Inquiry processing device, inquiry processing method, and program

Abstract

Previously, the inability to properly handle user inquiries about the target area meant that user inquiries were less likely to be useful in improving business operations and other areas. [Solution] An inquiry processing device 1 comprises an inquiry reception unit 121 that receives inquiry information, which is information about user inquiries from one or more channels; a department determination means 1312 that uses the department information of two or more departments stored in a department management unit 111 that stores department information for two or more departments to determine the department corresponding to the inquiry information received by the inquiry reception unit 121 and obtain the department identifier of the department; and a department inquiry processing unit 132 that processes the inquiry information using the department identifier obtained by the department determination means 1312. As a result of not being able to properly process user inquiries about a target, user inquiries can be used to improve management, etc.

Description

Technical Field

[0001] The present invention relates to an inquiry processing apparatus that processes inquiries from users and the like.

Background Art

[0002] Conventionally, there has been a call center operation support system that stores inquiry contents, inquiry histories, and customer information in a storage device, and reads out necessary data and displays it on an operator terminal when a call is received from a customer (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the prior art, inquiries from users regarding a target could not be appropriately processed, and as a result, inquiries from users were unlikely to be usefully linked to improvements in management and the like.

Means for Solving the Problems

[0005] The inquiry processing apparatus of the first invention includes an inquiry reception unit that receives inquiry information, which is information on inquiries from a user from one or two or more channels, and a department management unit that stores department information regarding two or more departments. Department determination means for determining a department corresponding to the inquiry information received by the inquiry reception unit using the department information of each of the two or more departments and obtaining a department identifier of the department, and a department inquiry processing unit for processing the inquiry information using the department identifier obtained by the department determination means. It is an inquiry processing apparatus.

[0006] With such a configuration, the target department for an inquiry from a user regarding a target can be automatically determined.

[0007] Furthermore, the inquiry processing device of the second invention, compared to the first invention, further comprises: an inquiry receiving unit that receives inquiry information from two or more channels, an inquiry storage unit that stores the inquiry information received by the inquiry receiving unit in an inquiry management unit that stores one or more pieces of inquiry information paired with a channel identifier that identifies the channel of the inquiry, a statistical processing unit that acquires statistical processing results for the inquiry information for each of the two or more channel identifiers, and a statistical output unit that outputs the statistical processing results acquired by the statistical processing unit.

[0008] This configuration allows for obtaining statistical processing results for queries from each query channel.

[0009] Furthermore, the inquiry processing device of this third invention, compared to the first invention, further comprises: an inquiry receiving unit that receives two or more inquiry pieces of information; a classification determination unit that uses part or all of the two or more inquiry pieces of information received by the inquiry receiving unit to obtain two or more reasons for the inquiry pieces of information; a classification unit that determines one or more reasons from the two or more reasons obtained by the classification determination unit to which each of the two or more inquiry pieces of information received by the inquiry receiving unit belongs; and an inquiry storage unit that stores each of the two or more inquiry pieces of information received by the inquiry receiving unit in an inquiry management unit that stores one or more inquiry pieces of information in association with the one or more reasons determined by the classification unit.

[0010] With this configuration, it is possible to automatically determine two or more reasons for a query using two or more queries, and to classify two or more queries using two or more reasons.

[0011] Furthermore, the inquiry processing device of the fourth invention, compared to the first invention, further comprises: an inquiry receiving unit that receives inquiry information from two or more channels, a group determination unit that groups the two or more inquiry information received by the inquiry receiving unit and determines two or more groups, a representative determination unit that determines a representative inquiry information which is representative of the group for each of the two or more groups, and a representative output unit that outputs the representative inquiry information determined by the representative determination unit.

[0012] This configuration allows for grouping two or more queries into two or more groups, and determining a representative query within each of those two or more groups.

[0013] Furthermore, the inquiry processing device of the fifth invention is an inquiry processing device that, for any one of the first to fourth inventions, refers to a page storage unit that stores one or more web pages relating to the subject of the inquiry, and uses the inquiry information received by the inquiry reception unit to obtain improvement suggestions for one or more web pages, and further comprises an improvement output unit that outputs improvement suggestions.

[0014] This configuration allows for suggestions to improve web pages based on user inquiries about the target topic.

[0015] Furthermore, the inquiry processing device of this sixth invention, in contrast to the fifth invention, is an inquiry processing device in which the inquiry receiving unit also receives response information corresponding to the inquiry information, and the improvement acquisition unit uses the inquiry information and response information received by the inquiry receiving unit to acquire improvement proposals.

[0016] This configuration allows for suggestions to improve web pages using user inquiries and responses regarding the subject matter.

[0017] Furthermore, the inquiry processing device of the seventh invention further comprises a determination unit that determines whether the inquiry information received by the inquiry receiving unit satisfies the processing conditions, with respect to any one of the first to sixth inventions, and the department determination means is an inquiry processing device that acquires a department identifier only for inquiry information that the determination unit has determined satisfies the processing conditions.

[0018] This configuration allows for the assignment of a department only to inquiries that require processing. [Effects of the Invention]

[0019] According to the inquiry processing device of the present invention, if user inquiries regarding a subject are not processed appropriately, the user inquiries can be used to improve management and other aspects of business operations.

Brief Description of the Drawings

[0020] [Figure 1] Conceptual diagram of information system A in Embodiment 1 [Figure 2] Block diagram of the same information system A [Figure 3] Block diagram of the inquiry processing device 1 [Figure 4] Flowchart for explaining an operation example of the inquiry processing device 1 [Figure 5] Flowchart for explaining an example of the judgment processing [Figure 6] Flowchart for explaining an example of the inquiry analysis processing [Figure 7] Flowchart for explaining an example of the department decision processing [Figure 8] Flowchart for explaining an example of the organization analysis processing [Figure 9] Flowchart for explaining an example of the reason creation processing [Figure 10] Flowchart for explaining an example of the classification processing [Figure 11] Flowchart for explaining an example of the grouping processing [Figure 12] Flowchart for explaining an example of the representative determination processing [Figure 13] Flowchart for explaining an example of the statistical processing [Figure 14] Flowchart for explaining an example of the improvement proposal acquisition processing [Figure 15] Diagram showing an example of the department management table [Figure 16] Diagram showing an example of the inquiry management table [Figure 17] Diagram showing an example of the prompt management table [Figure 18] Diagram showing an example of the prompt management table [Figure 19] Diagram showing an example of the improvement proposal[[ID=6I]] [Figure 20] Block diagram of the same computer system [Modes for carrying out the invention]

[0021] The following describes embodiments of the inquiry processing device and the like with reference to the drawings. Note that components denoted by the same reference numerals in these embodiments perform similar operations, and therefore, further explanation may be omitted.

[0022] (Embodiment 1) This embodiment describes a query processing device that receives query information from two or more channels, stores the query information in pairs with channel identifiers, and performs statistical processing for each channel.

[0023] This embodiment describes an inquiry processing device that obtains two or more reasons based on some or all of the two or more inquiry pieces received.

[0024] In this embodiment, a query processing device that determines one or more reasons for each of two or more query pieces of information will be described.

[0025] In this embodiment, a query processing device that groups two or more query information and determines a representative query within each group will be described.

