Automatic response device, automatic response method, and program

Abstract

An automatic response device according to one aspect of the present disclosure comprises: an automatic response unit configured to automatically handle a received inquiry on the basis of preset response flow information; a presentation unit configured to present, to a prescribed presentation destination, a reaction status of the inquiry handling by the automatic response unit when the subject handling the inquiry is switched; and a correction unit configured to correct the response flow information on the basis of reaction result information including the reaction result of the inquiry handling for which the execution subject is switched.

Description

Technical Field

[0001] The present disclosure relates to an automatic response device, an automatic response method, and a program.

Background Art

[0002] In a contact center (also called a call center), generally, an operator determines a talk script for dealing with customers so that there is no difference in customer service among operators. Here, the talk script refers to the content of speech, the speech procedure, etc. determined in the contact center. In the talk script, for example, in each item or each scene such as the first greeting (opening), the inquiry content, customer identification (name, date of birth, etc.), response, and the last greeting (closing), there are often defined sentences, keywords, phrases, etc. that need to be spoken.

[0003] On the other hand, in recent years, with automatic response technologies such as IVR (Interactive Voice Response) and BOT (bot), inquiries from customers are automatically responded to. For example, Patent Document 1 describes a technology in which an automatic response BOT responds to customers when providing customer support services. Also, this Patent Document 1 basically describes that an automatic response BOT responds to customers and switches to an operator as needed to provide customer support services.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the automated response technology described above, for example, anticipated customer interactions are prepared as scenarios, and these scenarios are set in IVR or BOTs, and automated responses are made according to these scenarios. Therefore, if there is an inquiry that is not defined in the scenario, for example, it will be escalated to an operator, increasing the cost of customer service. On the other hand, if scenarios that comprehensively anticipate various types of interactions can be created in advance, escalations to operators can be reduced, but creating such scenarios would require enormous costs and is considered practically impossible.

[0006] This disclosure is made in view of the above points and provides a technology that can reduce the costs associated with automated responses. [Means for solving the problem]

[0007] An automated response system according to one aspect of the present disclosure includes: an automated response unit configured to automatically execute a response to a received inquiry based on pre-set response flow information; a presentation unit configured to present the status of the response by the automated response unit to a predetermined recipient when the entity executing the response to the inquiry changes; and a modification unit configured to modify the response flow information based on response result information, including the response result of the response to the inquiry after the entity executing the response has changed. [Effects of the Invention]

[0008] Technology is provided that can reduce the costs associated with automated responses. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows an example of the overall configuration of the contact center system according to this embodiment. [Figure 2] This figure shows an example of the hardware configuration of the automatic response device according to this embodiment. [Figure 3] This figure shows an example of the functional configuration of the automatic response device according to this embodiment. [Figure 4] This diagram shows an example of a customer service flow. [Figure 5] This flowchart shows an example of the response flow setting process according to this embodiment. [Figure 6] This is a diagram (part 1) showing an example of a talk script. [Figure 7] This is a diagram (part 2) showing an example of a talk script. [Figure 8] This figure shows an example of a customer service flow visualized on the customer service flow settings screen. [Figure 9] This flowchart shows an example of the automated response process according to this embodiment. [Figure 10] This is a diagram (part 1) showing an example of the customer service status visualized on the customer service status screen. [Figure 11] This is a diagram (part 2) showing an example of the customer service status visualized on the customer service status screen. [Figure 12] This is Figure (3) showing an example of the customer service status visualized on the customer service status screen. [Figure 13] This flowchart shows an example of the response flow modification process according to this embodiment. [Figure 14] This figure shows an example of a revised customer service flow visualized on the customer service flow revision screen. [Modes for carrying out the invention]

[0010] The following describes one embodiment of the present invention. In the following embodiment, a contact center system 1 is described that includes an automated response device 10 that automatically responds to customer inquiries and the like, targeting a contact center.

[0011] In the contact center system 1 according to this embodiment, the costs associated with automated responses can be reduced by realizing the following (1) to (3).

[0012] (1) Make it possible to simply and efficiently set the response flow used for automatic response for the automatic response device 10. The response flow is information representing a series of processes from the start to the end of customer service, and automatic response is performed according to this response flow. Hereinafter, the response flow shall be defined as a scenario or a set of scenarios related to customer service. Here, a scenario is information representing the response content of a subject expected in advance to respond to customers (for example, not only a person such as an operator, but also a subject including an automatic response system or program such as IVR or BOT, etc., and an execution subject of any response including inquiry response), that is, information representing customer service. Scenarios can be defined in various ways. For example, it can be defined as an ordered set {(questionn, answer n)|n = 1, ···, N} containing one or more pairs of questions to customers and answers thereto (question, answer). Or, for example, it can also be defined as an ordered set {(scene n, question n, answer n)|n = 1, ···, N} containing one or more triples of a scene, a question, and an answer thereto. However, these definitions of scenarios are just examples and are not limited thereto, and any definition is possible as long as it represents some customer service.

[0013] According to (1) above, the setting of the response flow used for automatic response can be simply and efficiently set for the automatic response device 10, and as a result, the cost required for setting the response flow can be reduced.

[0014] (2) When an escalation to an operator occurs, visualize the response status by automatic response. Escalation means that when a difficult-to-handle event occurs for the entity handling the response (for example, not only a person such as an operator, but also an entity including an automatic response system or program such as an IVR or BOT, which is the entity executing any response including inquiry handling), another entity (for example, a higher-level entity than the currently handling entity) is requested to handle it or an instruction on the handling method is requested. In the following, assume that there is an inquiry that cannot be handled by automatic response as an event difficult to handle by automatic response, and escalation is to request the operator to handle the event (that is, switch the handling of the event to the operator) when the event occurs. However, this is just an example, and escalation is not limited to this.

[0015] According to (2) above, the operator can easily grasp the response status before the escalation occurs, and the cost required for customer response after the escalation can be reduced. Also, due to this visualization, the modification of the response flow in (3) described later can be made more efficient, so the cost required for modifying the response flow can also be reduced.

[0016] (3) Modify the response flow using the response result information including information such as the location where the escalation occurred and the content of the operator's response after the escalation so that the event where the escalation occurred can also be handled.

[0017] According to (3) above, recurrence of past escalations can be prevented or suppressed, so the cost required for customer response after the escalation can be reduced. Also, the cost required for modifying the response flow can be reduced.

[0018] However, a contact center is just one example, and this embodiment can be similarly applied to any system that provides automated responses to inquiries and where escalation may occur. For example, this embodiment can be similarly applied to manned sales such as stores and counter sales, unmanned sales such as vending machines, manned or unmanned inquiry handling, and various types of work support. Furthermore, the entity making the inquiry is not limited to a person; it may be other equipment, systems, programs, etc., and more generally, it can be applied to any system that makes some kind of inquiry.

