Customer service assistance device, customer service assistance method, and recording medium

Abstract

This customer service assistance device includes: an acquisition means for acquiring diagnosis results including a diagnosis image of the state of the skin of a customer and customer service history information related to the customer's skin; a generation means for generating customer service assistance information by inputting the diagnosis results and the customer service history information into a trained model capable of outputting customer service assistance information; and an output means for outputting the generated customer service assistance information.

Description

Customer service support device, customer service support method, and recording medium

[0001] The present disclosure relates to a customer service assistance device, a customer service assistance method, and a recording medium.

[0002] When selling cosmetics, the customer's skin condition may be diagnosed.

[0003] Patent document 1 describes a skin image analysis method that uses a trained discrimination model that has been machine-learned based on multiple training data including multiple combinations of correct answer information indicating common attributes or conditions of the subject skin of a training skin image and a group of training skin patch images of a predetermined image size extracted from the training skin image, and inputs each skin patch image processed from an evaluation skin image into the discrimination model to obtain index values ​​for the skin attributes or skin conditions for each skin patch image.

[0004] Japanese Patent Application Laid-Open No. 2022-078936

[0005] Incidentally, when selling cosmetics, a salesperson such as a beauty advisor (BA) may provide customer service by recommending products while referring to the diagnosis results of the customer's skin condition.

[0006] However, there is a wide range of information to be conveyed to customers, including skin condition diagnosis results, recommended products based on those diagnosis results, and information to be conveyed to customers to increase their desire to purchase the recommended products, making it difficult for store clerks to provide appropriate customer service using the skin condition diagnosis results.

[0007] An example of an objective of the present disclosure is to provide a technology that supports customer service using skin condition diagnosis results when selling cosmetics.

[0008] A customer service support device in one aspect of the present disclosure includes an acquisition means for acquiring a diagnosis result including a diagnostic image of the customer's skin condition and customer service history information related to the customer's skin, a generation means for generating customer service support information by inputting the diagnosis result and the customer service history information into a trained model capable of outputting customer service support information, and an output means for outputting the generated customer service support information.

[0009] A customer service support method in one aspect of the present disclosure obtains a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin, generates customer service support information by inputting the diagnosis result and customer service history information into a trained model capable of outputting customer service support information, and outputs the generated customer service support information.

[0010] A program in one aspect of the present disclosure causes a computer to execute a process of acquiring a diagnostic result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin, generating customer service support information by inputting the diagnostic result and customer service history information into a trained model capable of outputting customer service support information, and outputting the generated customer service support information.

[0011] Each program may be stored in a non-transitory computer-readable recording medium.

[0012] One example of the effect of the present disclosure is that it is possible to support customer service using the results of skin condition diagnosis when selling cosmetics.

[0013] 1 is a diagram showing the configuration of a system including a customer service support device; FIG. 2 is a block diagram showing the functional configuration of the customer service support device; FIG. 3 is a diagram showing an example of an output screen; FIG. 4 is a flowchart showing the operation of the customer service support device; FIG. 5 is a block diagram showing another example of the functional configuration of the customer service support device; FIG. 6 is a diagram showing a hardware configuration for realizing the customer service support device in the present disclosure with a computer and its peripheral devices;

[0014] Embodiments of the present disclosure will be described in detail with reference to the drawings.

[0015] [Embodiment] FIG. 1 is a diagram illustrating an example of a system including a customer service support device 100 according to the present disclosure.

[0016] In FIG. 1, customer service support device 100 is a device that outputs customer service support information that supports customer service using the results of skin condition diagnosis when selling cosmetics.

[0017] In Fig. 1 , customer service assistance device 100 is realized by a server. Furthermore, customer service assistance device 100 is communicably connected to a store clerk terminal and a trained model. Note that the configuration of a system including customer service assistance device 100 is not limited to the example shown in Fig. 1 . For example, the trained model may be included in customer service assistance device 100. Furthermore, for example, customer service assistance device 100 may be realized by a store clerk terminal.

[0018] The clerk terminal is a terminal device used by a clerk who serves a customer, and may be, for example, a tablet or a smartphone.

[0019] The trained model is a model that outputs customer service support information according to input information. A detailed description of the trained model will be given later.

[0020] Next, the configuration of the customer service support device 100 according to the embodiment will be described.

[0021] 2 is a block diagram showing the configuration of customer service assistance device 100. Referring to FIG. 2, customer service assistance device 100 includes an acquisition unit 101, a generation unit 102, and an output unit 103.

[0022] Next, the configuration of the customer service support device 100 in the first embodiment will be described in detail.

[0023] The acquisition unit 101 is an example of an acquisition means that acquires a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin.

[0024] The diagnosis result is information indicating the skin condition determined from a diagnostic image of the customer's face. The information indicating the skin condition is, for example, information in which conditions such as dryness of the skin, skin texture, skin blemishes, and sunburn are expressed in multiple stages or numerical values. The information indicating the skin condition may also be information indicating, for example, the color of the skin, eyes, or hair, or the color of cosmetics that suit the customer according to the color.