[0026] In this embodiment, an inquiry processing device that makes suggestions for improving a web page based on inquiry information will be described. Note that the inquiry processing device may also use the response to the inquiry information in addition to the inquiry information to make suggestions for improving the web page.

[0027] In this embodiment, a query processing device that excludes query information that does not require processing will be described.

[0028] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is irrelevant. Information X and information Y may be linked, may reside in the same buffer, may information X be contained in information Y, or information Y may be contained in information X, and so on.

[0029] Furthermore, in this specification, selecting or determining information Z means obtaining information Z, obtaining a pointer to information Z, obtaining the ID of information Z, setting a flag on information Z, etc., and it is sufficient to be able to access information Z.

[0030] Figure 1 is an example of a conceptual diagram of information system A in this embodiment. Information system A comprises an inquiry processing device 1, one or more terminal devices 2, and one or more generation AI devices 3.

[0031] Inquiry Processing Device 1 is a device that processes inquiry information. Inquiry Processing Device 1 is usually a server. Inquiry Processing Device 1 may be, for example, a cloud server or an ASP server, but the type is not limited. Inquiry Processing Device 1 may also be a terminal. If Inquiry Processing Device 1 is a terminal, it may be, for example, a personal computer, a smartphone, or a tablet device, but the type is not limited.

[0032] Terminal device 2 is a device used by the user. Terminal device 2 can be, for example, a personal computer, smartphone, or tablet device, but the type is not limited. In this context, the user is the person who obtains the processing results of the inquiry information. The user could be, for example, a manager, a web page creator, or a member of the web page creation department.

[0033] The generation AI device 3 is a device that has the functionality of a generation AI. In this context, the generation AI device 3 typically has the functionality of a text generation AI. The generation AI may be, for example, ChatGPT or Google Bard, but is not limited to that. Note that Google is a registered trademark. The generation AI device 3 may be, for example, a cloud server or an ASP server, but is not limited to that type. The generation AI device 3 will be referred to as the generation AI as appropriate. In addition, the inquiry processing device 1 may also have the functionality of a generation AI. In such a case, the generation AI device 3 is not required for information system A.

[0034] The inquiry processing device 1 and one or more terminal devices 2, and the inquiry processing device 1 and one or more generating AI devices 3 can communicate with each other via a network such as the Internet.

[0035] Figure 2 is an example of a block diagram of information system A in this embodiment. Figure 3 is an example of a block diagram of inquiry processing device 1.

[0036] The inquiry processing device 1 comprises a storage unit 11, a reception unit 12, a processing unit 13, and an output unit 14. The storage unit 11 comprises a department management unit 111, an inquiry management unit 112, a prompt management unit 113, and a page storage unit 114. The reception unit 12 comprises an inquiry reception unit 121. The processing unit 13 comprises a judgment unit 130, an inquiry analysis unit 131, a department inquiry processing unit 132, an inquiry storage unit 133, a classification determination unit 134, a classification unit 135, a group determination unit 136, a representative determination unit 137, an improvement acquisition unit 138, and a statistical processing unit 139. The inquiry analysis unit 131 comprises a channel acquisition means 1311, a department determination means 1312, and an inquiry attribute value acquisition means 1313. The output unit 14 comprises a representative output unit 141, an improvement output unit 142, and a statistical output unit 143.

[0037] The terminal device 2 includes a terminal storage unit 21, a terminal receiving unit 22, a terminal processing unit 23, a terminal transmission unit 24, a terminal receiving unit 25, and a terminal output unit 26.

[0038] The storage unit 11, which constitutes the inquiry processing device 1, stores various types of information. These types of information include, for example, department information (described later), inquiry information (described later), prompts (described later), and one or more learning models.

[0039] The learning model could be, for example, a query information judgment model, a department determination model, a model for acquiring one or more query attribute values, or a classification model.

[0040] A query information judgment model is a learning model for determining whether query information satisfies processing conditions. For example, a learning unit (not shown) performs machine learning training using two or more training datasets, with query information or information based on query information as explanatory variables and the judgment result (whether or not the processing conditions are met) as the objective variable. This learning model is used in machine learning prediction processing. The learning model may also be called a learner, classifier, or classification model. The learning unit (not shown) may be located in the query processing device 1 or in a learning device (not shown). Information based on query information may be, for example, information vectorized from query information, information vectorized from a set of one or more independent words obtained from query information, or one or more independent words obtained from query information. Since information based on query information originates from query information, it may also be called query information.

[0041] A department determination model is a learning model used to determine the department corresponding to inquiry information. For example, a learning unit (not shown in the diagram) performs machine learning training using two or more training datasets, with inquiry information or information based on inquiry information as explanatory variables and department identifiers as the dependent variable. This training process constructs the information, which is then used in the machine learning prediction process.

[0042] A query attribute value acquisition model is a learning model for acquiring one or more query attribute values. For example, a learning unit (not shown) performs machine learning training using two or more training datasets, with query information or information based on query information as explanatory variables and query attribute values ​​as the target variable. This machine learning model is used for prediction processing. Examples of query attribute values ​​include sentiment identifier, text type, urgency, and response result. It is also preferable to have a query attribute value acquisition model for each of the two or more query attribute values.

[0043] A sentiment identifier is information that identifies the sentiment of the user making the query. Examples of sentiment identifiers include 'compliment', 'positive', 'negative', and 'neutral'.

[0044] Text type refers to the type of text in the inquiry information. Examples of text types include 'request', which indicates an inquiry form or request content; 'conversation', which indicates the content of a conversation between a customer and an operator; and 'notes', which indicates operator's notes or records.

[0045] Urgency refers to the degree of urgency required to respond to an inquiry. Examples of urgency levels include 'critical' (indicating an urgent response), 'high' (indicating a prompt response), 'medium' (indicating a normal response), and 'low' (indicating a low priority).

[0046] The response result is information indicating the outcome of the response to the inquiry. For example, the response result may be 'resolved' to indicate that the problem has been resolved, 'pending' to indicate that the problem has not yet been resolved, or 'escalated' to indicate that the problem has been escalated.

[0047] A classification model is a learning model that identifies reasons corresponding to query information. For example, a learning unit (not shown in the diagram) performs machine learning training using two or more training datasets, with query information or information based on query information as explanatory variables and classification identifiers as the target variable. This training process constructs the information and is used in machine learning prediction.

[0048] A reason is information used to classify inquiry information. It can also be described as a classification item or keyword for the reason inquiry information. Examples of reasons include "lost card," "login-related," and "address change."

[0049] The machine learning algorithms used in this specification include deep learning, random forests, decision trees, SVMs, and others. Furthermore, various machine learning functions and existing libraries can be used for machine learning, such as the TensorFlow® library, the R language's random forest module, fastText, and TinySVM.

[0050] The department management unit 111 stores information for one or more departments. Typically, the department management unit 111 stores information for two or more departments. It is preferable that the department management unit 111 stores information for one or more departments corresponding to two or more organizations. The department information in the department management unit 111 is associated with, for example, an organization identifier. The department information is also associated with, for example, a notification flag and contact information.

[0051] Department information refers to information about departments within an organization. For example, department information may be associated with a department identifier. It may also be a vector containing, for example, a string describing the department, one or more keywords indicating the department's characteristics, or the names of one or more products handled by the department.

[0052] A department identifier is information that identifies a department. Examples of department identifiers include the department name and the department ID.

