers, or dynamically generating follow-up messages.

[0033] For example, the server (100) can provide customized consultation optimized for the customer's needs by analyzing the customer's questions, reactions, emotional state, etc. based on a large-scale language model and utilizing the latest procedure information, medical staff information, and price information in conjunction with the hospital database (300). In addition, the server (100) can manage interactions with the customer by transmitting generated consultation information and follow-up messages to the customer terminal (200) and can perform a key role in the aesthetic medical procedure consultation platform.

[0034] The customer terminal (200) is a terminal used by a customer who uses the platform provided by the server (100). The customer terminal (200) can provide an interface to access the platform provided by the server (100) to conduct consultations or to output information of the platform to the customer. The customer terminal (200) communicates with the server (100) through a web browser, a mobile application, or other digital interface, and can display customized procedure information or follow-up messages received from the server (100), or perform a relay role in transmitting customer input to the server (100).

[0035] The hospital database (300) is a database containing information about hospitals that have joined the platform. For example, the hospital database (300) may include the latest procedure information of the hospital, the professional experience and schedule information of medical staff, and price information for each procedure, and may provide such information to the server (100). The hospital database (300) may be updated regularly and may be used in real-time by linking with the server (100) during the process of generating customer-tailored procedure information.

[0036] Meanwhile, the specific configuration of the server (100) according to an embodiment of the present invention is as shown in Fig. 2 below.

[0037] FIG. 2 is a configuration diagram of a server (100) according to one embodiment.

[0038] Referring to FIG. 2, a server (100) according to one embodiment may include a memory (110), a processor (120), an input / output interface (130), and a communication interface (140).

[0039] The memory (110) can store data obtained from an external device or data generated by itself. The memory (110) can store instructions that can perform operations of the processor (120). For example, the memory (110) can store basic customer information, behavioral data, and information contained in the hospital database (300).

[0040] The processor (120) is a computing device that controls the overall operation. The processor (120) can execute instructions stored in memory (110). The operation of the server (100) of the present invention can be understood as an operation performed by each processor (120).

[0041] The input / output interface (130) may include a hardware interface or a software interface for inputting or outputting information.

[0042] The communication interface (140) enables the transmission and reception of information through a communication network. To this end, the communication interface (140) may include a wireless communication module or a wired communication module.

[0043] For example, the server (100) can be implemented as various types of devices capable of performing calculations through a processor (120) and transmitting and receiving information through a network. For example, it can be implemented in the form of a computer device, a portable communication device, a smartphone, a portable multimedia device, a laptop, a tablet PC, etc., but is not limited to these examples.

[0044] Next, when the server (100) provides platform services to the customer terminal (200), the operation performed by the server (100) is examined together with FIG. 3.

[0045] FIG. 3 is a flowchart showing the steps of an operation performed by a server (100) according to one embodiment.

[0046] Each step disclosed in FIG. 3 is merely a preferred embodiment for achieving the purpose of the present invention, and some steps may be added or deleted as needed, and any one step may be included in another step. The order of each operation disclosed in FIG. 3 is arranged for ease of understanding only, and such order is not limited to a chronological order, and the order may be changed and operated differently according to the designer's choice.

[0047] Referring to FIG. 3, in step S1010, the server (100) can obtain basic information and behavioral data of customers using the platform. For example, the basic information may include personal information such as the customer's age and gender, and aesthetic medical interests (e.g., skin care, lifting, plastic surgery procedures, etc.) that the customer has set or provided on the platform. For example, the behavioral data may include behavioral data taken by the customer on the platform, such as the customer's browsing history within the platform, history of visiting specific procedure pages, history of consultation requests, and attempts to make reservations.

[0048] The server (100) can identify the customer's preferences and interests through basic information and behavioral data, and provide customized treatment information to the customer in a subsequent stage. In addition, the server (100) can safely process the customer's sensitive information by applying security technologies such as data encryption and access control to protect the customer's privacy during the data acquisition process.

[0049] In step S1020, the server (100) can generate customized procedure information that a hospital providing cosmetic medical procedures can provide from the customer's basic information and behavioral data based on a large-scale language model linked with the hospital database (300) and provide it to the customer.

[0050] A large-scale language model is an artificial intelligence-based natural language processing model equipped with natural language processing and generation capabilities by learning large-scale text data, such as OpenAI's GPT-4, Google's PaLM 2, and Meta's LLaMA. In an embodiment of the present invention, the server (100) controls the large-scale language model to analyze the customer's basic information and behavioral data, and can generate customized procedure information by linking with the hospital database (300) or dynamically generate answers to the customer's questions.

[0051] In the description to be made below, the actions of the server (100) controlling the large-scale language model can be understood as being performed through a pre-configured prompt that includes the content described in the description of the specification, which causes the server (100) to cause the large-scale language model to perform a certain action. Such a prompt may be composed of natural language containing the content of the action described in the description of the present invention, and the server (100) may input a prompt containing the content of a specific natural language into the large-scale language model according to the situation according to the embodiment to be made below, thereby controlling the large-scale language model to take an action according to the embodiment.

[0052] The following briefly describes embodiments in which the server (100) controls a large-scale language model, and the description of specific scenarios for each embodiment will conclude the description of all steps of FIG. 3 and describe each in detail later.

[0053] For example, the server (100) can control a large-scale language model to provide answers to questions from customers. For example, if a customer asks about a specific procedure, the model can search for relevant information in the hospital database (300), summarize the information, or generate a response in a form that is easy for the customer to understand.

[0054] For example, the server (100) can control a large-scale language model based on interaction data and behavioral data, including questions and answers with the customer, to identify the customer's interests, emotional state, and preferences, and to generate appropriate customized treatment information based thereon. For example, the customized treatment information may include treatment options reflecting the customer's skin condition, beauty goals, and preferences, expected effects of the treatment, information on related medical staff, time required for the treatment, price information, etc.

[0055] For example, the server (100) inputs a hospital database (300) into a large-scale language model so that the response of the large-scale language model can reflect the latest procedure information, medical staff expertise, and price information of a specific hospital, and can control the generation of customized messages based on customer behavior data. Through this, unlike existing consultation systems that simply provide static responses, customized consultation that changes dynamically according to the customer's needs and situation can be provided.