[0019] In the following description, the automated response device 10 is assumed to provide automated responses to customers via text chat (including those that allow sending and receiving stamps, attachments, etc., in addition to text) or voice calls. However, it is not limited to these, and this embodiment can be similarly applied even when automated responses are provided by any means of communication.

[0020] <Example of the overall configuration of Contact Center System 1> Figure 1 shows the overall configuration of the contact center system 1 according to this embodiment. As shown in Figure 1, the contact center system 1 according to this embodiment includes an automated response device 10, an operator terminal 20, and a customer terminal 30. The automated response device 10, the operator terminal 20, and the customer terminal 30 are connected to communicate via a communication network 40, which includes, for example, the Internet, a VoIP (Voice over Internet Protocol) network, or a PSTN (Public Switched Telephone Network). Here, the automated response device 10 and the operator terminal 20 are installed in the contact center environment E, which is the system environment of the contact center. Note that the contact center environment E is not limited to a system environment in the same building, but may be, for example, a system environment in multiple buildings geographically separated from each other.

[0021] The automated response system 10 is a computer or computer system that automatically responds to customer inquiries according to a pre-configured response flow. In this case, if an issue arises that cannot be handled by automated response, the automated response system 10 requests an operator to handle the issue (escalates the issue).

[0022] The operator terminal 20 is various terminals such as a PC (personal computer) used by the operator. The operator terminal 20 also functions as an IP (Internet Protocol) telephone. The operator terminal 20 may also be, for example, a smartphone, tablet, or wearable device.

[0023] Customer terminals 30 are various devices used by customers, such as PCs (personal computers), smartphones, mobile phones, and landline phones.

[0024] The overall configuration of Contact Center System 1 shown in Figure 1 is an example and is not limited to it. For example, the Contact Center Environment E may include terminals used by supervisors who manage or monitor each operator, a PBX (Private Branch eXchange), various network switches, etc., and telephones used by operators. It may also include a voice recognition system that performs voice recognition for voice calls with customers, a CRM (Customer Relationship Management) system that manages various customer-related information, etc. In addition to these, various devices, equipment, and systems commonly used in contact centers may also be included.

[0025] <Example of hardware configuration for the automatic response device 10> Figure 2 shows an example of the hardware configuration of the automatic response device 10 according to this embodiment. As shown in Figure 2, the automatic response device 10 according to this embodiment includes an input device 101, a display device 102, an external I / F 103, a communication I / F 104, a RAM (Random Access Memory) 105, a ROM (Read Only Memory) 106, an auxiliary storage device 107, and a processor 108. Each of these hardware components is connected to each other via a bus 109 so as to be able to communicate.

[0026] The input device 101 is, for example, a keyboard, mouse, touch panel, or physical button. The display device 102 is, for example, a display or display panel. The automatic response device 10 does not necessarily have to have at least one of the input device 101 and the display device 102.

[0027] External I / F 103 is an interface to external devices such as recording media 103a. The automatic response device 10 can read from and write to the recording media 103a via the external I / F 103. Examples of recording media 103a include flexible disks, CDs (Compact Discs), DVDs (Digital Versatile Disks), SD memory cards (Secure Digital memory cards), and USB (Universal Serial Bus) memory cards.

[0028] The communication interface 104 is an interface for connecting the automatic response device 10 to various networks. The RAM 105 is a volatile semiconductor memory (storage device) that temporarily holds programs and data. The ROM 106 is a non-volatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. The auxiliary storage device 107 is, for example, a storage device (storage device) such as an HDD (Hard Disk Drive), SSD (Solid State Drive), or flash memory. The processor 108 is, for example, a computing device such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit).

[0029] The automatic response device 10 according to this embodiment can realize various processes described later by having the hardware configuration shown in Figure 2. Note that the hardware configuration shown in Figure 2 is just one example, and the hardware configuration of the automatic response device 10 is not limited to this. For example, the automatic response device 10 may have multiple auxiliary storage devices 107 and multiple processors 108, it may not have some of the hardware shown, or it may have various hardware other than the hardware shown.

[0030] <Example of the functional configuration of the automatic response device 10> Figure 3 shows an example of the functional configuration of the automatic response device 10 according to this embodiment. As shown in Figure 3, the automatic response device 10 according to this embodiment includes a response flow setting processing unit 201, an automatic response processing unit 202, an escalation processing unit 203, and a response flow modification processing unit 204. Each of these units is realized, for example, by processing that one or more programs installed in the automatic response device 10 cause a processor 108 or the like to execute. The automatic response device 10 according to this embodiment also has a storage unit 205. The storage unit 205 is realized, for example, by an auxiliary storage device 107. However, all or part of the storage area of ​​the storage unit 205 may be realized by a storage device (for example, a database server, etc.) that is connected to the automatic response device 10 in a communicative manner.

[0031] The response flow setting processing unit 201 executes a process (hereinafter also referred to as the response flow setting process) to set a response flow for the automated response device 10.

[0032] The automated response processing unit 202 executes processing (hereinafter also referred to as automated response processing) to automatically respond to customer inquiries, etc., according to the response flow.

[0033] The escalation processing unit 203 executes a process to visualize the status of the response by the automated response when an escalation occurs in the automated response process.

[0034] The response flow correction processing unit 204 uses the response result information to execute a process for correcting the response flow (hereinafter also referred to as the response flow correction process).

[0035] The memory unit 205 stores various information necessary for the various processes performed by the automatic response device 10 (for example, response flow, response result information, etc.).

[0036] The functional configuration of the automatic response device 10 shown in Figure 3 is merely an example and is not limited thereto. At least a portion of the components of the automatic response device 10 shown in Figure 3 may be present in other devices or equipment connected to the automatic response device 10 via a communication network or the like. For example, the automatic response device 10 may not have an automatic response processing unit 202, while other devices or equipment connected to the automatic response device 10 via a communication network or the like may have an automatic response processing unit 202.

[0037] <Response Flow> As mentioned above, a customer service flow is information that represents the entire process from the start to the end of a customer interaction, and is defined as a scenario related to customer interaction or a collection of such scenarios. Therefore, depending on the definition of the scenario, a customer service flow can be expressed in various forms.

[0038] An example of a response flow is shown in Figure 4. The response flow shown in Figure 4 is represented as a tree structure with directed edges, where each box is a node, with the scene as a single box. In the example shown in Figure 4, the root node is placed on the left and the leaf nodes are on the right, but the root node may be placed on the top and the leaf nodes on the bottom. More generally, the position of each node can be set arbitrarily.

[0039] Each box consists of a scene, one or more ordered slots, branching conditions, and exception handling conditions. A scene, as mentioned above, is a conversational moment such as "initial greeting (opening)," "inquiry confirmation," "customer identity verification (name, date of birth, etc.)," ​​"interaction," and "final greeting (closing)." A slot is a script that represents a pair of a question (or comment) and its answer (or a comment corresponding to that comment). For example, the scene "inquiry confirmation" and the slot "question: Please tell me your name. answer: " are concrete examples of a scene and a slot, respectively. In this example, the answer is set to blank, but depending on the content of the question, specific answer content or answer options may be set.