[0025] An example of a method for acquiring a diagnostic result will be described. For example, a sales clerk takes a diagnostic image of a customer's face using a sales clerk terminal and sends it to a system that diagnoses skin conditions. The sales clerk terminal receives the diagnostic result from the system that diagnoses skin conditions. The acquisition unit 101 acquires the diagnostic result received by the sales clerk terminal. Note that any known method may be used to diagnose the skin condition. Furthermore, the customer service support device 100 may have a function for diagnosing the skin condition.

[0026] The customer service history information is information that indicates the history of customer service provided to the customer. The customer service history information includes the content of conversations between the customer and the store clerk during past or current customer service sessions, as well as the customer's purchase history. The customer service history information may include the customer's identification information, the identification information of the store clerk who served the customer, the date and time of the customer service, and the duration of the customer service. The customer service history information may also include information about products that the customer tried during the customer service session. The current customer service session refers to the time when customer service assistance device 100 is about to output customer service assistance information.

[0027] The content of the conversation between the customer and the sales clerk includes information related to the customer's request for purchasing cosmetics. For example, information related to the customer's request for purchasing cosmetics may include information indicating the product or product category the customer is considering purchasing, and information indicating the customer's concerns or questions regarding cosmetics or skin. The content of the conversation between the customer and the sales clerk may be recorded, for example, by the sales clerk filling out a customer chart. For example, the content of the conversation between the customer and the sales clerk may be recorded as audio data recorded during the customer-to-sales clerk interaction or as text data transcribed from the audio. Note that the acquisition unit 101 may acquire the content of the conversation between the customer and the sales clerk by analyzing text data that has been previously converted into text using voice recognition or the like. The acquisition unit 101 may also include a function for converting the audio of the conversation between the customer and the sales clerk into text data and extracting the content of the conversation between the customer and the sales clerk by analyzing the text data.

[0028] For example, past customer service history information is associated with customer identification information and recorded for each customer in a customer information database. The acquisition unit 101 may acquire the past customer service history information using the customer identification information input to the store clerk terminal. The customer service history information for the current customer service session is information input to the store clerk terminal, and the acquisition unit 101 acquires the customer service history information from the store clerk terminal.

[0029] The acquisition unit 101 may also acquire other information used to generate customer service assistance information.

[0030] For example, the acquisition unit 101 may further acquire skill information indicating the sales clerk's customer service skills. For example, the skill information may be the number of years of experience or the number of customer service sessions the sales clerk has performed. Furthermore, for example, the skill information may be the amount of variation in the content of suggestions made to customers, estimated from the sales clerk's customer service history information. Furthermore, for example, the skill information may be information indicating skills appropriate to each situation, such as the number of customer service sessions performed by each customer attribute or the number of customer service sessions performed by each recommended cosmetic product, estimated from the sales clerk's customer service history information. The skill information may be expressed as skill levels classified into multiple stages according to the numerical values ​​of the various examples of skill information described above. Such skill information may be recorded in advance in association with the sales clerk's identification information, or may be identified using a known method by a skill identification unit provided inside or outside the customer service support device 100.

[0031] Furthermore, for example, the acquiring unit 101 may acquire information specifying the level of detail of the customer service assistance information selected by the store clerk. For example, the information specifying the level of detail of the customer service assistance information may be information indicating the level of detail classified into multiple stages. For example, the information specifying the level of detail of the customer service assistance information may be information indicating whether the form of output of the customer service assistance information is a talk script or items to be discussed. Furthermore, the information specifying the level of detail of the customer service assistance information may be information specifying the level of detail of the customer service assistance information for each item to be discussed.

[0032] Furthermore, for example, the acquisition unit 101 may acquire information indicating the customer's situation. The customer's situation may be the customer's emotions, the customer's physical condition, whether the customer is in a hurry, or the like. The information indicating the customer's emotions or physical condition may be information recorded in a medical record by the customer or a store clerk based on a conversation with the customer, or may be identified from the audio of the conversation between the customer and the store clerk, or may be identified from a diagnostic image of the customer used to diagnose the skin condition. The information indicating whether the customer is in a hurry may be information recorded in a medical record by the customer or a store clerk based on a conversation with the customer, or may be extracted from the audio of the conversation between the customer and the store clerk, as described above.

[0033] Furthermore, the acquisition unit 101 may acquire the customer service history information and the diagnosis results even after the output unit 103 (described later) outputs the customer service assistance information. That is, the acquisition unit 101 may continuously acquire at least one of the customer service history information and the diagnosis results in real time while serving customers. The respective components in this case will be described later in Modification 1.

[0034] The generation unit 102 is an example of a generation means that generates customer service assistance information by inputting a diagnosis result and customer service history information into a trained model that can output customer service assistance information. The generation unit 102 generates at least one of customer service assistance information to be output to a store clerk and customer service assistance information to be output to a customer. In the following description, the generation of customer service assistance information to be output to a store clerk will be mainly described, but the generation of customer service assistance information to be output to a customer may be performed in a similar manner.

[0035] The customer service support information is information for supporting the sales clerk in serving customers, i.e., information that the sales clerk should convey to the customer based on the diagnosis result and the customer service history information.