[0056] For example, the server (100) can analyze the customer's basic information and behavioral data using a large-scale language model to match the customer's needs with the hospital's availability. For example, if a customer shows interest in a lifting procedure, the server (100) can control the generation of procedure information most suitable for the customer's needs through the large-scale language model, based on the expertise of the medical staff providing the procedure, the equipment used, and the price in the hospital database (300). In this process, the large-scale language model can generate customized procedure information by reflecting the latest information in the hospital database (300) and considering the customer's basic information, behavioral data, and interaction data. The server (100) can transmit the generated customized procedure information to the customer terminal (200) to support the customer in checking the details of the procedure and proceeding with additional consultation or reservation steps.

[0057] For example, the server (100) can recommend products or services suitable for a customer based on customer data. For instance, if the customer's basic information (e.g., allergy information) indicates a preference for cosmetics composed of ingredients that do not contain specific allergens, and behavioral data shows a history of frequently searching for a specific brand of moisturizing cream, the server (100) can recommend an allergy-free moisturizing cream or whitening serum sold at the hospital to that customer. In addition, the server can suggest a skincare kit provided by the hospital (e.g., a soothing mask pack and cleansing product set) as a customized package that reflects the customer's skin condition and interests. This function reflects the customer's needs and preferences more precisely and can contribute to increasing the hospital's additional revenue and improving customer satisfaction.

[0058] In step S1030, the server (100) can control the large language model to dynamically generate a follow-up message to maintain the customer's interest or induce additional consultation based on interaction data including the customer's question about customized procedure information and the large language model's answer, at a time when a preset time has elapsed since the customer's last response occurred.

[0059] Here, a follow-up message refers to a message generated based on customer interaction data, meaning personalized information or guidance provided to maintain customer interest or encourage further consultation. By reflecting the context of the customer's last response or question, follow-up messages can ensure the continuity of the consultation by providing the information the customer needs in a timely manner. Follow-up messages may include additional information regarding the customer's question, details of cosmetic medical procedures, expected outcomes, medical staff expertise, appointment options, or the hospital's latest promotional information.

[0060] In this process, large-scale language models can analyze customer interaction data to understand the context of the customer's last question or response, and generate customized follow-up messages that reflect the customer's emotional state or behavioral patterns.

[0061] For example, if there is no response from the customer from the time the customer's last conversation occurred until a preset time has elapsed, the server (100) can classify the nature of the customer's last conversation using a large-scale language model and control the large-scale language model to generate a subsequent message containing a customized phrase that reflects the customer's interests, emotional state, or concerns based on interaction data.

[0062] The following briefly describes embodiments in which a server (100) controls a large-scale language model to generate subsequent messages, and the description of specific scenarios for each embodiment will conclude the description of all steps of FIG. 3 and describe each in detail later.

[0063] For example, if the last conversation classified by the large-scale language model is of a nature indicating the end of the conversation, the server (100) can control the large-scale language model to generate a follow-up message containing content that maintains the customer's interest or induces further consultation.

[0064] For example, if the last conversation classified by a large-scale language model is of the nature of a question about a procedure, the server (100) can control the large-scale language model to generate a subsequent message containing additional explanation about the procedure.

[0065] For example, if the last conversation classified by the large-scale language model is of the nature of a question about price, the server (100) can control the large-scale language model to generate a subsequent message including additional explanation or discount benefits regarding said price.

[0066] For example, if the last conversation classified through a large-scale language model is of the nature of a question about medical staff, the server (100) can control the large-scale language model to generate a subsequent message containing additional information about said medical staff.

[0067] For example, if the last conversation classified by a large-scale language model is of the nature of a question about a reservation, the server (100) can control the large-scale language model to generate a subsequent message containing additional information about the reservation option.

[0068] For example, a server (100) can use a large-scale language model to classify a customer's emotional state based on interaction data and control the large-scale language model to generate a subsequent message containing customized phrases to alleviate concerns or build trust based on the customer's emotional state.

[0069] For example, if there is no response from a customer until a pre-set time has elapsed after a pre-set behavior is detected from a customer, the server (100) may control a large language model to generate a follow-up message containing additional information related to the customer's pre-set behavior. Here, the pre-set behavior may include actions such as browsing the hospital's website, attempting to purchase services from the hospital, and attempting to schedule a consultation with the hospital.

[0070] As such, the S1030 step plays a key role in maintaining continuous interaction with customers and strengthening the consultation process, contributing to the improvement of customer satisfaction and the hospital's service quality.

[0071] In step S1040, the server (100) can send a follow-up message to the customer through the platform and multiple channels linked to the platform. Here, multiple channels refer to various communication means used to maximize interaction with the customer. For example, multiple channels may include email, mobile application notifications, website pop-ups, text messages, and social media channels.

[0072] For example, the server (100) may send a follow-up message by selecting the most suitable channel based on the customer's preference, and may send the follow-up message through an alternative channel if there is no response from a specific channel. For example, if the customer has set their preference to email, the generated follow-up message is delivered first via email, and if the customer does not respond, an additional message may be sent via SMS or app notification. For example, the server (100) may generate a follow-up message containing personalized content and an appropriate tone based on the customer's last interaction data, and may include procedure information, medical staff profiles, appointment support, or promotional information in the follow-up message.

[0073] As such, the S1040 step is an important process that contributes to maintaining continuous interaction with customers, providing personalized consultation experiences, and enhancing the efficiency of hospital services.

[0074] Next, we will explain one by one the various scenarios in which the server (100) provides platform services to the customer terminal (200).

[0075] As a first scenario, a personalized consultation is provided to the customer before visiting the hospital. In this situation, it is assumed that customer A is a 30-year-old woman who is interested in skin care and is searching for information on specific skin treatments. When customer A accesses the platform, selects a desired hospital, and starts the first consultation, the server (100) analyzes the customer's basic information and interests and sends a personalized greeting message to the customer terminal (200). For example, a message such as "Hello, customer! You are interested in Ulthera. How can I help you?" can be provided as a pop-up window to attract the customer's attention.