[0040] A branching condition is a condition used to determine the next destination box (or the next destination scene if there is a one-to-one correspondence between the scene and the box) based on the slots within the box. For example, if the current box is "Box 1", a concrete example of a branching condition is "If all the answers in the slots in Box 1 are filled, transition to Box 2-1". Note that this is just one example, and there may be branching conditions that do not require all the answers in the box to be filled.

[0041] An exception handling condition is a condition for transitioning to a box related to an exceptional response when an exceptional response is required based on the content of a slot within a box. For example, if the current box is "Box 1", a concrete example of an exception handling condition is "If the answer in the slot within Box 1 contains the word "AAA", transition to Box 2-3". In this example, "AAA" represents a negative emotion of the customer, and the transition to Box 2-3 is intended to take some kind of action regarding this negative emotion.

[0042] Furthermore, the box to which the user transitions when an exception handling condition is met may include a return condition to return to the original box. A concrete example of a return condition is, for instance, if the current box is "Box 2-3", then "when all the slots in Box 2-3 are filled with answers, transition to Box 1".

[0043] In the response flow shown in Figure 4, for example, a set of scripts (slots) that appear along the path from the root node to the leaf node represents one scenario. For example, Box 1 → Box 2-1 → Box 3-1 corresponds to one path from the root node to the leaf node, but each box can contain multiple slots. Therefore, if, for example, Box 1 contains two slots (slot 1-1, slot 1-2), Box 2-1 contains three slots (slot 2-1-1, slot 2-1-2, slot 2-1-3), and Box 3-1 contains one slot (slot 3-1-1), then the following ordered set of scripts represents one scenario.

[0044] Script set in slot 1-1 → Script set in slot 1-2 → Script set in slot 2-1-1 → Script set in slot 2-1-2 → Script set in slot 2-1-3 → Script set in slot 3-1-1 Furthermore, if there are branching conditions that do not require filling in the answers for all slots in a box, the set of scripts that appear in the paths branched by those branching conditions also represents a single scenario. For example, suppose there is a branching condition that if a certain condition is met in slot 1-1 of box 1, the path branches to box 2-1. In this case, the following ordered set of scripts represents a single scenario.

[0045] Script set in slot 1-1 → Script set in slot 2-1-1 → Script set in slot 2-1-2 → Script set in slot 2-1-3 → Script set in slot 3-1-1 Similarly, the set of scripts that appear in paths branched by exception handling conditions also represents a single scenario. The same applies to other paths from the root node to the leaf nodes.

[0046] However, in the response flow shown in Figure 4, one box may represent one scenario. For example, if Box 1 contains two slots (Slot 1-1, Slot 1-2), then a set of scripts in an ordered sequence, such as the script set in Slot 1-1 → the script set in Slot 1-2, may represent one scenario. The same applies to the other boxes.

[0047] The customer service flow shown in Figure 4 is just one example and is not limited to this. A customer service flow can be represented in various formats, as long as it shows the entire process from the start to the end of a customer interaction.

[0048] <Response Flow Setup Process> Below, an example of the response flow setting process according to this embodiment will be described with reference to Figure 5. In the following, it will be assumed that no response flow is set in the automatic response device 10 (i.e., no response flow is stored in the storage unit 205).

[0049] First, the response flow setting processing unit 201 displays the response flow setting screen on the operator terminal 20's display in response to a request from the operator terminal 20 (step S101). The response flow setting screen is a screen used to set the response flow for the automated response device 10 and to confirm the setting results.

[0050] Next, the response flow setting processing unit 201 inputs the scenario specified on the operator terminal 20 (step S102). Here, the operator operating the operator terminal 20 can specify the scenario in various formats on the response flow setting screen.

[0051] For example, an operator can specify a scenario that includes a scene, the number of slots, the script to be set for each slot, branching conditions, exception handling conditions, the branching scene when the branching conditions are met, the branching scene when the exception handling conditions are met, and return conditions to return to the original box or scene when transitioning from another box or scene due to the exception handling conditions.

[0052] Furthermore, for example, an operator can specify an existing talk script as a scenario, which defines the scene, the content of the utterances in that scene, and the order of utterances. A concrete example of such a talk script is shown below.

[0053] • Example 1 of a talk script In the talk script shown in Figure 6, for each item representing a scene, the script specifies the sentences that the speaker (typically the operator; hereafter, the speaker will be assumed to be the operator) needs to say in that item.

[0054] For example, the item "Opening Greeting" has a script that reads "Thank you for calling...". This means that the operator must say a phrase like "Thank you for calling..." as their opening greeting. The same applies to the other items: "Confirming Inquiry Details," "Customer Identification (Name, Date of Birth, etc.)," ​​"Handling the Call," and "Closing Greeting."

[0055] In the talk script shown in Figure 6, the inquiry process proceeds in the following order: "Initial greeting (opening)", "Confirmation of inquiry content", "Customer identity verification (name, date of birth, etc.)", "Response", and "Final greeting (closing)".

[0056] In the talk script shown in Figure 6, the script itself can be said to represent a single scenario. That is, the sequence of scripts "Thank you for calling..." → "Regarding the AAA service contract..." → "Your..." → "At the time of contract..." → "Thank you for calling..." represents a single scenario.

[0057] • Example 2 of a talk script In the talk script shown in Figure 7, similar to Figure 6, each item representing a scene or similar situation specifies the sentences, content, keywords, or phrases that the speaker (typically the operator; hereafter, the speaker will be assumed to be the operator) must utter. The number of turns for each item is also defined. A turn represents the exchange of utterances between the customer and the operator. For example, one turn is defined as the customer uttering a response to the operator's utterance, or the operator uttering a response to the customer's utterance. In the example shown in Figure 7, "Script (Example 1)", "Script (Example 2)", and "Script (Example 3)" each represent one talk script.

[0058] For example, the item "Opening" has a script (Example 1) that includes phrases like "Thank you for calling." This indicates, as in Figure 6, that the operator is required to say a sentence such as "Thank you for calling" during the opening.

[0059] Furthermore, for example, the item "Opening" has a script (Example 2) that specifies "Express gratitude." This indicates that the operator needs to express gratitude during the opening (for example, "Thank you very much," or "Thank you very much," etc.).

[0060] Furthermore, for example, the "Opening" section includes "Telephone" and "Thank you" as script examples (Example 3). This indicates that operators are required to include keywords (or phrases) such as "Telephone" and "Thank you" in their opening remarks.

[0061] Furthermore, the item "Opening" is defined as "the first 3 turns," meaning that the first 3 turns of the inquiry process constitute the opening.

[0062] The same applies to the other items: "Customer Verification," "Identity Verification," "Verification of a Phone Number Where You Can Call Back," and "Closing."