[0036] The information that the salesperson should convey to the customer is, for example, information about the cosmetics recommended to the customer based on the diagnosis results and customer service history information. The information about the cosmetics recommended to the customer includes at least the product name of the cosmetics. The information about the cosmetics recommended to the customer may include product information about the cosmetics such as price, content volume, release date, and stock quantity. The information about the cosmetics recommended to the customer may also include information indicating the characteristics of the cosmetics, information indicating how to use the cosmetics, and information about the ingredients of the cosmetics. The information about the characteristics of the cosmetics is, for example, the selling points of the cosmetics. The information about the ingredients of the cosmetics is, for example, information about ingredients contained in the cosmetics that are related to the customer's needs.

[0037] The customer service support information may also include information indicating a speaking style for the clerk to convey to the customer what the clerk should communicate to the customer. The information indicating the speaking style may be, for example, a talk script for the clerk to convey to the customer what the clerk should communicate to the customer. The talk script may be a sentence to be read by the clerk, or may be information indicating the order of topics or information indicating points to note about the speaking style.

[0038] Furthermore, the customer service support information may differ depending on the skills of the sales clerk. For example, the lower the skill indicated in the sales clerk's skill information, the more information and detailed the content of the customer service support information may be. For example, if the information that a sales clerk needs to convey to a customer has multiple levels, such as major items, medium items, minor items, and details, customer service support information that only includes major items has a low level of detail, while customer service support information that includes everything from major items to details has a high level of detail. For example, for information about cosmetics recommended by a customer, the major item is the product name of the cosmetics, the medium item is the effects of the cosmetics, the minor item is information indicating the ingredients of the cosmetics, and the details are a talk script. The generation of customer service support information according to the skills of the sales clerk by the generation unit 102 will be described later.

[0039] The generation unit 102 receives the diagnosis results and customer service history information and generates the above-mentioned customer service assistance information using a trained model that can output customer service assistance information.

[0040] A trained model capable of outputting customer service assistance information is, for example, a trained model that has been machine-learned using past customer service history information as training data. In this case, the training data is data in which the explanatory variables are customer service history information and diagnosis results from past customer service encounters, and the objective variable is information indicating customer reactions from past customer service encounters.

[0041] The information indicating the customer's response includes at least one of whether or not a cosmetic product was purchased as a result of the service and whether or not the customer was served again after the service. Whether or not a cosmetic product was purchased as a result of the service may be determined based on the customer's purchase history after the service indicated in the service history information. For example, whether or not a cosmetic product was purchased as a result of the service may be determined based on whether or not the cosmetic product included in the service history information is included in the customer's purchase history within a predetermined period, such as one week, after the service indicated in the service history information. Whether or not the customer was served again after the service may be determined based on the customer's service history information after the service indicated in the service history information. For example, whether or not the customer was served again after the service may be determined based on whether or not customer service history information for the customer is recorded within a predetermined period, such as six months, after the service indicated in the service history information.

[0042] In this case, the generation unit 102 generates information that the store clerk should communicate to the customer as customer service support information by inputting the customer service history information and the diagnosis results into a trained model that has been machine-learned using past customer service history information as training data.

[0043] Furthermore, for example, the trained model capable of outputting customer service support information may be a large-scale language model. For example, the trained model capable of outputting customer service support information is a large-scale language model that has additionally learned information about cosmetics and information about customer service. When using a large-scale language model, the generation unit 102 creates a prompt to be input to the large-scale language model and inputs the created prompt to the large-scale language model. For example, the generation unit 102 may create a prompt using a prompt template. For example, the prompt template includes a command statement that indicates the information to be created and an input field for inputting information to be used for the creation. Furthermore, the prompt may include wording for obtaining a better output, such as "You are an excellent cosmetics salesperson."

[0044] The generation unit 102 generates information that the salesperson should communicate to the customer as customer service support information, for example, by inputting a prompt including customer service history information and diagnosis results into a large-scale language model that has additionally learned information about cosmetics and information about customer service. In this case, the command statement of the prompt may be, for example, "Based on the customer service history information and diagnosis results, please create cosmetics to recommend to the customer and information to communicate to encourage the customer to purchase those cosmetics."

[0045] Furthermore, for example, the generation unit 102 may generate a talk script by inputting a prompt including customer service history information and diagnosis results into a large-scale language model that has additionally learned information about cosmetics and information about customer service. In this case, the command statement of the prompt may be, for example, "Based on the customer service history information and diagnosis results, please create a talk script to discuss the cosmetics you recommend to the customer and the information that should be communicated to encourage the customer to purchase the cosmetics."

[0046] Furthermore, for example, the generation unit 102 may generate a talk script for the sales clerk to communicate to the customer by inputting a prompt including information the sales clerk should communicate to the customer into a large-scale language model. In this case, the information the sales clerk should communicate to the customer may be generated using a trained model that has been machine-learned using past customer service history information as training data, or a large-scale language model that has been additionally trained with information about cosmetics and information about customer service. In this case, the command statement of the prompt may be, for example, "The information you should tell the customer when serving them is as follows. Please create a talk script for you to talk about these information as a cosmetics sales clerk."

[0047] The type and method of the trained model including the large-scale language model are not particularly limited, and may be used as appropriate.

[0048] Furthermore, the generation unit 102 may generate customer service assistance information by further using other information.