[0053] An organization identifier is information that identifies an organization. Examples of organization identifiers include the organization name and the organization's ID. While an organization is typically a company, it can also be a group, a sole proprietorship, etc.

[0054] A notification flag indicates whether or not to notify a department or person in charge of inquiry information. For example, a notification flag could be "Do not notify (0)" or "Notify (1)". Alternatively, a notification flag could be "Do not notify (0)", "Notify in case of emergency (1)", or "Always notify (2)".

[0055] Contact information refers to the information that indicates who should receive inquiry information. Examples of contact information include email addresses, phone numbers, and user IDs for chat apps and social media.

[0056] The inquiry management unit 112 stores one or more inquiry information. The inquiry information in the inquiry management unit 112 is associated with, for example, a channel identifier. The inquiry information in the inquiry management unit 112 is associated with, for example, an organization identifier or department identifier. The inquiry information in the inquiry management unit 112 is associated with, for example, one or more reasons. The inquiry information in the inquiry management unit 112 is associated with, for example, a group identifier. The inquiry information in the inquiry management unit 112 is associated with, for example, an inquiry vector. The inquiry information in the inquiry management unit 112 is associated with, for example, time information. Time information is information that identifies the time when the inquiry information was received. Time information is, for example, a date and time, or a date, but the granularity is not specified.

[0057] Inquiry information refers to information about a user's inquiry to a target from one or more channels. The target is the subject of the user's inquiry. The target can be, for example, a product sold to the user or a service provided to the user, but is not limited to that. The product can be, for example, software or a device, but is not limited to that. Inquiry information is information that shows the user's questions, opinions, views, etc. Inquiry information can be, for example, a string of characters, but may also be an image, audio information, video, etc.

[0058] A channel identifier is information that identifies a channel. Examples of channel identifiers include the channel name and channel ID. A channel is the route through which user inquiry information arrives. A channel can also be described as information that identifies the source from which the inquiry information was entered. Examples of channels include telephone, chat, email, and operator memos.

[0059] The prompt management unit 113 stores one or more prompts. A prompt may be a prompt template for constructing a prompt. A prompt template has one or more variables.

[0060] The prompt management unit 113 stores, for example, inquiry judgment prompts, department determination prompts, inquiry analysis prompts, classification determination prompts, classification prompts, inquiry analysis prompts, and improvement suggestion prompts.

[0061] An inquiry judgment prompt is a prompt used to obtain a result of determining whether or not the inquiry information meets the processing conditions. An example of an inquiry judgment prompt might be: "Process the inquiry data, which contains a mix of various customer support content, determine whether it includes the main topic of the user's inquiry, and respond with 'yes' or 'no'."

[0062] Processing conditions are the conditions for performing some kind of processing on inquiry information. For example, processing conditions may be the conditions for storing inquiry information. For example, processing conditions may be the conditions for performing statistical processing using inquiry information. For example, processing conditions may be the conditions for determining the department identifier associated with the inquiry information. For example, processing conditions may be the conditions for determining the reason for the inquiry information. For example, processing conditions may be the conditions for determining the group to which the inquiry information belongs. For example, processing conditions may be the conditions used to obtain suggestions for improving a web page.

[0063] Failure to meet the processing conditions may result in, for example, the absence of user voice information in the inquiry (silent information), or the inclusion of only a greeting in the inquiry.

[0064] The department determination prompt is a prompt used to determine the department corresponding to the inquiry information. For example, the department determination prompt is: "Please select and output the department name for the following inquiry information from the following department name list. [Inquiry Information]<Inquiry Information> [Department Name List]<Department Information>". In the prompt, the strings enclosed in "<" and ">" are variables, and information will be assigned to them. <Inquiry Information> will contain the received inquiry information. <Department Information> will contain the department information of Department Management Department 111.

[0065] An inquiry analysis prompt is a prompt for analyzing inquiry information and obtaining one or more inquiry attribute values. An example of an inquiry analysis prompt is: "Your role is to analyze user statements from customer support inquiry records. Please analyze the following inquiry information according to the following criteria: - sentiment: Select one of the user's sentiments ('compliment', 'positive', 'negative', 'neutral'). [Inquiry Information]<Inquiry Information>"

[0066] A classification decision prompt is a prompt that uses two or more query information to output two or more reasons. For example, a classification decision prompt might be: "Output 10 reasons to classify the following query set, which consists of two or more query information: [Query Set]<Query Set>". A reason is a keyword used to classify query information.

[0067] A classification prompt is a prompt used to determine one or more reasons to which the inquiry information belongs from two or more reasons. An example of a classification prompt is: "Output the reasons that the following inquiry information belongs to from the following 10 reasons. Note that you may output two or more reasons. [Inquiry Information]<Inquiry Information> [Reason]<Reason>"

[0068] An inquiry analysis prompt is a prompt for analyzing inquiry information. For example, an inquiry analysis prompt might say, "Your role is to analyze user statements from customer support inquiry records. Please perform the analysis according to the following criteria:..."

[0069] An improvement suggestion prompt is a prompt designed to elicit improvement suggestions. An example of an improvement suggestion prompt might be: "Considering the following set of inquiries, please tell us what is missing or insufficient in the following web pages! [Set of Inquiries]<Set of Inquiries> [Web Page]<Web Page>"

[0070] The page storage unit 114 stores one or more web pages. Preferably, the one or more web pages are associated with an organization identifier. In other words, it is preferable that the page storage unit 114 stores a web page for each organization. The storage of web pages can be considered equivalent to the storage of URLs for accessing those web pages.

[0071] The reception desk 12 receives various types of information and instructions. These types of information and instructions include, for example, inquiry information.

[0072] Here, "reception" typically refers to the reception of information transmitted via wired or wireless communication lines, but it may also be a concept that includes the reception of information input from input devices such as keyboards, mice, and touch panels, as well as the reception of information read from recording media such as optical discs, magnetic discs, and semiconductor memory.

[0073] The inquiry reception unit 121 accepts one or more inquiry pieces. The inquiry reception unit 121 usually accepts two or more inquiry pieces. The inquiry reception unit 121 accepts inquiry pieces from one or more channels. The inquiry reception unit 121 usually accepts inquiry pieces from two or more channels. It is preferable that the inquiry pieces accepted by the inquiry reception unit 121 are associated with channel identifiers. It is preferable that the inquiry reception unit 121 also accepts response information associated with the inquiry pieces. It is preferable that the inquiry pieces accepted by the inquiry reception unit 121 are associated with organization identifiers.

[0074] The inquiry reception unit 121 receives inquiry information from, for example, the servers of one or more call centers. The inquiry reception unit 121 also receives inquiry information from, for example, the terminals of operators at one or more call centers. The inquiry reception unit 121 also receives inquiry information from, for example, the terminal device 2 of the user making the inquiry.

[0075] The processing unit 13 performs various processes. These various processes include, for example, those performed by the judgment unit 130, department determination means 1312, department inquiry processing unit 132, inquiry storage unit 133, classification determination unit 134, classification unit 135, group determination unit 136, representative determination unit 137, improvement acquisition unit 138, or statistical processing unit 139.