[0063] Here, in the examples shown in Figure 7, "Script (Example 1)" is a so-called "reading script type," "Script (Example 2)" is a so-called "item enumeration type" (or "utterance content enumeration type"), and "Script (Example 3)" is a so-called "keyword type." Generally, a script is often defined using one of these types, but a script may be defined using two or more types. For example, for a certain item in a talk script, both utterance content and keywords may be defined.

[0064] In the talk scripts shown in Figure 7, "Script (Example 1)," "Script (Example 2)," and "Script (Example 3)," each talk script represents a single scenario, similar to Figure 6. However, the order of some items in the script may vary. For example, "Could you tell me your date of birth?" and "Could you tell me your phone number?" may be in any order.

[0065] Returning to the explanation of Figure 5, following step S102, the response flow setting processing unit 201 creates a response flow from the scenario entered in step S102, saves the response flow to the storage unit 205, and visualizes it on the response flow setting screen (step S103). As a result, the response flow is set in the automatic response device 10, and the response flow is visualized on the response flow setting screen displayed on the operator terminal 20's display.

[0066] For example, if a talk script is input as a scenario in step S102 above, the response flow setting processing unit 201 can interpret the structure of the talk script and determine or estimate the scene, number of slots, scripts to be set in each slot, branching conditions, exception handling conditions, return conditions, branch destination scene when the branching conditions are met, branch destination scene when the exception handling conditions are met, and transition destination scene when the return conditions are met. For example, if two scripts are defined within a certain item (scene) of the talk script, the number of slots in the box for that scene can be determined or estimated to be "2". Also, although the talk scripts exemplified in Figures 6 and 7 do not have branching scenes, in the case of a talk script that includes scene branching (for example, a talk script represented by a tree structure, etc.), the branching conditions can be determined or estimated as branching conditions. The same applies to exception handling conditions; in the case of a talk script that includes scene branching related to exception handling, the branching conditions can be determined or estimated as exception handling conditions. The same applies to return conditions.

[0067] Here, an example of a response flow visualized on the response flow settings screen is shown in Figure 8. In the example shown in Figure 8, response flow 1101 is visualized on the response flow settings screen 1100. Furthermore, when each box on this response flow 1101 is selected (for example, by hovering the mouse over it), the slots, branching conditions, exception handling conditions, etc. contained in that box are visualized. In the example shown in Figure 8, the slots, branching conditions, exception handling conditions, etc. contained in the "Opening" box are visualized.

[0068] In this way, an operator (or a user other than an operator, such as a person in charge of setting up response flows) can set the response flow represented by a scenario in the automated response device 10 by specifying the scenario. Moreover, since an existing talk script can be specified as the scenario, the operator can set up the response flow in the automated response device 10 in a simple and efficient manner. This reduces the cost required to set up the response flow. Of course, the operator may also input a tree structure with directed edges where each box is a node, scenes, the number of slots, scripts to be set in each slot, branching conditions, exception handling conditions, return conditions, branch destination scenes when branching conditions are met, branch destination scenes when exception handling conditions are met, transition destination scenes when return conditions are met, etc., by operating a GUI (Graphical User Interface) or the like.

[0069] <Automated response processing> An example of the automated response processing according to this embodiment will be described below with reference to Figure 9. In the following description, it will be assumed that a response flow is set in the automated response device 10 (i.e., the response flow is stored in the storage unit 205).

[0070] The automated response processing unit 202 sends a question to the customer terminal 30 according to the response flow stored in the memory unit 205 (step S201). For example, in the "opening" scene of a text chat, it might send a question such as, "What is your inquiry? Please select from the following..." Alternatively, for example, speech synthesis technology could be used to read the scenario aloud and output the question as synthesized speech.

[0071] Next, the automated response processing unit 202 receives the answer to the question sent in step S201 from the customer terminal 30 (step S202).

[0072] Next, the automated response processing unit 202 determines whether escalation is necessary (step S203). Escalation is necessary when an event occurs that is difficult to handle with automated responses, for example, when the response flow described above cannot respond to the answer received in step S202. However, in addition to the above, escalation may also be determined to be necessary if, for example, the customer's emotions are negative as a result of using known emotion estimation technology (for example, if the customer is angry or dissatisfied). In addition to the above, escalation may also be determined to be necessary if, in general, escalation to a human is necessary regarding the automated response (for example, if the customer wishes to speak with an operator).

[0073] If it is not determined in step S203 above that escalation is necessary, the automated response processing unit 202 determines whether the customer interaction has been completed (step S204). Here, it is assumed that the customer interaction has been completed if it has been successfully concluded. Successful conclusion of customer interaction includes, for example, when all of the customer's inquiries have been resolved and the voice call with the customer has been disconnected, or when the text chat with the customer has ended. On the other hand, if, for example, at least some of the customer's inquiries have not been resolved, or if, as a result of sentiment estimation, the customer has some kind of dissatisfaction, the customer interaction may be considered not to have been successfully concluded, and escalation may be carried out.

[0074] If it is not determined in step S204 that the customer interaction has been completed, the automated response processing unit 202 returns to step S201. This repeats the process of sending questions to the customer and receiving answers according to the interaction flow, and the customer inquiry is handled automatically.

[0075] On the other hand, if it is determined in step S203 that escalation is necessary, the escalation processing unit 203 requests the operator terminal 20 to handle the customer and displays a response status screen on the operator terminal 20 that visualizes the response status to date (step S205). This allows the operator to easily understand the status of the automated response to the customer on this response status screen, enabling them to take appropriate action quickly. As a result, the cost required for customer support after escalation can be reduced. The content of the automated response, the escalation location (for example, the scene or slot in which the escalation occurred), and the operator's response after the escalation are stored in the storage unit 205 as response result information. A specific example of the response status screen is shown below.

[0076] • Specific example of the response status screen (Part 1) In the response status screen 2100 shown in Figure 10, the response status 2101 of a customer in which an escalation occurred is visualized. In this response status 2101, scenes in which customer support was provided by automated responses and scenes in which customer support was not provided are visualized in a way that makes them distinguishable. Furthermore, scenes in which an escalation occurred are visualized in a way that makes them identifiable. As a result, in the example shown in Figure 10, the operator can see that customer support was provided in the following stages: "Opening" → "Customer Confirmation 1" → "Identity Verification 2", and that an escalation occurred in "Identity Verification 2". In the example shown in Figure 10, scenes in which customer support was not provided are also displayed, but for example, scenes in which customer support was not provided do not need to be displayed, or scenes in which customer support was not provided are displayed in a simplified manner (for example, only the scene name is displayed). On the other hand, for example, the flow of scenes in which customer support was provided (in the example shown in Figure 10, "Opening" → "Customer Confirmation 1" → "Identity Verification 2") may be highlighted. Furthermore, for example, boxes or arrows along the route may be highlighted along the flow of the scenes.