[0049] For example, the generation unit 102 may output customer service support information corresponding to the clerk's skill information by further inputting the clerk's skill information into a trained model capable of outputting customer service support information. For example, the generation unit 102 may generate customer service support information with higher detail as the skill indicated in the skill information becomes lower. Conversely, the generation unit 102 may generate customer service support information with lower detail as the skill indicated in the skill information becomes higher. This is because the more skilled a clerk is, the more knowledgeable they are about cosmetics and the more experience they have in customer service, and therefore the less detailed customer service support information they are likely to need. For example, when the clerk's skill is low, the generation unit 102 generates customer service support information that includes details such as a talk script. In this way, even an inexperienced and low-skilled clerk can easily provide appropriate customer service by providing service in accordance with the customer service support information. Furthermore, when the clerk's skill is high, the generation unit 102 generates customer service support information that includes, for example, only major categories. It is expected that highly skilled clerks can provide flexible customer service according to the situation with the customer. Therefore, by generating customer service assistance information with a low level of detail by the generating unit 102, it becomes possible to provide appropriate customer service according to the situation with the customer while using the customer service assistance information.

[0050] For example, the generation unit 102 may generate a talk script when the sales clerk has low skills. The talk script contains detailed information about what to say. A command statement of the prompt when the sales clerk has low skills may be, for example, "Please create a detailed talk script to explain which cosmetics you recommend to the customer based on the customer service history information and the diagnosis results, and the information you should convey to the customer to get them to purchase the cosmetics." Furthermore, for example, the generation unit 102 may generate information that the sales clerk should convey to the customer when the sales clerk has high skills. A command statement of the prompt when the sales clerk has high skills may be, for example, "Please create a concise command statement of which cosmetics you recommend to the customer based on the customer service history information and the diagnosis results, and the information you should convey to the customer to get them to purchase the cosmetics." Furthermore, for example, the generation unit 102 may generate information that the sales clerk should convey to the customer when the sales clerk has medium skills. For example, a prompt command for a salesperson with average skills might be, "Based on the customer service history information and diagnosis results, please recommend cosmetics to the customer and provide information to encourage the customer to purchase those cosmetics. Please also include key points to be discussed for each item."

[0051] Note that the customer service support information according to skills and the prompts for generating the customer service support information are not limited to these examples.

[0052] Furthermore, the generation unit 102 may generate customer service assistance information with a level of detail that corresponds to the selection by the store clerk by further inputting information that specifies the level of detail of the customer service assistance information selected by the store clerk. In this case, the generation unit 102 generates customer service assistance information with a level of detail that corresponds to the selection by the store clerk by including, in the command statement of the prompt, a statement that specifies the level of detail selected by the store clerk or a form of the customer service assistance information that corresponds to the level of detail.

[0053] Furthermore, the generation unit 102 may generate customer service support information according to the customer's situation by further inputting information indicating the customer's situation. For example, if the customer is not feeling well, the generation unit 102 may generate customer service support information to enable the customer to enjoy shopping by including a phrase such as "Please write a compliment for the diagnosis result" in the command sentence of the prompt. For example, if the customer is not feeling well or is in a hurry, the generation unit 102 may generate customer service support information to enable the customer to purchase cosmetics in a short time by including a phrase such as "Please output a talk script that can provide information about cosmetics in a short time" in the command sentence of the prompt.

[0054] Note that the models capable of outputting customer service assistance information used by the generation unit 102 are not limited to these examples. Furthermore, the generation unit 102 may generate customer service assistance information using one or more arbitrary number of models capable of outputting customer service assistance information.

[0055] The output unit 103 is an example of an output means for outputting the generated customer service assistance information. For example, the output unit 103 outputs the customer service assistance information to a display unit of a store clerk terminal of a store clerk who serves a customer.

[0056] When the customer service assistance information includes multiple items that the clerk should communicate to the customer, the output unit 103 may output the customer service assistance information depending on the status of the interaction between the clerk and the customer. The status of the interaction between the clerk and the customer may be, for example, whether the clerk has spoken about each of the multiple items that the clerk should communicate to the customer. For example, the output unit 103 may reduce the number of items of customer service assistance information displayed depending on the content of the conversation between the clerk and the customer acquired by the acquisition unit 101 or the content of the conversation between the clerk and the customer, as determined by the clerk's operation. In this case, the output unit 103 may increase the number of items that the clerk should speak about after an item that the clerk has already spoken about. For example, when reducing the number of items to be displayed, the output unit 103 may display only the item names, and when increasing the number of items to be displayed, the output unit 103 may display the item names and their details.

[0057] Furthermore, output unit 103 may output the customer service support information to be output to the sales clerk and the customer service support information to be output to the customer in a manner that allows them to change the information in response to an operation by the sales clerk. For example, output unit 103 may output the customer service support information to be output to the sales clerk to a sales clerk terminal, and output the customer service support information to be output to the customer to a terminal device that the customer can refer to. In this case, the customer service support information to be output to the sales clerk is output in a format that allows the sales clerk to easily converse with the customer while looking at the display unit. For example, the customer service support information to be output to the sales clerk may include, in addition to the skin condition diagnosis results and cosmetic information, customer service history information, information indicating how to speak to the customer, information about how to use the cosmetic product, and information about its ingredients. In contrast, the customer service support information to be output to the customer is output in a format that allows the customer to easily consider purchasing the cosmetic product. For example, the customer service support information to be output to the customer may include the skin condition diagnosis results and cosmetic product information, and may further include a button that allows the customer to display information about how to use the cosmetic product and its ingredients.