[0076] Furthermore, if the inquiry information received by the inquiry reception unit 121 is voice information, it is preferable for the processing unit 13 to perform voice recognition processing on the inquiry information to obtain the inquiry information as a string. Such string inquiry information can also be said to be inquiry information received by the inquiry reception unit 121. In addition, the processing unit 13 may process the inquiry information received by the inquiry reception unit 121 to obtain information based on the inquiry information. Such information based on the inquiry information may be, for example, an inquiry vector or one or more keywords. Again, such inquiry information can also be said to be inquiry information received by the inquiry reception unit 121.

[0077] The determination unit 130 determines whether the inquiry information received by the inquiry reception unit 121 satisfies the processing conditions and obtains the determination result. The determination result is, for example, "Yes (e.g., "1")" or "No (e.g., "0")".

[0078] The determination unit 130 obtains the determination result by, for example, one of the following methods (1) to (3). (1) Method using generative AI

[0079] The decision unit 130 acquires inquiry information received by the inquiry reception unit 121. The decision unit 130 acquires the inquiry decision prompt from the prompt management unit 113. The decision unit 130 provides the inquiry information and the inquiry decision prompt to the generating AI. Then, the decision unit 130 acquires the decision result from the generating AI.

[0080] Furthermore, providing information and prompts to the generating AI can be considered equivalent to substituting information into one or more variables in the prompt template, constructing a prompt, and then providing that prompt to the generating AI. (2) Methods using machine learning

[0081] The decision unit 130 acquires inquiry information received by the inquiry reception unit 121. The decision unit 130 acquires an inquiry information decision model from the storage unit 11. The decision unit 130 provides the inquiry information and the inquiry information decision model to a prediction module that performs machine learning prediction processing, executes the prediction module, and acquires the decision result. (3) Methods using natural language processing

[0082] The judgment unit 130 acquires inquiry information received by the inquiry reception unit 121. If the inquiry information is audio information, the judgment unit 130 performs speech recognition processing on the audio information to obtain the inquiry information as a string. Next, the inquiry reception unit 121 performs morphological analysis on the inquiry information as a string to obtain one or more independent words. Next, the inquiry reception unit 121 acquires the result of determining whether any of the one or more independent words match any of the one or more keywords in the storage unit 11.

[0083] The inquiry analysis unit 131 analyzes the inquiry information received by the inquiry reception unit 121 and obtains one or more inquiry attribute values. These inquiry attribute values ​​include, for example, a channel identifier that identifies the channel through which the inquiry information was received, a department identifier that identifies the department corresponding to the inquiry information, a sentiment identifier, text type, urgency, and response result.

[0084] The channel acquisition means 1311 acquires a channel identifier that identifies the channel through which the inquiry information was received. The channel identifier is information associated with the inquiry information received by the inquiry reception unit 121.

[0085] The department determination means 1312 obtains a department identifier for the inquiry information received by the inquiry reception unit 121. In other words, the department determination means 1312 determines the department that will handle the inquiry indicated by the inquiry information. The department determination means 1312 may also obtain an organization identifier associated with the inquiry information received by the inquiry reception unit 121.

[0086] It is preferable for the department determination means 1312 to acquire department identifiers only for inquiry information that the judgment unit 130 has determined to satisfy the processing conditions.

[0087] The department determination means 1312 obtains, for example, two or more department information items from the department management unit 111 that are paired with the received organization identifier. Using these two or more department information items, the department determination means 1312 obtains one or more department identifiers.

[0088] The department determination means 1312 obtains a department identifier by, for example, one of the following methods (1) to (3). (1) Methods using natural language processing (1-1) Method using keywords

[0089] The department determination means 1312 obtains one or more keywords from the inquiry information received by the inquiry reception department 121. For each of the two or more department information entries of the department management department 111, the department determination means 1312 obtains the number of keywords included in the department information from the one or more keywords obtained. The department determination means 1312 obtains one or more department identifiers that correspond to the department information that satisfies the corresponding conditions for the number of keywords. The corresponding conditions are that the number of keywords is equal to or greater than a threshold, or that the number of keywords is the maximum. (1-2) Vector method

[0090] The department determination means 1312 vectorizes the inquiry information received by the inquiry reception unit 121 and obtains an inquiry vector. For each of the two or more department information entries in the department management unit 111, the department determination means 1312 obtains the distance between the vector based on the department information and the inquiry vector, and obtains one or more department identifiers that correspond to the department information for which the distance satisfies the relevant conditions. The relevant conditions here are that the distance is less than or equal to a threshold, or the distance is the minimum. (2) Method using generative AI

[0091] The department determination means 1312 acquires inquiry information received by the inquiry reception unit 121. The department determination means 1312 acquires information on two or more departments from the department management unit 111. The department determination means 1312 acquires a department determination prompt from the prompt management unit 113. The department determination means 1312 provides the inquiry information, information on two or more departments, and the department determination prompt to the generating AI, and acquires one or more department identifiers from the generating AI. (3) Methods using machine learning

[0092] The department determination means 1312 acquires inquiry information received by the inquiry reception unit 121. The department determination means 1312 acquires a department determination model from the storage unit 11. The department determination means 1312 provides the inquiry information and the department determination model to a prediction module that performs machine learning prediction processing, executes the prediction module, and acquires a department identifier.

[0093] The inquiry attribute value acquisition means 1313 acquires one or more inquiry attribute values ​​using inquiry information. Inquiry attribute values ​​are attribute values ​​of inquiry information. For example, inquiry attribute values ​​may include emotion identifier, text type, urgency, and response result.

[0094] The inquiry attribute value acquisition means 1313 acquires one or more inquiry attribute values ​​by any of the following methods (1) to (2). (1) Method using generative AI

[0095] The inquiry attribute value acquisition means 1313 acquires inquiry information received by the inquiry reception unit 121. The inquiry attribute value acquisition means 1313 acquires the inquiry analysis prompt from the prompt management unit 113. The inquiry attribute value acquisition means 1313 provides the inquiry information and the inquiry analysis prompt to the generating AI and acquires one or more inquiry attribute values ​​from the generating AI. (2) Methods using machine learning

[0096] The inquiry attribute value acquisition means 1313 acquires inquiry information received by the inquiry reception unit 121. For each of the one or more inquiry attribute values, the inquiry attribute value acquisition means 1313 acquires an inquiry attribute value acquisition model corresponding to the inquiry attribute value to be acquired. For each of the one or more inquiry attribute values, the inquiry attribute value acquisition means 1313 provides the acquired inquiry information and the acquired inquiry attribute value acquisition model to a prediction module that performs machine learning prediction processing, executes the prediction module, and acquires the inquiry attribute value.

[0097] The department inquiry processing unit 132 processes the inquiry information using the department identifier obtained by the department determination means 1312. Such processing includes, for example, notification processing, storage processing, and statistical processing. (1) Notification processing performed by the department inquiry processing unit 132

[0098] The department inquiry processing unit 132 obtains contact information from the department management unit 111 that corresponds to the department identifier obtained by the department determination means 1312. The department inquiry processing unit 132 transmits the inquiry information received by the inquiry reception unit 121, or information based on the inquiry information, to the contact indicated by the contact information. The information based on the inquiry information is, in this case, for example, information that summarizes the inquiry information.