[0076] Subsequently, proactive questions are presented based on the customer's behavioral data. For example, if the customer browses a page related to skin treatment, additional information is obtained through questions such as, "Can you tell me a little more about your current skin condition? e.g., acne, wrinkles, elasticity, etc." Using this information, the server (100) provides customized treatment information suitable for the customer's skin condition and interests and suggests scheduling a consultation. For example, a message such as, "We recommend the laser treatment most suitable for your skin condition. Would you like us to help you schedule a consultation?" is generated and sent to the customer terminal (200).

[0077] If a customer does not reply for a certain period of time, the server (100) dynamically generates and sends a follow-up message depending on the situation. For example, if a customer asks about a procedure and does not respond, a message such as "Do you have any additional questions regarding the procedure? Do you have any concerns?" can be sent after 20 minutes. If there is no response to a question about the price, a message such as "Do you have any concerns about the price? What part are you worried about?" can be provided after 15 minutes. If there is no response to an inquiry about available dates, a follow-up message such as "Is the date you want not right? What is your preferred date?" can be sent after 15 minutes to induce a response from the customer.

[0078] Additionally, if the customer's last response implies the end of the conversation, the server (100) can send an appropriate follow-up message after a certain period of time to maintain the customer's interest or induce further consultation. For example, if the customer replies, "I will think about it more and let you know," a message such as "Please feel free to ask if you have any additional questions" is provided after 1 to 2 days.

[0079] In addition, the server (100) is linked in real-time with the hospital's dashboard to reflect the latest procedure information, medical staff profiles, and price information, thereby providing accurate and reliable consultation information to the customer terminal (200). For example, it helps the customer make a choice through a message such as, "If you book laser treatment performed by our specialist medical staff member, Director Kim Jin-soo, you can currently receive a special discount."

[0080] Through such a scenario, the server (100) can effectively satisfy the interests and needs of customers and play an important role in providing a personalized consultation experience.

[0081] The following scenario involves providing customer follow-up and satisfaction management after a hospital visit. In this situation, customer B is a 45-year-old male residing in the United States who received a lifting procedure last week. The server (100) can manage customer satisfaction and suggest additional consultations or services by sending customized messages for aftercare to the customer terminal (200) based on the customer's procedure history.

[0082] First, the server (100) sends post-procedure precautions based on the customer's age and the information on the procedure received. For example, it can send a message regarding precautions for skin care after the lifting procedure or general reactions that may occur after the procedure. Afterward, when a certain amount of time has passed since the procedure (e.g., after 2 days), it can check the customer's condition through a message such as, "Hello, customer. How is your skin condition after the last lifting procedure? Do you have any additional questions or discomfort?" and, if necessary, provide additional information regarding side effects in the form of a button.

[0083] When a customer's response is confirmed, the server (100) can analyze the customer's condition using a large-scale language model and suggest appropriate follow-up measures. For example, if the customer responds that they are satisfied with their skin condition, the server suggests follow-up services by sending a message such as, "I am glad to hear that your skin condition has improved a lot! How about considering additional treatments to enhance elasticity?" In addition, it can recommend related products or services based on the customer's condition and needs, and provide a message such as, "The effect is even better if you use our hospital's specialized nutritional supplements together to enhance elasticity. Would you like me to send you more detailed information?"

[0084] Finally, the server (100) establishes a continuous relationship with the customer through messages that encourage revisiting. For example, it sends messages such as, "How about visiting again next month for regular skin care? We will help you make an appointment," and increases customer trust by reflecting the latest information linked to the hospital's dashboard. This scenario can contribute to continuing interaction with the customer even after the procedure, managing customer satisfaction, and maximizing the hospital's service value.

[0085] The following scenario involves providing support for customized marketing campaigns. In this situation, it is assumed that a hospital aims to launch a new cosmetic procedure (e.g., the latest laser procedure) and promote it to existing customers. Based on data linked to the hospital's dashboard, the server (100) uses a large-scale language model to analyze customer data and identify customers who are likely to be interested in the new procedure. For example, target customer groups are selected based on customers who have received skin treatment in the past, or specific age groups and genders.

[0086] Personalized messages are sent to selected customers, and these messages are created by reflecting the latest procedure information, medical staff information, prices, etc., on the hospital's dashboard. For example, a message such as "Hello, customer! We would like to introduce [Latest Laser Procedure Name], newly launched by our hospital. Experience this procedure optimized for your skin condition." is sent to attract the customer's attention. When a customer responds to the message, the server (100) monitors the response in real time and induces interaction by providing additional information such as "Please click for more detailed information" or "Book a consultation now for a special discount."

[0087] Finally, the server (100) analyzes the campaign results to collect data such as which messages were effective and which customer groups showed a high response. Based on this data, the large-scale language model can improve future marketing strategies and design optimized campaigns. This scenario can contribute to maximizing the hospital's customer targeting and marketing efficiency, and strengthening the bond with customers.

[0088] In the following scenario, we will explain efficient management through integration with the integrated customer management system.

[0089] In this scenario, it is assumed that the hospital manages customer data using a Customer Relationship Management (CRM) system, and that the server (100) performs more effective customer management by interacting with this system through a large-scale language model. For example, customer D, who lives in Germany, has received two procedures in the past six months and has a history of purchasing specific beauty products.

[0090] The server (100) analyzes customer data linked to a CRM system in real time to comprehensively identify the customer's past visit records, treatment history, purchase history, etc. Based on this, it can provide personalized interactions to the customer. For example, it checks the customer's satisfaction and condition by sending a message such as, "Customer, has your skin condition improved a lot since the last treatment? Please let us know if you need additional care."

[0091] In addition, the server (100) utilizes automated reminders and notification functions to help customers avoid missing important appointment dates. For example, it can encourage customers to make appointments through notification messages such as, "Your next procedure appointment is coming up soon. If you would like to make an appointment, please book right now." These reminders contribute to efficiently managing customers' schedules and increasing the operational efficiency of the hospital.

[0092] After the procedure is completed, the server (100) sends a feedback request message through the customer terminal (200), and the collected feedback is automatically reflected in the CRM system. Through this, the hospital can analyze customer satisfaction and continuously improve service quality. For example, customer feedback is elicited through a message such as, "Please provide a satisfaction survey after the procedure. Your valuable feedback is of great help in improving services."

[0093] This type of CRM integration scenario systematizes the interaction between the hospital and customers and provides personalized services, thereby increasing customer satisfaction while significantly improving the hospital's management efficiency.