[0058] An example of an output screen of customer service assistance information output by the output unit 103 will be described with reference to FIG.

[0059] FIG. 3 shows an example of an output screen of customer service assistance information to be output to a sales clerk on a sales clerk terminal. In FIG. 3, the output screen includes, as a sales clerk screen, customer information, skin diagnosis results, customer service assistance information, and staff information. Also, in FIG. 3, the output screen includes, as a sales clerk screen, a button for recapturing a diagnostic image and an end button for ending the display of the customer service assistance information. Also, in FIG. 3, the output screen includes tabs for switching the displayed screen, a display of the name of the staff member logged in to the sales clerk terminal, and a button for logging out the staff member. In FIG. 3, the tabs include a sales clerk screen tab for displaying customer service assistance information to be output to the sales clerk, a customer screen tab for displaying customer service assistance information to be output to the customer, and a data confirmation tab for checking other data.

[0060] In the store clerk screen shown in Figure 3, customer information is information about the customer. In Figure 3, the customer information displays the customer's name, age, and gender, which are previously registered in association with the customer's identification information. Also in Figure 3, the customer information includes current and previous customer service history information (counseling information in Figure 3). In Figure 3, the current customer service history information displays the product category the customer wishes to purchase, the customer's concerns related to the desired purchase, and the target location of the customer's concerns marked on the face illustration. The current customer service history information displayed on the screen is input into the store clerk terminal by the store clerk or the customer. Also in Figure 3, the previous customer service history information displays the date of the service, the customer's concerns related to the previous purchase, and the previous purchase history.

[0061] The skin diagnosis results in Fig. 3 include a diagnostic image and diagnostic results shown by a graph. In Fig. 3, some of the diagnostic results that have been diagnosed as numerical values ​​indicating poor condition are highlighted with star-shaped marks. In Fig. 3, the positions on the diagnostic image that have been diagnosed as numerical values ​​indicating poor condition are also highlighted with star-shaped marks. This allows the store clerk to easily and quickly understand the skin diagnosis results.

[0062] The customer service support information in FIG. 3 is a talk script for recommending products AAA and BBB to a customer based on the diagnosis results and customer service history information. The talk script includes the diagnosis results, an overview of the recommended products, information about product AAA, information about product BBB, and instructions for using the recommended products. In FIG. 3, a collapse button for reducing the display is included at the top of each item in the talk script. In FIG. 3, the talk script for an item for which the collapse button is selected may be controlled so that it is not displayed. The store clerk selects the collapse button to reduce the display of items that have been discussed or that do not require detailed customer service support information, making other information easier to view. In FIG. 3, instructions for allowing the customer to select at least one of the recommended products and a selection field are displayed to display instructions for using the recommended products. In FIG. 3, when a product is selected, a talk script explaining how to use the selected product may be displayed.

[0063] The staff member information in Fig. 3 includes the name and skill information of the staff member who is the store clerk in charge of customer service. Fig. 3 shows that the staff member's skill level is level 1 out of 5, i.e., low skill. The customer service support information in Fig. 3 is generated in talk script format according to the staff member's skill, and includes information on ingredients of the recommended product, information based on the purchase history, and a button for displaying instructions on how to use the recommended product.

[0064] 3, when the customer screen is selected, the same skin analysis results as those on the clerk screen and customer service support information for the customer may be displayed, in which the talk script format of the clerk screen has been changed to a bulleted format. The customer screen may also be configured not to display an end button or a logout button. Furthermore, the customer screen may display, as customer service support information for the customer, more detailed information about the products, such as product images of recommended products and product prices.

[0065] The output screen example in Figure 3 is an example, and the customer service assistance information and the output screen output by the output unit 103 are not limited to this example. Furthermore, the output format and output destination of the output unit 103 are not limited to the example in Figure 3. The output unit 103 may record the customer service assistance information in a database as customer service history information in association with the customer's identification information. The output unit 103 may output text or audio customer service assistance information to a terminal device used by the customer as a customer service in the form of chat or audio conversation with the trained model used by the generation unit 102.

[0066] The operation of the customer service support device 100 configured as above will be described with reference to the flowchart of FIG.

[0067] The acquisition unit 101 acquires a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin (step S101).

[0068] Next, the generation unit 102 receives the diagnosis result and the customer service history information and generates customer service assistance information using a trained model that can output customer service assistance information (step S102).

[0069] Next, the output unit 103 outputs the generated customer service assistance information (step S103).

[0070] This completes the series of operations of customer service support device 100.

[0071] The customer service assistance device in the above-described embodiment includes an acquisition unit, a generation unit, and an output unit. The acquisition unit acquires a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin. The generation unit receives the diagnosis result and the customer service history information and generates customer service assistance information using a trained model capable of outputting customer service assistance information. The output unit then outputs the generated customer service assistance information.