[0099] It is preferable for the department inquiry processing unit 132 to transmit the inquiry information received by the inquiry reception unit 121, or information based on such inquiry information, to the contact indicated by the contact information only if the notification conditions are met. The notification conditions are, for example, conditions based on a notification flag paired with a department identifier acquired by the department determination means 1312. The notification conditions are, for example, conditions based on one or more inquiry attribute values ​​(e.g., urgency). (2) Storage processing performed by the department inquiry processing unit 132

[0100] The department inquiry processing unit 132 stores the inquiry information received by the inquiry reception unit 121, or information based on said inquiry information, in the inquiry management unit 112, paired with the department identifier acquired by the department determination means 1312. The information based on the inquiry information is, in this case, for example, information summarizing the inquiry information. The department inquiry processing unit 132 may also store one or more inquiry attribute values ​​paired with the department identifier acquired by the department determination means 1312. (3) Statistical processing performed by the department inquiry processing unit 132

[0101] The department inquiry processing unit 132 performs statistical processing on one or more inquiry pieces paired with each of the two or more department identifiers. For example, the department inquiry processing unit 132 obtains the number or percentage of inquiry pieces for each of the two or more department identifiers. For example, the department inquiry processing unit 132 obtains the number or percentage of inquiry pieces for each of the two or more department identifiers and for each of the one or two or more reasons. The content of the statistical processing is not specified.

[0102] The inquiry storage unit 133 stores the inquiry information received by the inquiry reception unit 121 in the inquiry management unit 112, paired with a channel identifier that identifies the channel of the inquiry. The inquiry storage unit 133 may also store the inquiry information received by the inquiry reception unit 121 in the inquiry management unit 112, paired with a department identifier and one or more inquiry attribute values.

[0103] The inquiry storage unit 133, for example, associates each of the two or more inquiry pieces received by the inquiry reception unit 121 with one or more reasons determined by the classification unit 135, and stores them in the inquiry management unit 112.

[0104] The classification determination unit 134 uses some or all of the two or more inquiry information received by the inquiry reception unit 121 to obtain two or more reasons for the inquiry information.

[0105] The classification determination unit 134 obtains two or more reasons, for example, by the following method (1) or (2). (1) Method using generative AI

[0106] The classification determination unit 134 obtains a classification determination prompt from the prompt management unit 113. The classification determination unit 134 obtains two or more inquiry pieces from the inquiry management unit 112, or information based on each of the two or more inquiry pieces. The classification determination unit 134 provides the two or more inquiry pieces, or information based on each of the two or more inquiry pieces, and the classification determination prompt to the generating AI, and obtains two or more reasons from the generating AI. (2) Methods using natural language processing

[0107] The classification determination unit 134 acquires two or more inquiry pieces from the inquiry management unit 112, or information based on each of the two or more inquiry pieces. Next, the classification determination unit 134 acquires one or more characteristic words from each of the two or more inquiry pieces, or information based on each of the two or more inquiry pieces. The classification determination unit 134 aggregates the acquired two or more characteristic words to obtain two or more characteristic words.

[0108] Characteristic terms are independent words that are distinctive terms. For example, distinctive terms are those whose tf / idf values ​​are above or greater than a certain threshold.

[0109] Aggregating characteristic terms involves uniquely processing two or more characteristic terms, or consolidating synonyms or related terms among two or more characteristic terms into a single term.

[0110] The classification unit 135 determines one or more reasons to which each of the two or more inquiry pieces received by the inquiry reception unit 121 belongs. The one or more reasons are one or more of the two or more reasons obtained by the classification determination unit 134.

[0111] The classification unit 135 obtains two or more reasons by, for example, one of the following methods (1) to (3). (1) Method using generative AI

[0112] The classification unit 135 acquires inquiry information to be classified. The classification unit 135 acquires two or more reasons. The classification unit 135 acquires a classification prompt from the prompt management unit 113. The classification unit 135 provides the inquiry information, the two or more reasons, and the classification prompt to the generating AI, and acquires one or more classification identifiers from the generating AI. (2) Methods using machine learning

[0113] The classification unit 135 acquires query information to be classified. The classification unit 135 acquires a classification model from the storage unit 11. The classification unit 135 provides the query information and the classification model to a prediction module that performs machine learning prediction processing, executes the prediction module, and acquires a classification identifier. (3) Methods using natural language processing

[0114] The classification unit 135 acquires query information to be classified. The classification unit 135 vectorizes the query information. The classification unit 135 calculates the similarity between the vector and the vector paired with each of the two or more reasons. The classification unit 135 acquires the reasons paired with the vector corresponding to each of the one or more similarities that satisfy the selection criteria. The selection criteria include, for example, the highest similarity, or the similarity being equal to or greater than a threshold. The vector paired with a reason is, for example, a vector of the reason which is a word (for example, a vector obtained by Word2vec), or a vector obtained from a string that describes the reason.

[0115] The group determination unit 136 groups two or more inquiry information received by the inquiry reception unit 121 and determines two or more groups.

[0116] The group determination unit 136, for example, vectorizes the query information for each of the two or more query information from the query management unit 112 and obtains a query vector. Next, the group determination unit 136 groups the two or more query vectors, for example, using the k-means method. The group determination unit 136 associates a group identifier with each of the two or more query information. The group determination unit 136 may also group the two or more query vectors using a clustering method other than the k-means method. Examples of clustering methods other than the k-means method include DBSCAN (Density-Based Spatial Clustering of Applications with Noise), Gaussian Mixture Model (GMM), Self-Organizing Map (SOM), and Spectral Clustering. The group identifier is information that identifies a group, for example, the group's ID. Furthermore, the group determination unit 136 may determine that each inquiry information other than the representative inquiry information belongs to the group to which the closest (most similar) representative inquiry information belongs, from among the two or more representative inquiry information that have been determined.

[0117] The representative determination unit 137 determines representative inquiry information for each of the one or more groups, which is the inquiry information that represents the group. The representative determination unit 137 may determine one or more representative inquiry information before the groups are determined.

[0118] The representative determination unit 137 calculates, for example, for each of two or more groups, the sum of the distances between each query vector of two or more query information belonging to the group and one or more other query vectors in the same group. The representative determination unit 137 determines the query information corresponding to the query vector with the smallest sum of distances as the representative query information for the group.

[0119] The improvement acquisition unit 138 refers to the page storage unit 114, which stores one or more web pages related to the target, and uses one or more inquiry pieces of information received by the inquiry reception unit 121 to acquire improvement suggestions for one or more web pages.

[0120] The improvement acquisition unit 138, for example, refers to the page storage unit 114, which stores one or more web pages related to the target, and uses the inquiry information received by the inquiry reception unit 121 to acquire improvement suggestions for web pages that contain missing information from one or more web pages.

[0121] The improvement acquisition unit 138 is preferably able to acquire improvement suggestions using the inquiry information and response information received by the inquiry reception unit 121.

[0122] The improvement acquisition unit 138, for example, refers to the page storage unit 114 to acquire one or more web pages that are paired with the acquired organization identifier. The improvement acquisition unit 138, for example, uses inquiry information received by the inquiry reception unit 121 to acquire improvement suggestions for web pages that are missing information in the said one or more web pages.

[0123] The improvement acquisition unit 138 may, for example, acquire improvement proposals using two or more representative inquiry pieces determined by the representative determination unit 137.