[0094] As a next scenario, a preemptive response can be given when a customer does not respond to an initial message. In this case, the server (100) dynamically generates a follow-up message when there is no response from the customer for a certain period of time to maintain the customer's interest and continue the consultation.

[0095] For example, if a customer leaves a question about a procedure and does not respond for 20 minutes, the server (100) sends a follow-up message saying, "Hello, customer. Do you have any further questions about the procedure? Do you want to know more about the procedure process or effects?" to induce the customer to ask further questions. If there is no response after 1 hour, a personalized consultation is suggested through a message saying, "I would like to recommend the optimal procedure that suits your skin condition and goals. Please feel free to contact me anytime!" If there is no response even by the morning of the next day, a message saying, "Hello, customer. If you would like more detailed information about the procedure you inquired about yesterday, please contact me anytime. I will do my best to ensure your satisfaction," is sent to continuously express interest and induce a consultation appointment.

[0096] This scenario can help maintain customer interest and support the more systematic operation of the hospital's consultation process by providing follow-up messages at appropriate intervals and tones, even if the customer does not respond after the initial inquiry.

[0097] In the following scenario, we will explain how to handle the case where a customer inquires about a price and does not receive a response within a certain period of time.

[0098] In this scenario, it is assumed that a customer inquires about the price of a specific procedure through the platform and does not respond for a certain period of time. In such cases, the server (100) sends a follow-up message at an appropriate time to maintain the customer's interest and induce further interaction.

[0099] The first follow-up message is sent 15 minutes after the customer's price inquiry. This message includes the text, "Hello, customer. Do you have any further questions about pricing or would you like to know about discount benefits?" It encourages the customer to inquire about additional pricing information or the hospital's discount offers. This provides a link to prevent the customer from stopping the consultation after simply checking the price.

[0100] A second follow-up message is sent after two hours and includes the content, "We offer various treatment options to fit your budget. Would you like us to help you find a suitable treatment?" The purpose of this message is to help the customer explore treatments that suit their situation by suggesting various options based on their budget.

[0101] The third follow-up message is sent the following afternoon and includes the statement, "Our hospital offers a special discount to first-time visitors. If you are interested, we will send you the details." This message focuses on re-engaging the customer's attention by highlighting the discount benefit and encouraging further interaction with the hospital.

[0102] This scenario encourages continued customer interaction with the platform by providing appropriate follow-up messages at timed intervals to maintain sustained interest. Additionally, it supports customers in making decisions tailored to their needs through various treatment options and discount benefits, while enabling the efficient operation of the hospital's consultation process.

[0103] In the following scenario, we will explain the case where a customer inquires about available reservation dates and receives no response within a certain period of time.

[0104] In this scenario, it is assumed that a customer inquires about available dates for a specific procedure through the platform and does not respond for a certain period of time. In such cases, the server (100) sends an appropriate follow-up message to maintain interaction with the customer and continue the booking process.

[0105] The first follow-up message is sent 15 minutes after the customer inquires about a reservation date. This message includes the content, "Hello, customer. Do you need help finding the optimal time for your desired reservation date?" and supports the customer in proceeding with the reservation smoothly. This helps the customer coordinate a reservation time that fits their schedule.

[0106] A second follow-up message is sent after one hour and includes the content, "We offer various reservation options for your convenience. Please let us know your preferred date and time, and we will assist you." This message emphasizes customer convenience and suggests various options to help the customer match their schedule with the hospital's availability.

[0107] A third follow-up message is sent the following morning and includes the statement, "Please let us know at any time if you need to adjust or change your appointment date. We will respond flexibly to fit your schedule." This message informs customers that appointments can be adjusted or changed according to their circumstances, providing an environment where customers can comfortably interact with the hospital.

[0108] This scenario structures the reservation process around the customer to encourage bookings that fit their schedules and induces re-engagement by expressing continuous interest and support even in the absence of a response. Through this, the hospital can efficiently manage reservation interactions with customers and contribute to increasing the success rate of consultation and procedure appointments.

[0109] In the following scenario, we will explain the case where there is no response within a certain period of time after a customer asks a question about the medical staff.

[0110] In this scenario, it is assumed that a customer inquires about the hospital's medical staff through the platform and does not respond for a certain period of time. In such cases, the server (100) generates and sends an appropriate follow-up message to maintain the customer's interest, build trust, and induce a consultation appointment.

[0111] The first follow-up message is sent 15 minutes after the customer asks the medical staff. This message includes the content, "Hello, customer. Is there anything else you would like to know about our medical team? We can explain their expertise and experience in detail." This message is designed to gain the customer's trust by emphasizing the medical staff's expertise and experience, and to encourage further questions.

[0112] A second follow-up message is sent after two hours and includes the message, "I would like to recommend the medical professional best suited for your skin condition. Please feel free to contact us at any time if you need additional information." This message helps the customer continue the consultation by recommending a medical professional tailored to their individual condition and needs.

[0113] The third follow-up message is sent the following afternoon and includes the content, "We will do our best to help you achieve your skin goals through a consultation with our medical team. Please let us know if you would like to schedule a consultation." This message strengthens the connection with the customer, generates customer interest based on trust with the medical staff, and encourages scheduling a consultation.

[0114] This scenario supports the efficient operation of the hospital's consultation process and builds customer trust by providing contextually appropriate follow-up messages at appropriate intervals, even when customers do not respond to inquiries about medical staff. This enables the hospital to maintain continuous interaction with customers and effectively connect its services.

[0115] In the following scenario, we will explain the case where there is no response after a customer performs a specific action (e.g., browsing the website, attempting to purchase a product, attempting to schedule a consultation, etc.).

[0116] In this scenario, it is assumed that a customer performs a specific action within the platform and does not engage in further interaction. In such cases, the server (100) generates preemptive questions and follow-up messages based on the customer's behavior data to maintain the customer's interest and induce interaction.

[0117] First, if there is no response after the customer browses the website, the server (100) sends a message after 10 minutes saying, "Hello, customer. Would you like to know more about a specific procedure on our website? We can assist you with additional information or consultation." This message provides additional information about the procedure of interest based on the customer's browsing history and encourages the customer to request a consultation.