[0077] Furthermore, selecting each box on the response status 2101 (for example, by hovering the mouse over it) visualizes the slots, branching conditions, exception handling conditions, etc. contained within that box. In the example shown in Figure 8, the slots, branching conditions, exception handling conditions, etc. contained within the "Opening" box are visualized. However, this is just one example; for example, in the response status 2101, only the answers in the slots contained within the boxes of scenes where customer support was provided by automated responses may be visualized at all times, or both the questions and answers may be visualized at all times, or information summarizing the questions and answers may be visualized at all times. In addition, other information useful for operators to handle customer support after escalation (for example, the product / service name related to the inquiry, the reason for escalation, customer personal information obtained from the CRM system, etc.) may also be visualized.

[0078] This allows operators to easily find out what kind of customer support was provided through automated responses for a particular customer that has been escalated.

[0079] • Specific examples of the response status screen (Part 2) In the customer service status screen 2200 shown in Figure 11, the customer service status 2201 of a customer who has experienced an escalation is visualized. Unlike Figure 10, the customer service status screen 2200 in Figure 11 also visualizes the candidate branching scenes. Candidate branching scenes are scenes to which the operator could have transitioned using the branching conditions included in the boxes of the scenes to which customer service was performed, even though no customer service was performed in those scenes. For example, in the example shown in Figure 11, it was possible to transition from "Opening" to three scenes: "Customer Confirmation 1," "Customer Confirmation 2," and "Customer Confirmation 3," but the operator actually transitioned only to "Customer Confirmation 1." In this case, "Customer Confirmation 2" and "Customer Confirmation 3" become candidate branching scenes. When an escalation occurs, transitioning to such candidate branching scenes may have been the correct customer service action. Therefore, by presenting the operator with candidate branching scenes, it is thought that the operator can perform a more appropriate customer service action.

[0080] Furthermore, the response status screen 2200 shown in Figure 11 may be the same as the response status screen 2100 shown in Figure 10, except for the parts specifically mentioned.

[0081] • Specific examples of response status screens (Part 3) The response status screen 2300 shown in Figure 12 visualizes the response status 2301 for multiple customers where escalation has occurred. For example, in the example shown in Figure 12, for customers A, B, C, X, and Y, the automated response process of "Opening" → "Customer Verification 1" → "Identity Verification 2" is performed, and it is visualized that an escalation occurred at "Identity Verification 2". Also in the example shown in Figure 12, for customers D and E, the automated response process of "Opening" → "Customer Verification 2" is performed, and it is visualized that an escalation occurred at "Identity Verification 2".

[0082] Furthermore, in response status 2301, information indicating that customers X and Y require "caution" is visualized. This indicates, for example, that customers X and Y should be treated with caution (e.g., customers who require careful handling) based on information obtainable from a CRM system, etc. Alternatively, information indicating that customers require "caution" may be visualized based on, for example, the results of sentiment estimation technology, the urgency of the customer's inquiry, etc., rather than a CRM system.

[0083] This allows operators to easily understand the status of multiple customer interactions via automated responses, enabling them to simultaneously handle multiple customers who have received the same customer support via automated responses (in other words, multiple customers who made the same inquiry). This reduces the number of operators needed to handle escalations, leading to expected cost savings. To handle multiple customers simultaneously, for example, in the case of text chat, this can be done by sending a mass message based on the operator's instructions, or in the case of voice calls, this can be done by outputting synthesized speech based on the operator's text input or speech. Furthermore, for example, a sentence for simultaneously handling multiple customers may be generated using natural language processing technology based on the operator's instructions or one or more keywords or ordered word sequences provided by the operator, and then the message or synthesized speech representing that sentence may be used to simultaneously handle those multiple customers.

[0084] Furthermore, warning information such as "Caution" is made visible for customers who require careful attention, making it possible to handle only these customers individually and thus preventing a decline in customer satisfaction. In addition to the information shown in Figure 12, various other types of information may be presented. For example, information such as who initiated the escalation, whether from the chatbot or IVR system or the customer side, the customer's emotions and satisfaction level that matched the customer's wishes or the conditions set in the system, the reason and circumstances of the escalation, the scene or slot in which the escalation occurred, the time elapsed since the customer last responded, and the priority level such as urgency may be presented. Moreover, when handling multiple customers simultaneously or when multiple escalations are progressing in parallel, the status of each customer's interaction may be dynamically sorted and displayed using attribute information related to the interaction, such as the emotions and satisfaction level of each customer being handled, the content of each interaction, the priority and urgency of the interaction, and whether or not an escalation occurred.

[0085] Furthermore, the response status screen 2300 shown in Figure 12 may be the same as the response status screen 2100 shown in Figure 10, except for the parts specifically mentioned.

[0086] Furthermore, the presentation of customer response status via automated responses, such as the response status screen 2300 shown in Figure 12, is not limited to escalation; for example, it may be presented at times other than escalation. For instance, it may be presented to allow users, such as operators, to understand customer response information.

[0087] <Response Flow Correction Process> The following describes an example of the response flow modification process with reference to Figure 13. In the following, it is assumed that at least one response result information is stored in the storage unit 205.

[0088] The response flow correction processing unit 204 receives a response flow correction instruction from the operator terminal 20 (step S301).

[0089] Next, the response flow correction processing unit 204 obtains the response result information from the storage unit 205 (step S302).

[0090] Next, the response flow correction processing unit 204 uses the response result information obtained in step S302 to estimate the parts of the response flow that need to be corrected (step S303). A specific example of the method for estimating the parts of the response flow that need to be corrected will be described later.

[0091] Next, the response flow modification processing unit 204 modifies the parts of the response flow that need to be modified using the response result information obtained in step S302 (step S304). The modification includes adding a box (scene) or slot to the part to be modified, changing a box (scene) or slot in the part to be modified, or deleting a box (scene) or slot in the part to be modified.

[0092] Here, the response flow modification processing unit 204 can estimate and modify the parts of the response flow that need modification, for example, by the following method.

[0093] • Estimation of areas requiring correction and methods for correction (Part 1) Step 1-1: The response result information includes the escalation location (for example, the scene or slot where the escalation occurred). Therefore, the response flow correction processing unit 204 estimates that the escalation location is the location to be corrected. For example, if an escalation occurs in a slot contained within a box in a certain scene, that scene and slot are estimated to be the locations to be corrected. In addition, for example, if an escalation occurs immediately after a scene transition or in the first slot after a scene transition, the scene or slot with a high degree of similarity may be estimated as the location to be corrected based on the operator's response content.

[0094] Step 1-2: The response result information includes the operator's response after the escalation (for example, the spoken text or text entered in the chat that represents the response). Therefore, the response flow correction processing unit 204 identifies the scene and question represented by this response from within the response flow using natural language processing or the like.

[0095] Step 1-3: If the scene identified in Step 1-2 is the same as the scene to be modified, the response flow modification processing unit 204 adds the slot containing the question identified in Step 1-2 as the next slot after the slot to be modified. If, for example, adding this slot necessitates modifications to other slots or boxes (including slot expansion), those modifications will also be performed.