[0124] The improvement acquisition unit 138 acquires improvement suggestions by, for example, one of the following methods (1) to (2). (1) Method using generative AI

[0125] The improvement acquisition unit 138 acquires two or more inquiry pieces from the inquiry management unit 112. The improvement acquisition unit 138 acquires one or more web pages from the page storage unit 114. The improvement acquisition unit 138 acquires improvement suggestion prompts from the prompt management unit 113. Next, the improvement acquisition unit 138 provides the inquiry set, which is a collection of two or more inquiry pieces, one or more web pages, and the improvement suggestion prompts to the generating AI, and acquires one or more improvement suggestions from the generating AI. (2) Methods using natural language processing

[0126] The improvement acquisition unit 138 acquires two or more inquiry pieces from the inquiry management unit 112. The improvement acquisition unit 138 vectorizes each of the two or more inquiry pieces and acquires an inquiry vector. For each of the one or more processing units that each of the one or more web pages has, the improvement acquisition unit 138 vectorizes the processing unit and acquires a processing unit vector. For each of the two or more inquiry vectors, the improvement acquisition unit 138 calculates the similarity between the inquiry vector and each of the one or more processing unit vectors. If the similarity of one or more for each inquiry vector does not meet the acceptance criteria, the improvement acquisition unit 138 assumes that the inquiry information paired with that inquiry vector is information not contained on the web page and acquires an improvement suggestion that encourages adding the answer to that inquiry information to the web page. The acceptance criteria are, for example, that the similarity is above or above a threshold or greater than a threshold. The processing unit can be, for example, a page or a frame, but is not limited to that.

[0127] The statistical processing unit 139 performs statistical processing on two or more inquiry pieces and obtains the statistical processing results.

[0128] The statistical processing unit 139 obtains, for example, statistical processing results for query information for each of two or more channel identifiers. The statistical processing unit 139 obtains, for example, the number or percentage of query information for each of two or more channel identifiers. The statistical processing unit 139 obtains, for example, the number or percentage of query information for each of two or more channel identifiers and for each reason. The content of the statistical processing is not specified.

[0129] The output unit 14 outputs various types of information. These types of information include, for example, representative inquiry information, improvement suggestions, and statistical processing results.

[0130] The output here refers to, for example, transmission to terminal device 2, but may also be a concept that includes display on a screen, projection using a projector, printing with a printer, sound output, transmission to external devices other than terminal device 2, storage on a recording medium, and delivery of processing results to other processing devices or other programs.

[0131] The representative output unit 141 outputs one or more representative inquiry pieces determined by the representative determination unit 137.

[0132] The improvement output unit 142 outputs one or more improvement suggestions acquired by the improvement acquisition unit 138.

[0133] The statistical output unit 143 outputs one or more statistical processing results obtained by the statistical processing unit 139.

[0134] Various types of information are stored in the terminal storage unit 21 of the terminal device 2. These types of information include, for example, an organization identifier and a user identifier. The organization identifier may also serve as the user identifier. In other words, if the user identifier is determined, the organization identifier may also be determined.

[0135] The terminal reception unit 22 receives input such as instructions and information from the user. These instructions and information include, for example, a suggestion output instruction. A suggestion output instruction is an instruction to output improvement suggestions.

[0136] The means of inputting instructions and information can be anything, such as a touch panel, keyboard, mouse, or menu screen.

[0137] The terminal processing unit 23 performs various processes. These processes include, for example, changing the instructions and information received by the terminal receiving unit 22 into instructions and information for transmission, and changing the information received by the terminal receiving unit 25 into an output structure.

[0138] The terminal transmission unit 24 transmits various information and instructions to the inquiry processing device 1. These various information and instructions include, for example, an organization identifier, a user identifier, and a suggestion output instruction.

[0139] The terminal receiving unit 25 receives various types of information from the inquiry processing device 1. These types of information include, for example, inquiry information, representative inquiry information, and improvement suggestions.

[0140] The terminal output unit 26 outputs various types of information. These types of information include, for example, inquiry information, representative inquiry information, and improvement suggestions.

[0141] Here, output refers to, for example, display on a screen, but may also be a concept that includes projection using a projector, printing with a printer, sound output, storage on a recording medium, transmission to other processing devices, and delivery of processing results to other processing devices or other programs.

[0142] The storage unit 11, department management unit 111, inquiry management unit 112, prompt management unit 113, page storage unit 114, and terminal storage unit 21 are preferably made of non-volatile recording media, but can also be made of volatile recording media.

[0143] The process by which information is stored in the storage unit 11, etc. is not relevant. For example, information may be stored in the storage unit 11, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 11, etc., or information input via an input device may be stored in the storage unit 11, etc.

[0144] The reception unit 12 and the inquiry reception unit 121 are preferably implemented by wireless or wired communication means, but may also be implemented by means of receiving broadcasts, device drivers for input means such as touch panels and keyboards, or control software for menu screens.

[0145] The processing unit 13, judgment unit 130, department determination means 1312, department inquiry processing unit 132, inquiry storage unit 133, classification determination unit 134, classification unit 135, group determination unit 136, representative determination unit 137, improvement acquisition unit 138, and statistical processing unit 139 can typically be implemented using a processor, memory, etc. The processing procedures of the processing unit 13, etc., are typically implemented in software, and this software is recorded on a recording medium such as ROM. However, it may also be implemented in hardware (dedicated circuitry). The processor can be a CPU, MPU, GPU, etc., and the type is not limited.

[0146] The output unit 14, representative output unit 141, improved output unit 142, and statistical output unit 143 are preferably implemented by wireless or wired communication means, but may also be implemented by driver software for an output device such as a display or speaker, or by driver software for an output device and the output device itself.

[0147] The terminal reception unit 22 can be implemented using device drivers for input means such as touch panels and keyboards, or control software for menu screens, etc.

[0148] The terminal transmission unit 24 is usually implemented by wireless or wired communication means, but it may also be implemented by broadcasting means.

[0149] The terminal receiving unit 25 is usually implemented by wireless or wired communication means, but it may also be implemented by means of receiving broadcasts.

[0150] The terminal output unit 26 may or may not be considered to include output devices such as a display or speakers. The terminal output unit 26 can be implemented using driver software for an output device, or a driver software for an output device and an output device.

[0151] Next, an example of the operation of the inquiry processing device 1 will be explained using the flowchart in Figure 4.

[0152] (Step S401) The inquiry reception unit 121 determines whether or not it has received the inquiry information. If it has received the inquiry information, it proceeds to step S402; if it has not received the inquiry information, it proceeds to step S411.

[0153] (Step S402) The processing unit 13 determines whether the inquiry information received in step S401 is voice information or not. If it is voice information, proceed to step S403; if it is a string, proceed to step S404.

[0154] (Step S403) The processing unit 13 performs speech recognition processing on the inquiry information received in step S401 and obtains the inquiry information in the form of a string. Note that the inquiry information in the form of a string can also be simply called inquiry information.

[0155] (Step S404) The decision unit 130 performs a decision process using the query information in the string. An example of the decision process will be explained using the flowchart in Figure 5.

[0156] (Step S405) If the result of the judgment in step S404 is that "the processing conditions are met", proceed to step S406; if the result is that "the processing conditions are not met", return to step S401.

[0157] (Step S406) The inquiry analysis unit 131 analyzes the inquiry information and obtains two or more inquiry attribute values. An example of such inquiry analysis processing will be explained using the flowchart in Figure 6.