[0072] The customer service support device of this embodiment allows a salesperson serving a customer to provide service while referring to the diagnosis results, including a diagnostic image of the customer's skin condition, and customer service support information corresponding to the customer's skin-related customer service history information. As a result, the customer service support device of this embodiment can provide a technology to support customer service using the skin condition diagnosis results when selling cosmetics. For example, even an inexperienced salesperson can provide service while referring to the customer service support information, so the customer service support device of this embodiment can support the provision of better customer service.

[0073] [Modification 1] Next, Modification 1 of the present disclosure will be described in detail with reference to the drawings. Below, explanations of content that overlaps with the above explanation will be omitted to the extent that the explanation of this embodiment is not unclear.

[0074] In the first modification, the customer service assistance device 100 may update the customer service assistance information in real time using customer service history information acquired during customer service.

[0075] In this case, the acquisition unit 101 acquires at least one of customer service history information during the customer service and a diagnosis result during the customer service. For example, the customer service history information during the customer service may be information entered by the salesperson into a chart on the salesperson terminal or audio data of a conversation between the salesperson and the customer. Furthermore, for example, the diagnosis result during the customer service may be a diagnosis result when a skin condition is diagnosed again. The second diagnosis may be, for example, a diagnosis of the condition after trying out a product. For example, the second diagnosis may be a diagnosis of whether the skin condition has improved as a result of trying out the product, whether the color of the tried out product suits the customer, etc.

[0076] The generation unit 102 updates the customer service assistance information by further inputting at least one of the acquired customer service history information and diagnosis results during customer service into the trained model.

[0077] The generation unit 102 may further use a trained model that estimates emotions from images. The generation unit 102 updates the customer service support information in accordance with the customer's emotions based on the images during customer service, which are estimated by inputting the images into the trained model that estimates emotions from images.

[0078] The processing of customer service support device 100 in Modification 1 will be described with reference to the flowchart of FIG.

[0079] The acquisition unit 101 acquires a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin (step S201).

[0080] Next, the generation unit 102 receives the diagnosis result and the customer service history information and generates customer service assistance information using a trained model that can output customer service assistance information (step S202).

[0081] Next, the output unit 103 outputs the generated customer service assistance information (step S203).

[0082] Next, the acquisition unit 101 further acquires diagnostic results or customer service history information during the customer service session (step S204). If the acquisition unit 101 further acquires diagnostic results or customer service history information during the customer service session in step S204, the process proceeds to step S202 using the further diagnostic results or customer service history information during the customer service session. That is, the generation unit 102 generates updated customer service support information using the further acquired diagnostic results or customer service history information during the customer service session. In the following step S203, the output unit 103 outputs the updated customer service support information. Note that if the acquisition unit 101 does not acquire diagnostic results or customer service history information during the customer service session in step S204, the process may end.

[0083] This completes the series of operations of customer service support device 100.

[0084] The customer service support device in Variation 1 generates and outputs updated customer service support information using information obtained in real time. This increases the likelihood of generating customer service support information that will more satisfy the customer based on the customer's reaction. For example, if the customer's reaction to customer service using previously generated customer service support information is poor, the customer service support device in Variation 1 can generate revised customer service support information that is more likely to improve the customer's reaction. Therefore, the customer service support device in Variation 1 can better support customer service.

[0085] [Modification 2] Next, Modification 2 of the present disclosure will be described in detail with reference to the drawings. Below, explanations of content that overlaps with the above explanation will be omitted to the extent that the explanation of this embodiment is not unclear.

[0086] FIG. 6 is a block diagram showing the configuration of the customer service support device 100A.

[0087] In the second modification, customer service assistance device 100A further includes a learning unit 104 in addition to the configuration of customer service assistance device 100 .

[0088] The learning unit 104 is an example of a learning means that learns a model capable of outputting customer service support information.

[0089] The learning unit 104 may learn, for example, customer service history information and diagnosis results as explanatory variables, and information indicating customer reactions during past customer service as objective variables, as training data.

[0090] For example, the learning unit 104 may train the large-scale language model using customer service history information, diagnosis results, and information indicating customer reactions as training data. The learning unit 104 may also train the large-scale language model using information on cosmetics, skin care methods, and cosmetic ingredients.

[0091] The learning unit 104 may perform learning to create a model, or may perform learning to update a model.

[0092] The learning unit 104 may also have the model learn information about newly released cosmetics, thereby making it possible to output customer service support information about newly released cosmetics as well.

[0093] [Hardware Configuration] Some or all of the components of each device or system in each embodiment of the present disclosure described above are realized by any combination of an information processing device 1000 and a program, for example, as shown in Fig. 7. The information processing device 1000 includes, as an example, the following configuration.

[0094] - CPU (Central Processing Unit) 1001 - ROM (Read Only Memory) 1002 - RAM (Random Access Memory) 1003 - Program 1004 loaded into RAM 1003 - Storage device 1005 for storing program 1004 - Drive device 1007 for reading and writing from and to recording medium 1006 - Communication I / F 1008 for connecting to communication network 1009 - Input / output I / F 1010 for inputting and outputting data - Bus 1011 for connecting each component. Note that I / F is an abbreviation for Interface.