[0118] Next, if there is no response after the customer attempts to purchase a product, the server (100) sends a message after 5 minutes saying, "Customer, do you have any questions about the products you have added to your shopping cart? We will help you with your purchase." This message provides additional information to help the customer make a purchase decision and supports the customer in completing the purchase process.

[0119] Finally, if there is no response after the customer attempts to make a consultation appointment, the server (100) sends a message after 10 minutes saying, "Do you need help with the consultation appointment process? We will coordinate the date and time you want." This message guides the customer to proceed smoothly with the appointment process and prioritizes the customer's convenience.

[0120] This scenario helps customers maintain their connection with the platform by analyzing customer behavioral data in real time and providing proactive interactions. This enhances the customer experience, improves accessibility to hospital services, and supports seamless interaction between customers and the hospital.

[0121] In the following scenario, we will explain the response strategy for the case where a customer does not respond to multiple follow-up messages.

[0122] In this scenario, it is assumed that the customer does not respond to subsequent messages sent multiple times. The server (100) sends appropriate messages at regular intervals to maintain interaction with the customer, and finally provides a message that encourages re-engagement while respecting the customer's decision.

[0123] The first follow-up message is sent after 20 minutes. This message includes the content "Hello, customer. Do you have any further questions about the procedure? Please feel free to contact us at any time," and gently encourages interaction to allow the customer to continue the consultation.

[0124] A second follow-up message is sent after one hour has passed, and it re-engages the customer's interest with the message, "If you need help achieving your skin goals, we are happy to assist you at any time. Please let us know if you would like to schedule a consultation," and suggests specific actions (e.g., scheduling a consultation).

[0125] The third follow-up message is sent the following morning and includes the content, "Thank you for your interest. Please feel free to contact us at any time if you have any further questions or need consultation. We are happy to assist you." This message expresses gratitude to the customer and maintains an open attitude toward participating in the platform.

[0126] The final message is sent after 3 days and contains the content, "Hello, customer. Thank you for your interest in our service. Please feel free to contact us anytime you need us in the future. Have a nice day!" This message is intended to respect the customer's decision while encouraging them to revisit the platform or use the service in the future.

[0127] This scenario focuses on maintaining a positive customer experience by continuing interactions in a polite and consistent manner, even in the absence of a direct response. This provides customers with the confidence to connect with the hospital whenever needed, while highlighting the platform's flexibility and accessibility.

[0128] In the following scenario, we will explain a strategy to send follow-up messages using other channels when a customer does not respond via a specific channel (e.g., email, chat, mobile app notifications).

[0129] This scenario deals with a method of sending follow-up messages using various channels to maintain interaction with a customer. If there is no response from the channel initially selected by the customer, the server (100) continuously maintains contact with the customer by sending follow-up messages through another channel.

[0130] First, if a customer leaves an inquiry via email and does not respond for one hour, the server (100) sends a message saying, "Hello, customer. If you need additional information about the procedure you inquired about via email, please contact us at any time." This message encourages the customer to continue the conversation via email and may suggest other channels as needed.

[0131] Next, if a customer starts a real-time consultation via chat but does not respond for 15 minutes, the server (100) sends a message saying, "Customer, you can start a real-time consultation via chat at any time. Would you like me to help?" This message encourages the customer to recognize the convenience of real-time consultation again and resume the consultation.

[0132] In addition, if a customer leaves an inquiry via a mobile app notification and does not respond for 30 minutes, the server (100) sends a message saying, "Hello, customer. Please let us know if you need additional information about the procedure you inquired about through the app." This message helps maintain the customer's interest by providing additional information suitable for the app usage environment.

[0133] This scenario focuses on flexibly switching customer interactions to other channels to continue the relationship, even if there is no response from a specific channel. This allows for the provision of appropriate communication methods tailored to customer preferences and situations, thereby improving hospital service accessibility and enhancing customer satisfaction.

[0134] In the following scenario, we will explain customized responses based on the customer's emotional state.

[0135] This scenario describes how, when a customer expresses dissatisfaction or concern in a previous interaction, the server (100) analyzes the customer's emotional state through a large-scale language model and responds appropriately. This allows for the restoration of the customer's trust and the provision of a positive counseling experience.

[0136] First, the server (100) detects an emotional state by analyzing the customer's message using natural language processing (NLP) technology. For example, if the customer inputs "I think the results of the procedure will not meet my expectations," a large-scale language model identifies the emotions expressing worry and anxiety in the message.

[0137] Subsequently, generate a message that immediately acknowledges and comforts the customer's emotions. For example, build trust by empathizing with the customer's feelings through a message such as, "Hello, customer. I understand your concerns regarding the treatment results. How can I help you achieve better results?"

[0138] Next, it includes proactive suggestions to resolve the customer's problem. For example, it supports the customer in resolving the problem by sending a message such as, "I would like to help resolve your concerns through additional consultation. Would you like me to help you schedule a consultation at a time convenient for you?"

[0139] Finally, alleviate customer anxiety by providing additional support. For example, ensure customers receive the necessary information through a message such as, "We can also provide detailed guidance on post-procedure care. We can send you relevant materials if you wish."

[0140] This scenario focuses on strengthening trust between customers and the hospital and providing a positive experience by carefully analyzing the customer's emotional state and offering customized responses accordingly. Through this, the hospital can enhance customer satisfaction and contribute to maintaining long-term relationships.

[0141] In the following scenario, we will explain how to encourage the re-engagement of long-term inactive customers.

[0142] This scenario describes a process that detects when a customer has not interacted with the platform for a certain period and uses a large-scale language model to send appropriate messages to encourage re-engagement. Through this, the hospital can restore relationships with existing customers and encourage repeat visits.

[0143] First, the server (100) identifies customers who have not responded for more than 3 months since their last interaction through data analysis. Based on this, a large-scale language model detects the customer's inactivity and begins sending customized follow-up messages.

[0144] The first follow-up message is sent three months after inactivity is detected. This message includes the content, "Hello, customer! It has been a while. I would like to inform you about a new procedure recently introduced at our clinic," and aims to attract the customer's attention by providing new information.