[0158] (Step S407) The inquiry storage unit 133 obtains two or more inquiry attribute values ​​obtained in step S406.

[0159] (Step S408) The inquiry storage unit 133 associates two or more inquiry attribute values ​​with inquiry information and stores them in the inquiry management unit 112. It is also preferable for the inquiry storage unit 133 to store inquiry vectors obtained from inquiry information in the inquiry management unit 112, associating them with the inquiry information.

[0160] (Step S409) The department inquiry processing unit 132 determines whether the inquiry information received in step S401 matches the department notification conditions. If it matches the department notification conditions, the process proceeds to step S410; otherwise, the process returns to step S401.

[0161] Department notification conditions are the conditions for notifying a department of inquiry information. Department notification conditions may be the same as notification conditions.

[0162] (Step S410) The department inquiry processing unit 132 obtains a department identifier that corresponds to the inquiry information. The department inquiry processing unit 132 obtains contact information that corresponds to the department identifier from the department management unit 111. The department inquiry processing unit 132 notifies the contact person indicated by the contact information of the inquiry information received in step S401. The process returns to step S401. Note that the inquiry information may consist of the inquiry information alone, or it may consist of the inquiry information and one or more inquiry attribute values.

[0163] (Step S411) The processing unit 13 determines whether or not it is time to analyze the inquiry information. If it is time to analyze the inquiry information, it proceeds to step S412; otherwise, it proceeds to step S421.

[0164] The timing for analyzing inquiry information may include, for example, when a predetermined time has passed, when a predetermined number of inquiry entries have been accumulated, or when instructions from a user have been received.

[0165] (Step S412) The processing unit 13 assigns 1 to counter i.

[0166] (Step S413) The processing unit 13 determines whether or not the i-th organization exists. If the i-th organization exists, the process proceeds to step S414; otherwise, the process returns to step S401.

[0167] Note that if an i-th organization exists, for example, if the i-th organization identifier exists in the Department Management Division 111.

[0168] (Step S414) The processing unit 13 obtains the organizational identifier of the i-th organization from the department management unit 111.

[0169] (Step S415) The processing unit 13 analyzes the query information of the i-th organization. An example of such organization analysis processing will be explained using the flowchart in Figure 8.

[0170] (Step S416) The improvement acquisition unit 138 acquires improvement suggestions for the i-th organization's web page. An example of such improvement suggestion acquisition process will be explained using the flowchart in Figure 14.

[0171] (Step S417) The determination unit 130 determines whether or not the organizational notification conditions are met. If the organizational notification conditions are met, the process proceeds to step S418; otherwise, the process proceeds to step S420.

[0172] Organizational notification conditions are the conditions for notifying an organization of the results of the organizational analysis process or improvement suggestions. Organizational notification conditions are, for example, based on the content of the results of the organizational analysis process or improvement suggestions obtained. Organizational notification conditions may include, for example, the number of inquiries from an organization exceeding a threshold, or the existence of improvement suggestions.

[0173] (Step S418) The department inquiry processing unit 132 obtains information to be notified. The information to be notified is, for example, representative inquiry information and improvement suggestions.

[0174] (Step S419) The department inquiry processing unit 132 obtains one or more contact information from the department management unit 111 that corresponds to the organization identifier of the i-th organization. The department inquiry processing unit 132 notifies the contacts indicated by the one or more contact information of the information obtained in step S418.

[0175] (Step S420) The processing unit 13 increments counter i by 1. The process returns to step S413.

[0176] (Step S421) The reception unit 12 determines whether or not it has received an information output instruction. If it has received an information output instruction, it proceeds to step S422; if it has not received an information output instruction, it returns to step S401. At this point, the reception unit 12 determines whether or not it has received an information output instruction from the terminal device 2.

[0177] (Step S422) The processing unit 13 acquires one or more pieces of information specified by the information output instruction.

[0178] (Step S423) The output unit 14 outputs one or more pieces of information acquired in step S422. Here, the output unit 14 transmits one or more pieces of information to the terminal device 2 that sent the information output instruction.

[0179] In the flowchart shown in Figure 4, processing is terminated by power-off or processing termination interrupts.

[0180] Next, an example of the decision-making process in step S404 will be explained using the flowchart in Figure 5.

[0181] (Step S501) The decision unit 130 acquires the inquiry information received by the inquiry reception unit 121.

[0182] (Step S502) The decision unit 130 obtains a query decision prompt from the prompt management unit 113.

[0183] (Step S503) The decision unit 130 provides the inquiry information obtained in step S501 and the inquiry decision prompt obtained in step S502 to the generating AI.

[0184] (Step S504) The determination unit 130 determines whether or not an answer has been obtained from the generating AI. If an answer has been obtained, the process proceeds to step S505; otherwise, it returns to step S504.

[0185] (Step S505) The decision unit 130 obtains a decision result from the answer obtained in step S504. It returns to the higher-level processing.

[0186] Next, an example of the inquiry analysis process in step S406 will be explained using the flowchart in Figure 6.

[0187] (Step S601) The inquiry analysis unit 131 acquires inquiry information.

[0188] (Step S602) The channel acquisition means 1311 acquires a channel identifier that identifies the channel on which the inquiry information was received.

[0189] (Step S603) The department determination means 1312 determines the department that should respond to the inquiry information. An example of such department determination process will be explained using the flowchart in Figure 7.

[0190] (Step S604) The inquiry attribute value acquisition means 1313 acquires an inquiry analysis prompt from the prompt management unit 113.

[0191] (Step S605) The inquiry attribute value acquisition means 1313 provides the inquiry information and the inquiry analysis prompt to the generating AI.

[0192] (Step S606) The inquiry attribute value acquisition means 1313 determines whether or not it has obtained a response from the generating AI. If a response has been obtained, it proceeds to step S607; otherwise, it returns to step S606.

[0193] (Step S607) The inquiry attribute value acquisition means 1313 acquires one or more inquiry attribute values ​​from the answer obtained in step S606. It returns to the higher-level processing.

[0194] Next, an example of the department determination process in step S603 will be explained using the flowchart in Figure 7.

[0195] (Step S701) The department determination means 1312 performs morphological analysis on the acquired inquiry information.

[0196] (Step S702) The department determination means 1312 obtains one or more keywords from the results of morphological analysis. Keywords are usually independent words. Keywords are, for example, terms of a specific part of speech such as nouns.

[0197] (Step S703) The department determination means 1312 assigns 1 to counter i.

[0198] (Step S704) The department determination means 1312 determines whether the i-th department information exists in the department management unit 111. If the i-th department information exists, the process proceeds to step S705; otherwise, it returns to the higher-level process.

[0199] (Step S705) The department determination means 1312 obtains the i-th department information from the department management unit 111.

[0200] (Step S706) The department determination means 1312 uses one or more keywords and department information to determine whether the inquiry information corresponding to the one or more keywords satisfies the conditions for the i-th department information. If the conditions are met, proceed to step S707; otherwise, proceed to step S708.

[0201] The applicable conditions include, for example, that the department information contains one or more keywords, that the number of keywords contained in the department information is the maximum possible, that the distance between the vectorized department information and the vector obtained from one or more keywords is within a threshold, or that the distance between the vectorized department information and the vector obtained from one or more keywords is the minimum possible.