[0096] Step 1-4-1: On the other hand, if the scene identified in Step 1-2 above is different from the scene to be corrected, the response flow correction processing unit 204 adds a box to the response flow that is the box of the scene identified in Step 1-2 above and has a slot set with the question identified in Step 1-2 above. The response flow correction processing unit 204 also sets a branching condition for this added box, which is a branching condition to a destination box identified from the operator's response after escalation. This destination box may be identified by natural language processing or the like from the operator's response after escalation, as described above, or it may be identified as the box of the first scene after returning to the automated response if the system returns to the automated response after the operator's response.

[0097] Step 1-4-2: The response flow modification processing unit 204 sets a condition for the branching conditions included in the scene box estimated in Step 1-1 above, stating that if the answer in the slot estimated in Step 1-1 above is filled, the system will transition to the box added in Step 1-4-1 above.

[0098] • Estimation of areas requiring correction and methods for correction (Part 2) Step 2-1: The response flow correction processing unit 204 estimates the escalation location as the correction location (scene and slot), similar to Step 1-1 above.

[0099] Step 2-2: The response result information includes the operator's response after escalation (for example, the spoken text or text entered in the chat that represents the response). The response flow modification processing unit 204 then matches the text representing the operator's response after escalation with the script set in the slots included in each box of the response flow. Here, such matching can be performed using known natural language processing. For example, matching can be done by solving the elastic matching problem, or a pre-trained language model (for example, BERT: Bidirectional Encoder Representations from Transformers) can be fine-tuned for the question answering task and then matched using that language model.

[0100] Step 2-3: If a slot exists that matches the matching in Step 2-2 above, the response flow modification processing unit 204 modifies the modification location estimated in Step 2-1 above so that this slot becomes the destination. In other words, the response flow modification processing unit 204 sets a condition for the branching conditions included in the scene box estimated in Step 2-1 above, stating that if the answer for the slot estimated in Step 2-1 above is filled, the system will transition to the matching slot.

[0101] Step 2-4-1: On the other hand, if there is no slot that matches the matching in Step 2-2 above, the response flow correction processing unit 204 identifies the scene and question from the operator's response after escalation, similar to Step 1-2 above.

[0102] Step 2-4-2: Next, the response flow modification processing unit 204 adds a box to the response flow that is a scene box identified in Step 2-4-1 and has a slot set with the question identified in Step 2-4-1, similar to Step 1-4-1 above. The response flow modification processing unit 204 also sets a branching condition for this added box, which is a branching condition to the destination box identified from the operator's response after escalation.

[0103] Step 2-4-3: The response flow modification processing unit 204, similar to Step 1-4-2 above, sets a condition for the branching conditions included in the scene box estimated in Step 2-1 above, stating that if the answer in the slot estimated in Step 2-1 above is filled, the system will transition to the box added in Step 2-4-2 above.

[0104] • Estimation of areas requiring correction and methods for correction (Part 3) Step 3-1: The response flow modification processing unit 204 uses the response result information to classify the text representing the operator-customer interaction after escalation into groups from the beginning. Here, the groups correspond to boxes (scenes), the operator-customer interaction corresponds to slots, and each text representing the operator-customer interaction after escalation is separated by scene transitions based on branching conditions and exception handling conditions. The response flow modification processing unit 204 classifies the text representing the operator-customer interaction into one or more groups based on comparison with existing scenarios, talk scripts, and response flows, the relationships between texts representing the operator's interaction, the similarity between texts representing the operator's interaction and existing scene slots, the similarity with the slot configuration and order of existing scenes, the similarity with branching conditions, exception handling conditions, return conditions, and other topics and scene transitions in the operator-customer interaction.

[0105] Step 3-2-1: The response flow correction processing unit 204 identifies the areas to be corrected (scenes and slots) for the first group. Next, the response flow correction processing unit 204 identifies the transitions between scenes from the scene where the escalation occurred until the customer escalates, in order, until a scene or slot to be corrected is found. Candidate areas to be corrected are ranked based on the similarity between the response content expressed in the text of the first group and the representation in the slots within the scene, the omission or excess of the response content and slots, the similarity between the response content and the order of multiple slots, and whether the series of operator and customer interactions when transitioning to the next group match the branching conditions, exception handling conditions, return conditions, etc., and the scene and slot that best matches are selected as the scene and slot to be corrected.

[0106] For example, if the operator's response is similar to or related to a slot in the current scene, that slot may be designated as the area to be corrected. If the response is not similar to or related to a slot in the current scene, the area to be corrected may be determined by considering that the area to be corrected is likely to be in a scene or slot prior to the escalation. In this case, the area to be corrected may be scored based on the similarity, relevance, and inconsistencies between the operator's response after the escalation and the slot, and the one with the highest score may be designated as the area to be corrected. However, if the score falls below a predetermined threshold or no corresponding slot is found, the search area may be expanded to include scenes prior to the current scene from the customer's scene transition history. Alternatively, the area to be corrected may be determined after scoring all scenes in the customer's scene transition history and selecting the one with the highest score.

[0107] Step 3-2-2: The response flow correction processing unit 204 identifies whether the second group corresponds to an existing scene. Similar to step 3-2-1 performed for the first group, this can be determined based on whether the group matches an existing scene, the similarity between the group and the corresponding slot within the group, branching conditions, exception handling conditions, etc. If no existing scene corresponding to the second group is found, it is determined that a new scene needs to be created. At this stage, the correction location may also be identified again based on the determination result of the transition destination for the second group. If the transition destination scene does not match, it may be excluded from the list of candidates for correction.

[0108] Step 3-3: The response flow correction processing unit 204 determines the correction content and makes the correction once it has identified the correction location and the destination of the next scene. The content of the operator-customer interaction after escalation is analyzed using natural language processing technology as explained above. An example of the correction method is described below.

[0109] • Slot correction (Part 1): If the interaction between the operator and the customer after escalation involves confirming the meaning and wording of the question in the slot that needs correction, or what kind of answer should be given, then one of the following actions should be taken.

[0110] Correction A: If the customer's answer is among the current slot options, modify the wording and content of the slot in question. No changes are required to existing branching conditions, exception handling conditions, return conditions, or transition destinations.

[0111] Correction Response b: If the answer is not among the current choices, in addition to correcting the expression and content of the slot for the correction, correct or add branching conditions, exception handling conditions, and return conditions corresponding to the newly added choices, and set the transition destination to an existing scene or a newly added scene corresponding to the second group that has already been identified.

[0112] • Slot modification (part 2): If the interaction between the operator and customer after escalation involves additional questions and answers, add the questions asked by the operator as a slot and the customer's answers as options after the modified slot in the existing scene. Modify branching conditions, exception handling conditions, and return conditions as needed, and set the destination to an already identified scene. If the destination is a new scene, add the new scene. In other words, take one of the following actions:

[0113] Correction c: If the answer is among the current options, add a slot. Branching conditions and exception handling conditions will be modified or added as needed, taking into account the customer's response to the added slot, and the transition destination will be set to the already identified transition scene.