[0095] Each component of each device or system in each embodiment is realized by the CPU 1001 acquiring and executing a program that realizes the function of that component. The program that realizes the function of each component of each device is stored in the storage device 1005 or RAM 1003 in advance, for example, and is read by the CPU 1001 as needed. The program 1004 may be supplied to the CPU 1001 via a communication network, or may be stored in the recording medium 1006 in advance, and the drive device 1007 may read the program and supply it to the CPU 1001.

[0096] There are various variations in the method of realizing each device. For example, each device or system may be realized by any combination of a separate information processing device 1000 and a program for each component. Furthermore, multiple components included in each device may be realized by any combination of a single information processing device 1000 and a program.

[0097] Furthermore, some or all of the components of each device or system may be realized by general-purpose or dedicated circuits (circuitry) including a processor or the like, or a combination of these. Examples of circuits include a CPU, a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), and an AI (Artificial Intelligence) processing LSI (Large Scale Integration). These may be configured by a single chip or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits and a program.

[0098] When some or all of the components of each device or system are realized by multiple information processing devices, circuits, etc., the multiple information processing devices, circuits, etc. may be centrally or decentralized. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, or the like, in which they are connected via a communication network.

[0099] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0100] Furthermore, although the operations are described in a sequential order in the form of a flowchart, the order of the description does not limit the order in which the operations are performed. Therefore, when implementing each embodiment, the order of the operations may be changed as long as it does not affect the content.

[0101] Some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0102] [Supplementary Note 1] A customer service support device comprising: an acquisition means for acquiring a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin; a generation means for generating customer service support information by inputting the diagnosis result and the customer service history information into a trained model capable of outputting customer service support information; and an output means for outputting the generated customer service support information.

[0103] [Supplementary Note 2] The customer service support device described in Supplementary Note 1, wherein the customer service support information is information indicating a speaking style in which the store clerk will convey to the customer information that the store clerk should convey to the customer, based on the diagnosis result and the customer service history information.

[0104] [Supplementary Note 3] The customer service support device according to Supplementary Note 2, wherein the information indicating the speaking style is a talk script for the store clerk to convey the matter to the customer.

[0105] [Supplementary Note 4] The customer service assistance device according to Supplementary Note 3, wherein the talk script is information obtained by inputting the customer service assistance information into a large-scale language model.

[0106] [Supplementary Note 5] The customer service support device described in any one of Supplementary Notes 1 to 4, wherein the acquisition means further acquires skill information indicating the skills of the store clerk, and the generation means further inputs the skill information and generates the customer service support information according to the skill information using the trained model.

[0107] [Supplementary Note 6] The customer service assistance device according to Supplementary Note 5, wherein the generating unit generates the customer service assistance information with a higher level of detail as the skill indicated in the skill information becomes lower.

[0108] [Supplementary Note 7] The customer service support device described in any one of Supplementary Notes 1 to 6, wherein the acquisition means acquires information specifying a level of detail of customer service support information selected by a store clerk, and the generation means generates the customer service support information of the level of detail.

[0109] [Supplementary Note 8] The customer service support device described in any of Supplementary Notes 1 to 7, wherein the acquisition means further acquires at least one of an image of the customer during service and the customer service history information during service, and the generation means updates the customer service support information using the trained model by further inputting at least one of the image during service and the customer service history information.

[0110] [Supplementary Note 9] The customer service support device according to Supplementary Note 8, wherein the generation means updates the customer service support information according to the emotion of the customer based on the image during the customer service, which emotion is estimated using a trained model that estimates emotions from images.

[0111] [Supplementary Note 10] The customer service support device described in any one of Supplementary Notes 1 to 9, wherein the output means outputs the customer service support information according to the status of the interaction between the store clerk and the customer when the customer service support information includes multiple items that the store clerk should communicate to the customer.

[0112] [Supplementary Note 11] The customer service support device according to any one of Supplementary Notes 1 to 10, wherein the customer service support information is information that a store clerk should convey to the customer in accordance with the diagnosis result and the customer service history information.

[0113] [Supplementary Note 12] The customer service support device according to Supplementary Note 11, wherein the item is information about cosmetics recommended to the customer.

[0114] [Supplementary Note 13] The customer service support device according to Supplementary Note 12, wherein the information about the cosmetic product recommended to the customer is information indicating how to use the cosmetic product.

[0115] [Supplementary Note 14] The customer service support device described in Supplementary Note 12 or 13, wherein the information about the cosmetics recommended to the customer is information about ingredients contained in the cosmetics that are related to the customer's request included in the customer service history information.

[0116] [Supplementary Note 15] The customer service support device described in any one of Supplementary Notes 1 to 14, wherein the generation means generates at least one of the customer service support information to be output to a store clerk serving the customer and the customer service support information to be output to the customer.

[0117] [Supplementary Note 16] The customer service support device according to Supplementary Note 15, wherein the output means outputs the customer service support information to be output to the store clerk and the customer service support information to be output to the customer in a changeable manner in response to an operation by the store clerk.