[0145] The second follow-up message is sent one week after the first message is sent. This message includes the content, "I would like to recommend a customized treatment tailored to your skin condition. If you are interested, I can help you schedule a consultation," and encourages the customer's re-engagement through a personalized consultation.

[0146] The third follow-up message is sent two weeks after the second message is sent. This message includes the content, "Our hospital offers a special discount to returning customers. Why not take this opportunity to visit us?" and actively encourages customer re-engagement by emphasizing the discount benefits.

[0147] The final follow-up message is sent one month after the third message is sent. This message includes the content, "We are always doing our best for your health and beauty. Please contact us whenever you need us," and respects the customer's decision while encouraging future re-participation.

[0148] This scenario focuses on restoring relationships with customers through messages designed in stages to revive the interest of inactive customers. Through this, the hospital can retain existing customers and improve return visit rates based on long-term trust.

[0149] In the following scenario, we will explain an example of inducing customer participation through the provision of educational content.

[0150] This scenario describes a method in which, when a customer has shown interest in a specific procedure but has not yet made a decision, the server (100) provides educational content through a large-scale language model to help the customer understand and encourage participation. Through this, the customer can obtain in-depth information about the procedure and make a decision based on trust.

[0151] First, the server (100) analyzes customer behavior data to recognize situations where educational content is needed. For example, if a customer visits a specific procedure page or asks a question about the procedure but does not make an immediate decision, the large language model determines that the customer needs additional information.

[0152] The first educational message is sent 30 minutes after the customer's behavior is detected. This message includes the content, "Hello, customer. Would you like to know more about [Procedure Name]? We will send you materials regarding the procedure and expected effects." By providing specific information about the procedure, this message enhances the customer's understanding and offers the background knowledge necessary to make a decision about the treatment.

[0153] Next, additional educational materials are provided one day after the initial message is sent. This message includes the content, "Dear customer, please check out the success stories of [Procedure Name]. You can understand the effectiveness of the procedure through the experiences of actual customers. [Link]," thereby reinforcing the credibility of the procedure through these success stories. Through this material, customers can understand the value provided by the procedure based on actual cases.

[0154] Finally, an interaction-inducing message is sent two days after additional materials are provided. This message includes the content, "Have you reviewed the materials? Please feel free to contact us at any time if you have any further questions. You can also schedule a consultation." This message encourages customers to ask questions or schedule a consultation based on the provided materials, thereby strengthening interaction with the platform.

[0155] This scenario is designed to go beyond simply providing information to customers; it aims to aid understanding, build trust, and ultimately facilitate consultation bookings. Through this, the hospital actively supports customers in making their own decisions and further strengthens the relationship with them.

[0156] In the following scenario, we will explain a scenario that provides cross-selling and upselling strategies through a recommendation system.

[0157] This scenario describes a method for increasing the hospital's revenue by having the server (100) recommend related additional procedures or products through a large-scale language model when a customer books a specific procedure or purchases a product. Through this, the customer is presented with additional options that meet their needs and goals, and the hospital can simultaneously improve revenue and customer satisfaction.

[0158] First, the server (100) analyzes data after a customer's treatment reservation or purchase. This data includes the customer's skin condition, the scheduled treatment, the type of product purchased, etc., and a large-scale language model identifies suitable additional treatments or products based on this.

[0159] The first recommendation message is sent one day after the customer books the procedure. This message includes the content, "Hello, customer! Thank you for helping us book [Procedure Name]. We recommend [Additional Procedure Name] for post-procedure skin care." This message induces upselling by suggesting an additional procedure that complements the treatment the customer has already booked.

[0160] The second recommendation message is sent three days after the customer completes the procedure. This message includes the content, "Try [Related Product Name] for post-procedure care. We are offering a special discount to you. Are you interested?" and recommends products necessary to maintain or enhance the effects of the procedure. This message focuses on increasing the hospital's sales of supplementary products through cross-selling.

[0161] Additionally, a third message designed to engage the customer is sent one week after the completion of the procedure. This message includes the statement, "We can recommend additional customized products tailored to your skin condition. Please let us know if you would like to learn more." This message maintains continuous interaction with the customer and simultaneously pursues increased sales and customer satisfaction through recommendations for additional products or procedures.

[0162] This scenario strengthens the hospital's cross-selling and upselling strategies by providing personalized recommendation messages based on customer reservation and purchase data. This enables customers to make choices better suited to their skin goals, while allowing the hospital to strengthen relationships with customers and maximize revenue.

[0163] In the following scenario, we will describe a scenario that provides a response induction strategy through multi-channel integrated follow-up messages.

[0164] This scenario describes a method in which, if a customer does not respond on a specific channel (e.g., email, SMS, app notification, social media), the server (100) induces a response from the customer by sending a follow-up message through another channel using a large-scale language model. Through this, the hospital can maintain interaction with the customer and provide a suitable communication method considering the customer's preferences and convenience.

[0165] First, the server (100) analyzes the customer's preferred channel and sets the priority. For example, if the customer is set to prefer email, email is designated as the default channel, and if there is no response, other channels such as SMS, app notifications, and social media are used sequentially.

[0166] The first follow-up message is sent if there is no response from the primary channel. For example, if email is set as the primary channel, a message stating, "Hello, customer. Would you like to know more about the procedure? Please feel free to contact us via email at any time," is sent to elicit a response from the primary channel.

[0167] If there is no response even in the primary channel, the server (100) uses another channel to send a follow-up message. For example, it sends a message via SMS saying, "Customer, we can also assist you with consultation via SMS. Please choose the method that is most convenient for you." to enable the customer to choose a more convenient channel.

[0168] If there is no response from any channel, the server (100) sends a final follow-up message. This message includes the content, "We provide consultation through various channels for your convenience. Please contact us whenever you need us," and shows an attitude that respects the customer's decision while emphasizing the accessibility of the hospital.

[0169] This scenario prioritizes customer convenience through a multi-channel integration strategy and provides flexible and comprehensive communication methods to maintain continuous interaction with customers. Through this, the hospital can effectively meet diverse customer preferences and improve accessibility to consultations and services.

[0170] In the following scenario, we will explain proactive proposals based on customer behavior prediction.

[0171] This scenario describes a method in which a large-scale language model analyzes customers' past behavioral data to predict procedures or products they are likely to need, and proactively suggests them based on this analysis. Through this, hospitals can identify customer needs in advance and provide customized services, thereby simultaneously increasing customer satisfaction and revenue.