[0114] Correction d: If the answer is not among the current options, add a slot, and modify or add branching conditions, exception handling conditions, and return conditions considering the customer's response to the added slot, and set the destination to the slot scene of the already identified destination.

[0115] • Correction of the response: If the content of the operator-customer interaction after escalation is not a question but a response, or if there is a problem with the scene transition control such as branching conditions, exception handling conditions, or return conditions based on the response, correct the relevant part. In other words, take one of the following actions:

[0116] Correction e: If the customer's answer to the operator's question is not already set as a response to the slot, add the customer's answer to the answer (option) for the slot. Furthermore, branching conditions, exception handling conditions, and return conditions are modified or added considering the customer's answer to the added slot, and the transition destination is set to the already identified transition destination slot / scene.

[0117] Correction f: If the content of the operator-customer interaction after escalation is unrelated to the questions and answers asked before the escalation, there is a problem with the branching conditions, exception handling conditions, return conditions, or destination scene settings. In this case, the transition to the scene where the escalation occurred is incorrect; the correct transition should be to another scene at some point among the scenes the customer transitioned to. Therefore, the correction is needed in the scene before the escalation, as the transition to a scene matching the second group is set, and this should be corrected. Alternatively, identify and correct the parts of the branching conditions, exception handling conditions, return conditions, and destination scene that need correction based on the scene slot and its relationship to the second group. In addition, in correction actions a-f, if branching conditions, exception handling conditions, return conditions, and transition destinations are not used to identify the areas to be corrected, or if the conditions for identifying the second group and the existing scene are not used, the branching conditions, exception handling conditions, return conditions, and transition destinations may differ. In such cases, corrections may be made based on the response after escalation, or the operator may be presented with areas that are presumed to require correction and proposed corrections for confirmation and correction.

[0118] We have now processed the first set of corrections. From here on, we will recursively process the data from the second group to the last group.

[0119] Procedure 3-4-1: The response flow correction processing unit 204 shall process the following groups:

[0120] Step 3-4-2: The response flow correction processing unit 204, similar to step 3-2-2, identifies the transition scene for the next group after the group being processed.

[0121] Procedure 3-4-3: The response flow correction processing unit 204 sets the necessary settings for the group to be processed, such as branching conditions, exception handling conditions, and return conditions, by one of the following methods.

[0122] Correction g: For example, if the group following the group being processed corresponds to an existing scene, modify the branching conditions, exception handling conditions, and return conditions of the group being processed so that the existing scene becomes the destination.

[0123] Correction h: If the group following the group being processed does not correspond to an existing scene, add the group following the group being processed as a new scene and set slots, branching conditions, exception handling conditions, and return conditions.

[0124] Repeat steps 3-4-1 through 3-4-3 until there are no more groups to process.

[0125] The above processing example automatically corrects only the parts of the response flow that arise from newly added options, etc., resulting from escalation inquiries or customer responses. For example, when a question is added as a new slot, the options the customer answered, the corresponding branching conditions, exception handling conditions, and transition destinations can be set, but everything else remains undefined. Therefore, in response to additions or modifications to answer options, branching conditions, exception handling conditions, and return conditions related to the modified areas, the system may present the operator with other areas that need to be reviewed, such as reviewing other answer options, branching conditions, exception handling conditions, and return conditions, or it may present the operator with suggested modifications, modification options, or requests for input related to the presented areas, which can be inferred from similar scenes or slots. The timing of presenting areas for review can be done during the individual modification process for each group, or after all groups have been processed. Regarding suggested modifications such as areas for review, the system may display the entire scene transition history for the customer handled this time, including the escalation area, the modified area, and the scene transition after the modification, or it may display the differences in scene transitions before and after the modification in a way that allows switching or side-by-side comparison. Furthermore, when presenting revisions or revision proposals, it is desirable to suppress the display of information that is not affected by the revisions or that does not directly affect the customer's scene transitions, selectively display information such as the scene name, customer responses and summaries, and information regarding transition conditions, and to present the information in a way that allows for efficient revision and revision confirmation.

[0126] Furthermore, while the above processing example presented the response flow in the form of boxes (scenes) and slots, it may also be presented as a scenario, which is a collection of scripts that add responses and answers from the other party, branching conditions, exception handling conditions, return conditions, and destination scenes to a scene-based talk script as shown in Figure 7. In this case, the script such as lines, action items, utterances, keywords, etc., can be generated from the response flow scenario using natural language processing.

[0127] While the above describes visualizing and displaying all information, this is not necessarily the only option. Information may be presented using audio or other means besides visualization, or by combining multiple presentation methods.

[0128] • Estimation of areas requiring correction and methods for correction (Part 4) In the method for estimating and correcting correction areas (part 3), the process was carried out sequentially starting from the first group in step 3-2-1. However, it is also possible to process the correspondence of the last group first, and then finally identify and correct the correction areas (scenes and slots) corresponding to the first group. Specifically, the response flow correction processing unit 204, in step 3-1, classifies the text representing the operator and customer interaction after escalation from the response result information into groups sequentially from the beginning. Then, it proceeds through steps 3-4-1 to 3-4-3 sequentially from the last group to the second group. The difference here from the method for estimating and correcting correction areas (part 3) is that, since the correspondence between the groups and the destination scenes has already been identified, the correspondence between the previous group and the source scene is identified. The correction is completed by performing steps 3-2-1 to 3-3 for the first and second groups, excluding step 3-2-2.

[0129] However, the above estimations and methods for correcting the correction areas are merely examples and are not limited to them. For example, unnecessary scenes or slots may be deleted based on the response result information.

[0130] Returning to the explanation of Figure 13, following step S304, the response flow correction processing unit 204 displays a response flow correction screen on the operator terminal 20, in which the corrected response flow is visualized (step S305). An example of a response flow correction screen in which the corrected response flow is visualized is shown in Figure 14. In the response flow correction screen 3100 shown in Figure 14, the response flow 3101 corrected in step S304 is visualized. In this response flow 3101, the scenes corrected in step S304 (in the example shown in Figure 14, "Customer Confirmation 1" and "Identity Confirmation 4") are visualized in a different display manner than the scenes that were not corrected. This allows the operator to know the corrected response flow and the details of the corrections.

[0131] In this way, operators (or other users, such as those responsible for modifying response flows) can modify the response flow. Moreover, by using the response result information from when an escalation occurred, the response flow can be automatically modified to prevent similar escalations from occurring in the past, thereby reducing the cost that operators would otherwise have to pay to modify the response flow.