[0118] [Supplementary Note 17] A customer service support method comprising: acquiring a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin; generating the customer service support information using a trained model capable of outputting customer service support information by inputting the diagnosis result and the customer service history information; and outputting the generated customer service support information.

[0119] [Appendix 18] A recording medium that records a program that causes a computer to execute the following processes: acquiring a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin; generating the customer service support information using a trained model that can output customer service support information by inputting the diagnosis result and the customer service history information; and outputting the generated customer service support information.

[0120] Furthermore, some or all of the configurations described in Supplementary Notes 2 to 16 that are subordinate to Supplementary Note 1 (customer service support device) may also be subordinate to Supplementary Note 17 (customer service support method) and Supplementary Note 18 (program) in the same subordinate relationship as Supplementary Note 2 to 16. Furthermore, not limited to Supplementary Note 1, Supplementary Note 17, and Supplementary Note 18, some or all of the configurations described as supplementary notes may be subordinate to various hardware, software, various recording means for recording software, or systems, as long as they do not deviate from the respective embodiments described above.

[0121] REFERENCE SIGNS LIST 100 Customer service support device 101 Acquisition unit 102 Generation unit 103 Output unit 100A Customer service support device 104 Learning unit 1000 Information processing device 1001 CPU 1002 ROM 1003 RAM 1004 Program 1005 Storage device 1006 Recording medium 1007 Drive device 1008 Communication I / F 1009 Communication network 1010 Input / output I / F 1011 Bus

Claims

1. A customer service support device comprising: an acquisition means for acquiring a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin; a generation means for generating customer service support information by inputting the diagnosis result and the customer service history information into a trained model capable of outputting customer service support information; and an output means for outputting the generated customer service support information.

2. The customer service support device of claim 1, wherein the customer service support information is information indicating a manner of speaking in which the store clerk will communicate to the customer what the store clerk should communicate to the customer, based on the diagnosis result and the customer service history information.

3. The customer service support device according to claim 2, wherein the information indicating the speaking style is a talk script for the store clerk to convey the matter to the customer.

4. The customer service support device according to claim 3, wherein the talk script is information obtained by inputting the customer service support information into a large-scale language model.

5. A customer service support device as described in any one of claims 1 to 4, wherein the acquisition means further acquires skill information indicating the skills of the store clerk, and the generation means further inputs the skill information and uses the trained model to generate the customer service support information corresponding to the skill information.

6. The customer service assistance device according to claim 5, wherein the generating means generates the customer service assistance information with a higher level of detail as the skill indicated in the skill information becomes lower.

7. A customer service support device as described in any one of claims 1 to 6, wherein the acquisition means acquires information specifying the level of detail of the customer service support information selected by the store clerk, and the generation means generates the customer service support information of said level of detail.

8. A customer service support device as described in any one of claims 1 to 7, wherein the acquisition means further acquires at least one of an image of the customer during service and the customer service history information during service, and the generation means updates the customer service support information using the trained model by further inputting at least one of the image during service and the customer service history information.

9. The customer service support device according to claim 8, wherein the generation means updates the customer service support information in accordance with the customer's emotion based on the image during the customer service, estimated using a trained model that estimates emotions from images.

10. A customer service support device as described in any one of claims 1 to 9, wherein the output means outputs the customer service support information according to the status of the interaction between the store clerk and the customer when the customer service support information contains multiple items that the store clerk should communicate to the customer.

11. A customer service support device according to any one of claims 1 to 10, wherein the customer service support information is information that a store clerk should convey to the customer depending on the diagnosis result and the customer service history information.

12. The customer service support device according to claim 11, wherein the item is information about cosmetics recommended to the customer.

13. The customer service support device according to claim 12, wherein the information about the cosmetics recommended to the customer is information indicating how to use the cosmetics.

14. The customer service support device according to claim 12 or 13, wherein the information regarding the cosmetics recommended to the customer is information regarding ingredients contained in the cosmetics that are related to the customer's request contained in the customer service history information.

15. A customer service support device according to any one of claims 1 to 14, wherein the generation means generates at least one of the customer service support information to be output to a store clerk serving the customer and the customer service support information to be output to the customer.

16. The customer service support device according to claim 15, wherein the output means outputs the customer service support information to be output to the store clerk and the customer service support information to be output to the customer in a manner that allows them to be changed in response to an operation by the store clerk.

17. A customer service support method comprising: acquiring a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin; generating customer service support information by inputting the diagnosis result and the customer service history information into a trained model capable of outputting customer service support information; and outputting the generated customer service support information.

18. The customer service support method described in claim 17, wherein the customer service support information is information indicating a manner of speaking in which the store clerk will convey to the customer information that the store clerk should convey to the customer, based on the diagnosis result and the customer service history information.

19. A recording medium that records a program that causes a computer to execute the following process: obtaining a diagnosis result including a diagnostic image of a customer's skin condition and customer service history information related to the customer's skin; generating customer service support information by inputting the diagnosis result and the customer service history information into a trained model that can output customer service support information; and outputting the generated customer service support information.

20. The recording medium described in claim 19, wherein the customer service support information is information indicating how the store clerk should communicate to the customer what the store clerk should communicate to the customer, based on the diagnosis result and the customer service history information.

Images