[0172] First, the server (100) applies the customer's behavioral data to a large-scale language model to comprehensively analyze the customer's previous treatment history, purchase patterns, website browsing history, etc., and predicts the treatment or product that is likely to be needed next. For example, by analyzing the data of a customer who has received a lifting treatment to improve skin elasticity, it can recommend related products that can additionally enhance moisturizing effects.

[0173] Personalized suggestion messages are sent at the anticipated time of need. These messages include content such as, "Hello, customer! Did your skin elasticity improve after the last [Procedure Name]? We would like to recommend [related product / procedure name] as well. If you are interested, we can provide more details," and they attract customer interest by suggesting customized options based on the customer's anticipated needs.

[0174] In addition, additional information is provided to encourage interaction. For example, a message such as, "For better results, using [related product / procedure name] together will enhance the effect. Please let us know if you would like more information," helps customers make decisions that meet their needs.

[0175] This scenario focuses on personalizing the customer experience and maximizing the value of the hospital's services by utilizing AI-based behavioral prediction to proactively identify customer needs and provide appropriate suggestions when they are required. Through this, the hospital can strengthen trust with customers and enhance their long-term satisfaction.

[0176] In the following scenario, we will explain a customized proposal scenario based on seasons or events.

[0177] This scenario deals with a method to increase the hospital's revenue by recommending suitable procedures or products to customers tailored to specific seasons or events (e.g., summer, Christmas, Valentine's Day, etc.), thereby attracting their interest.

[0178] First, the server (100) analyzes data on the current season or upcoming events to prepare customized suggestions related to the customer's needs. For example, during the summer, it may recommend treatments or products to moisturize the skin and prevent sun damage, and during the Christmas season, it may suggest beauty treatments for special events or outings.

[0179] The first personalized offer message is sent at the start of a season or event. This message includes the content, "Hello, customer! To celebrate summer, we are offering a special discount on [specific treatment / product name]. Don't miss out on this opportunity!" and encourages customers to discover services that meet their seasonal needs.

[0180] If an event is in progress, the server (100) sends a message in the middle of the event informing of additional benefits. For example, a message such as "This summer promotion is ending soon. If you book [procedure / product name], you will receive additional benefits. Book now!" can create a sense of urgency for the customer and induce a quick decision.

[0181] After the event concludes, a thank-you message is sent to customers to maintain a positive relationship. This message includes the message, "Thank you for participating in this summer promotion. We will continue to do our best for your health and beauty," reinforcing trust with customers and encouraging future participation.

[0182] This scenario focuses on attracting customer interest based on seasons or events, personalizing the customer experience through timely offers, and simultaneously improving the hospital's revenue and engagement. Through this, customers can discover services tailored to their situations and needs, while the hospital can strengthen long-term relationships with customers.

[0183] In the following scenario, we will explain a scenario for providing special care and benefits to VIP customers.

[0184] This scenario deals with how a hospital identifies VIP customers who meet specific criteria and provides them with special care and benefits to increase customer loyalty and strengthen long-term relationships.

[0185] First, the server (100) analyzes customer data to identify VIP customers. Customers with a high purchase history, those who frequently request consultations, and those who have maintained long-term interaction with the hospital may be classified as VIPs.

[0186] A first welcome message is sent to customers identified as VIPs. This message includes the content, "Hello, customer! We would like to offer special benefits to our VIP customers. Please check out the exclusive information on new procedures," acknowledging the value of the customer and allowing them to feel a special sense of worth by providing VIP-exclusive benefits.

[0187] Subsequently, messages offering VIP-exclusive benefits are sent regularly. For example, a message stating, "As part of this month's VIP-exclusive event, we are offering [Procedure Name] at a special discounted price. If you would like to make a reservation, we can assist you right now," allows customers to continuously experience the benefits of being a VIP and encourages re-participation.

[0188] In addition, personalized consultation and support are provided to VIP customers. For example, through a message such as, "We recommend customized treatments tailored to your skin condition. Please let us know anytime if you would like a private consultation," individual needs are met, and deeper trust is built.

[0189] Finally, exclusive content is provided for VIP customers only. This message includes the statement, "We are sending you the latest beauty trends and treatment information prepared for our VIP customers. Please check it out if you are interested," ensuring that customers feel a continuous connection with the hospital.

[0190] This scenario contributes to strengthening customer loyalty, maintaining the hospital's luxurious image, and forming long-term relationships by providing differentiated services to VIP customers. Through this, customers feel that they are receiving special treatment, and the hospital can achieve continuous engagement and high satisfaction.

[0191] In the following scenario, we will explain an emergency situation or a scenario for providing emergency assistance.

[0192] This scenario describes how, when a customer reports side effects or urgent problems after a procedure, the server (100) quickly detects and responds to the problem through a large-scale language model to provide support to the customer. This ensures customer safety and strengthens trust in the hospital.

[0193] First, the server (100) analyzes the customer's message to detect an emergency situation. A large-scale language model uses natural language processing to detect words indicating urgency in the message, such as "side effects," "urgent," and "pain," and determines that the customer's situation is an emergency.

[0194] When an emergency is detected, the large-scale language model sends an immediate response message. This message includes the content, "Hello, customer. We apologize for the inconvenience. We will immediately connect you with medical staff to assist you," providing the customer with confidence that they can receive immediate help.

[0195] Subsequently, the large-scale language model connects customers with medical professionals or provides additional support. For example, it facilitates rapid communication between customers and medical staff through messages such as, "Medical professionals will be contacting you soon. Please let us know if you need any further information," and takes measures to resolve the customer's emergency situation.

[0196] Afterward, the customer's condition is monitored and feedback is collected. For example, the customer's status is continuously checked through messages such as, "How are you feeling after your recent procedure? Please let us know anytime if you need further assistance," thereby securing the information necessary to improve the hospital's service quality.

[0197] This scenario focuses on prioritizing customer safety by responding quickly and effectively to customers in emergency situations, while emphasizing the hospital's credibility and accountability. Through this, the hospital can enhance customer satisfaction and provide a high level of service even in emergencies.