[0132] The following additional information is disclosed regarding the embodiments described above. (Note 1) Memory and At least one processor connected to the memory, Includes, The aforementioned processor, Based on pre-configured response flow information, the system automatically handles incoming inquiries. If the entity responsible for handling the inquiry changes, the status of the inquiry handled by the automated system will be presented to the designated recipient. An automated response system that modifies the response flow information based on response result information, including the response result of the inquiry response after the execution entity has been switched. (Note 2) The aforementioned processor, The automated response device described in Appendix 1 switches the entity responsible for handling the inquiry when an event occurs that makes it impossible to handle the inquiry based on the response flow information, or when a request is made to switch the entity responsible for handling the inquiry. (Note 3) The aforementioned processor, Using the scenario for handling inquiries as input, the response flow information is created. An automated response device as described in Appendix 1 or 2, which presents the response flow information for handling the inquiry to a designated recipient from among the created response flow information. (Note 4) The aforementioned processor, An automated response device according to Appendix 1 or 2, which presents, in different ways, the portion of the response flow represented by the response flow information described above that was responded to by the automated execution during the inquiry response and the portion that was not responded to by the automated execution. (Note 5) The aforementioned processor, An automated response device according to Appendix 1 or 2, which determines the location and content of the correction in the response flow represented by the response flow information based on the response result information, and corrects the response flow information based on the location and content of the correction. (Note 6) The response result information includes at least information representing the point in the response flow represented by the response flow information where the entity executing the inquiry response changed, and information representing the content of the inquiry response executed by the new entity. The aforementioned processor, The automated response device described in Appendix 5 determines the point where the entity executing the inquiry response changes as the point of modification, and determines the content of the modification based on the content of the inquiry response executed by the entity executing the inquiry response after the change. (Note 7) The aforementioned processor, An automated response device as described in Appendix 1 or 2, which, with respect to one or more inquiries whose response status is the same until the entity responsible for handling the inquiries changes, collectively instructs the new entity to handle the one or more inquiries. (Note 8) Memory and At least one processor connected to the memory, Includes, The aforementioned processor, Based on pre-configured response flow information, the system automatically handles incoming inquiries. If the entity responsible for handling the inquiry changes, the status of the inquiry handled by the automated response unit is presented to a designated recipient. An automated response system that, with respect to one or more inquiries whose response status is the same until the entity responsible for handling the inquiry changes, instructs the new entity to handle the response to one or more of those inquiries collectively. (Note 9) A non-temporary storage medium that stores a program executable by a computer to perform automated response processing, The aforementioned automated response process is: Based on pre-configured response flow information, the system automatically handles incoming inquiries. If the entity responsible for handling the inquiry changes, the status of the inquiry handled by the automated system will be presented to the designated recipient. A non-temporary storage medium that modifies the response flow information based on response result information, including the response result of the inquiry response after the execution entity has been switched. (Note 10) A non-temporary storage medium that stores a program executable by a computer to perform automated response processing, The aforementioned automated response process is: Based on pre-configured response flow information, the system automatically handles incoming inquiries. If the entity responsible for handling the aforementioned inquiry changes, the status of the inquiry response will be presented to the designated recipient. A non-temporary storage medium that, with respect to one or more inquiries whose response status is the same until the entity executing the inquiry response changes, instructs the new entity to handle one or more inquiries collectively.

[0133] The present invention is not limited to the embodiments specifically disclosed above, and various modifications, changes, and combinations with known technologies are possible without departing from the scope of the claims. [Explanation of Symbols]

[0134] 1. Contact Center System 10. Automated response system 20 Operator terminals 30 Customer terminals 40 Communication Networks 101 Input Device 102 Display device 103 External I / F 103a Recording medium 104 Communication I / F 105 RAM 106 ROM 107 Auxiliary storage 108 processors 109 Bus 201 Response Flow Setting Processing Unit 202 Automated Response Processing Unit 203 Escalation Processing Unit 204 Response Flow Correction Processing Unit 205 Storage section

Claims

1. An automated response unit is configured to automatically handle received inquiries based on pre-configured response flow information, When the entity responsible for handling the inquiry changes, the presentation unit is configured to display the status of the inquiry response by the automated response unit to a predetermined recipient. A modification unit is configured to modify the response flow information based on response result information, including the response result of the inquiry response after the execution entity has been switched. An automatic response device having the following features.

2. The aforementioned automatic response unit The automated response device according to claim 1, configured to switch the entity executing the inquiry response when an event occurs that makes it impossible to respond to the inquiry based on the response flow information, or when a request is made to switch the entity executing the inquiry response.

3. A creation unit is configured to create the aforementioned response flow information using a scenario related to handling inquiries as input, The aforementioned display unit is, The automated response device according to claim 1 or 2, wherein the response flow information for handling the inquiry, from among the response flow information created by the creation unit, is configured to be presented to a predetermined recipient.

4. The aforementioned display unit is, The automated response device according to claim 1 or 2, wherein the response flow represented by the response flow information is configured to present in different ways the portion that was responded to by the automated response unit during the inquiry handling and the portion that was not responded to by the automated response unit.

5. The aforementioned modification section is, The automated response device according to claim 1 or 2, which is configured to determine the location and content of the correction in the response flow represented by the response flow information based on the response result information, and to correct the response flow information based on the location and content of the correction.

6. The response result information includes at least information representing the point in the response flow represented by the response flow information where the entity executing the inquiry response changed, and information representing the content of the inquiry response executed by the new entity. The aforementioned modification section is, The automated response device according to claim 5, configured to determine the point where the entity executing the inquiry response changes as the point of modification, and to determine the content of the modification based on the content of the inquiry response executed by the entity executing the inquiry response after the change.

7. The automated response device according to claim 1 or 2, further comprising an instruction unit configured to collectively instruct the new implementing entity to respond to one or more inquiries whose response status is common until the implementing entity for the inquiry response changes.

8. An automated response unit is configured to automatically handle received inquiries based on pre-configured response flow information, When the entity responsible for handling the inquiry changes, the presentation unit is configured to display the status of the inquiry response by the automated response unit to a predetermined recipient. An instruction unit is configured to collectively instruct the new implementing entity to handle one or more inquiries that share the same response status until the implementing entity switches over, with respect to one or more inquiries that share the same response status until the new implementing entity switches over. An automatic response device having the following features.

9. An automated response procedure that automatically executes responses to received inquiries based on pre-configured response flow information, If the entity responsible for handling the inquiry changes, a presentation procedure is provided to present the status of the inquiry handling by the automated response procedure to a designated recipient. A correction procedure for correcting the response flow information based on response result information including the response result of the inquiry response after the execution entity has been switched, An automated response method performed by a computer.

10. An automated response procedure that automatically executes responses to received inquiries based on pre-configured response flow information, If the entity responsible for handling the inquiry changes, a presentation procedure is provided to present the status of the inquiry handling by the automated response procedure to a designated recipient. A procedure for giving instructions to the new implementing entity to collectively instruct the handling of one or more inquiries that share the same response status until the implementing entity switches over, An automated response method performed by a computer.

11. A program that causes a computer to function as an automatic response device as described in claim 1.

Images