[0198] According to the above-described embodiment, the present invention has the effect of maintaining customer interest and inducing additional consultation by providing personalized cosmetic medical procedure consultations utilizing customers' basic information and behavioral data.

[0199] Specifically, the present invention can significantly improve the quality and satisfaction of counseling by providing more reliable information to customers through the analysis of customer questions and interaction data based on a large-scale language model linked to a hospital database and the dynamic generation of customized messages that reflect the customer's emotional state and concerns.

[0200] In addition, the present invention enhances consultation efficiency by generating the most appropriate follow-up message based on a large-scale language model according to the customer's final response, providing it through multiple channels to diversify touchpoints with the customer, and supporting interaction in a manner preferred by the customer.

[0201] Furthermore, the present invention can enhance the flexibility and adaptability of the system by continuously learning customer behavioral data and optimizing consultation strategies. Through this, the present invention can contribute to improving the customer experience, systematizing the hospital's consultation and customer management processes, and strengthening competitiveness in the field of aesthetic medicine.

[0202] The various embodiments of this document and the terms used therein are not intended to limit the technical features described in this document to specific embodiments, and should be understood to include various modifications, equivalents, or substitutions of said embodiments. In connection with the description of the drawings, similar reference numerals may be used for similar or related components. The singular form of a noun corresponding to an item may include one or more items unless the relevant context clearly indicates otherwise.

[0203] In this document, each of the phrases such as “A or B,” “at least one of A and B,” “at least one of A or B,” “A, B or C,” “at least one of A, B and C,” and “at least one of A, B, or C” may include all possible combinations of items listed together in the corresponding phrase. Terms such as “1,” “2,” or “first” or “second” may be used simply to distinguish a component from another component and do not limit the components in any other aspect (e.g., importance or order). Where any (e.g., 1st) component is referred to as “coupled” or “connected” to another (e.g., 2nd) component, with or without the terms “functionally” or “communicationly,” it means that the component may be connected to the other component directly (e.g., wired), wirelessly, or through a third component.

[0204] As used in this document, the term "module" may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be a component formed integrally, or a minimum unit of a component or part thereof that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0205] Various embodiments of this document may be implemented as software (e.g., a program) comprising one or more instructions stored in a storage medium (e.g., memory) that can be read by a device (e.g., an electronic device). The storage medium may include random access memory (RAM), a memory buffer, a hard drive, a database, erasable programmable read-only memory (EPROM), electrically erasable read-only memory (EEPROM), read-only memory (ROM), and / or the like.

[0206] Additionally, the processor of the embodiments of the present invention may call at least one instruction among one or more instructions stored from a storage medium and execute it. This enables the device to be operated to perform at least one function according to at least one called instruction. Such one or more instructions may include code generated by a compiler or code that can be executed by an interpreter. The processor may be a general-purpose processor, a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), and / or the like.

[0207] A device-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' simply means that the storage medium is a tangible device and does not contain signals (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily.

[0208] Methods according to the various embodiments disclosed in this document may be provided as part of a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created on a device-readable storage medium, such as a manufacturer's server, an application store's server, or the server's memory.

[0209] According to various embodiments, each component (e.g., module or program) of the described components may include a singular or multiple entities. According to various embodiments, one or more of the components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Generally or additionally, multiple components (e.g., module or program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the components of the multiple components in the same or similar manner as they were performed by the corresponding component among the multiple components prior to integration. According to various embodiments, operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically; one or more of the operations may be executed in a different order; omitted; or one or more other operations may be added. Explanation of the symbols

[0210] 10: System 100: Server 200: Customer terminal 300: Hospital Database 110: Memory 120: Processor 130: Input / Output Interface 140: Communication interface

Claims

Claim 1 A method performed by a customer-customized conversation provider server operated by a processor, comprising: acquiring basic information and behavioral data of a customer using a beauty medical procedure consultation platform; generating customized procedure information that the hospital can provide from the customer's basic information and behavioral data based on a large-scale language model linked to a database of a hospital providing beauty medical procedures, and providing the information to the customer; controlling the large-scale language model to dynamically generate a follow-up message to maintain the customer's interest or induce additional consultation based on interaction data including the customer's question regarding the customized procedure information and the answer of the large-scale language model, at a time when a preset time has elapsed since the customer's last response occurred; and transmitting the follow-up message to the customer through the platform and multiple channels linked to the platform. Claim 2 A method according to claim 1, wherein the operation of dynamically generating the subsequent message includes, when there is no response from the customer from the time the customer's last conversation occurred until a preset time has elapsed, classifying the nature of the customer's last conversation using the large-scale language model, and controlling the large-scale language model to generate a subsequent message including a customized phrase reflecting the customer's interests, emotional state, or concerns based on the interaction data. Claim 3 In paragraph 2, the operation of dynamically generating the subsequent message comprises: an operation of controlling the large-scale language model to generate a subsequent message containing content that maintains customer interest or induces additional consultation when the classified last conversation is of a nature indicating the end of the conversation; an operation of controlling the large-scale language model to generate a subsequent message containing additional explanation of the procedure when the classified last conversation is of a nature questioning about a procedure; an operation of controlling the large-scale language model to generate a subsequent message containing additional explanation of the price or discount benefits when the classified last conversation is of a nature questioning about the price; an operation of controlling the large-scale language model to generate a subsequent message containing additional information about the medical staff when the classified last conversation is of a nature questioning about the medical staff; and an operation of controlling the large-scale language model to generate a subsequent message containing additional information about the reservation options when the classified last conversation is of a nature questioning about a reservation. Claim 4 A method according to paragraph 3, wherein the operation of dynamically generating the subsequent message includes the operation of classifying the emotional state of the customer based on the interaction data using the large-scale language model, and controlling the large-scale language model to generate a subsequent message including customized phrases to alleviate concerns or build trust according to the emotional state of the customer. Claim 5 In claim 4, the operation of dynamically generating the subsequent message includes the operation of controlling the large-scale language model to generate a subsequent message containing additional information related to the behavior when the customer's response is absent until a pre-set time has elapsed after a pre-set behavior is detected from the customer, and the pre-set behavior includes behaviors regarding browsing the hospital's website, attempting to purchase services for the hospital, and attempting to schedule a consultation for the hospital.

Images