[0202] (Step S707) The department determination means 1312 obtains a department identifier that is paired with the i-th department information, associates it with the inquiry information, and stores the department identifier.

[0203] (Step S708) The department determination means 1312 increments the counter i by 1. Return to step S704.

[0204] Next, an example of the tissue analysis process in step S415 will be explained using the flowchart in Figure 8.

[0205] (Step S801) The processing unit 13 retrieves all inquiry information to be analyzed from the inquiry information of the relevant organization. The inquiry information retrieved may be all inquiry information that is paired with the organization identifier of the relevant organization, or it may be inquiry information for a specific period (for example, the last month, or this fiscal year) from the inquiry information that is paired with the organization identifier of the relevant organization.

[0206] (Step S802) The classification determination unit 134 determines whether there are two or more reasons that correspond to a department identifier in the department management unit 111. If there are two or more reasons, the process proceeds to step S803; otherwise, the process proceeds to step S804.

[0207] (Step S803) The classification determination unit 134 determines whether or not to create a reason for the organization identified by the organization identifier. If a reason is to be created, the unit proceeds to step S804; otherwise, it proceeds to step S805.

[0208] Furthermore, a reason will be created in the Department Management Department 111 even if two or more reasons already exist, for example, if a period of time exceeding a threshold has elapsed since the creation of two or more reasons, if the number of inquiry pieces has increased by more than a threshold since the creation of two or more reasons, or if a flag indicating that a reason should always be created is managed in conjunction with the organizational identifier during organizational analysis.

[0209] (Step S804) The classification determination unit 134 creates a reason for the organization identified by the organization identifier. An example of such reason creation process will be explained using the flowchart in Figure 9.

[0210] (Step S805) The classification unit 135 obtains two or more reasons from the department management unit 111 that correspond to the department identifier.

[0211] (Step S806) The classification unit 135 classifies the query information to be analyzed using two or more reasons. An example of such classification processing will be explained using the flowchart in Figure 10.

[0212] (Step S807) The group determination unit 136 groups the inquiry information to be analyzed. An example of this grouping process will be explained using the flowchart in Figure 11.

[0213] (Step S808) The representative determination unit 137 determines representative inquiry information for one or more groups. An example of such representative determination processing will be explained using the flowchart in Figure 12.

[0214] (Step S809) The statistical processing unit 139 performs statistical processing on the organization's inquiry information. It then returns to the higher-level processing unit. An example of such statistical processing will be explained using the flowchart in Figure 13.

[0215] Next, an example of the reason creation process in step S804 will be explained using the flowchart in Figure 9.

[0216] (Step S901) The classification determination unit 134 obtains from the inquiry management unit 112 two or more inquiry pieces that are paired with the organization identifier of interest and serve as the basis for creating a reason. The two or more inquiry pieces that serve as the basis for creating a reason may be all the inquiry pieces that are paired with the organization identifier, or they may be some of the inquiry pieces. In the case of some inquiry pieces, for example, the classification determination unit 134 obtains inquiry pieces from recent inquiry pieces within a threshold period, or inquiry pieces from recent inquiry pieces within a threshold number.

[0217] (Step S902) The classification determination unit 134 obtains a classification determination prompt from the prompt management unit 113.

[0218] (Step S903) The classification decision unit 134 passes the two or more inquiry pieces obtained in step S901 and the classification decision prompt obtained in step S902 to the generation AI.

[0219] (Step S904) The classification determination unit 134 determines whether or not it has obtained an answer from the generating AI. If an answer has been obtained, it proceeds to step S905; otherwise, it returns to step S904.

[0220] (Step S905) The classification determination unit 134 obtains two or more reasons from the answers obtained in step S904.

[0221] (Step S906) The classification determination unit 134 stores the two or more reasons obtained in step S905 in the department management unit 111, associating them with the organization identifier. It then returns to the higher-level processing.

[0222] Next, an example of the classification process in step S806 will be explained using the flowchart in Figure 10.

[0223] (Step S1001) The classification unit 135 obtains two or more reasons.

[0224] (Step S1002) The classification unit 135 obtains a classification prompt from the prompt management unit 113.

[0225] (Step S1003) The classification unit 135 assigns 1 to counter i.

[0226] (Step S1004) The classification unit 135 determines whether or not the i-th query information to be classified exists. If the i-th query information exists, the process proceeds to step S1005; otherwise, it returns to the higher-level processing.

[0227] (Step S1005) The classification unit 135 obtains the i-th query information.

[0228] (Step S1006) The classification unit 135 passes the i-th query information, two or more reasons obtained in step S1001, and the classification prompt obtained in step S1002 to the generation AI.

[0229] (Step S1007) The classification unit 135 determines whether or not it has obtained an answer from the generating AI. If an answer has been obtained, it proceeds to step S1008; otherwise, it returns to step S1007.

[0230] (Step S1008) The classification unit 135 obtains the reason from the answer obtained in step S1007.

[0231] (Step S1009) The classification unit 135 associates the reason obtained in step S1008 with the i-th query information.

[0232] (Step S1010) The classification unit 135 increments the counter i by 1.

[0233] Next, an example of the grouping process in step S807 will be explained using the flowchart in Figure 11.

[0234] (Step S1101) The group determination unit 136 obtains two or more inquiry pieces to be grouped.

[0235] (Step S1102) The group determination unit 136 assigns 1 to counter i.

[0236] (Step S1103) The group determination unit 136 determines whether or not the i-th query information exists. If the i-th query information exists, the unit proceeds to step S1104; otherwise, the unit proceeds to step S1106.

[0237] (Step S1104) The group determination unit 136 vectorizes the i-th query information, obtains a query vector, and associates the query vector with the i-th query information.

[0238] (Step S1105) The group determination unit 136 increments counter i by 1. Return to step S1103.

[0239] (Step S1106) The group determination unit 136 groups two or more query vectors using a clustering method such as the k-means method.

[0240] (Step S1107) The group determination unit 136 associates a group identifier that identifies a group with each of the two or more query vectors. It then returns to the higher-level processing.

[0241] Next, an example of the representative determination process in step S808 will be explained using the flowchart in Figure 12.

[0242] (Step S1201) The representative determination unit 137 assigns 1 to counter i.

[0243] (Step S1202) The representative determination unit 137 determines whether or not the i-th group exists. If the i-th group exists, the process proceeds to step S1203; otherwise, it returns to the higher-level process.

[0244] (Step S1203) The representative determination unit 137 assigns 1 to counter j.

[0245] (Step S1204) The representative determination unit 137 determines whether or not the j-th query information is included in the i-th group. If the j-th query information is included, proceed to step S1205; otherwise, proceed to step S1214.

[0246] (Step S1205) The representative determination unit 137 obtains a query vector that is paired with the j-th query information included in the i-th group.

[0247] (Step S1206) The representative determination unit 137 assigns 1 to counter k.

[0248] (Step S1207) The representative determination unit 137 determines whether there is a k-th other query information in the i-th group, excluding the j-th query information. If there is a k-th other query information, the unit proceeds to step S1208; otherwise, it proceeds to step S1212.

[0249] (Step S1208) The representative determination unit 137 obtains a query vector that pairs with the kth other query information, excluding the jth query information in the i-th group.

[0250] (Step S1209) The representative determination unit 137 calculates the distance between the two query vectors. The two query vectors are the vectors obtained in steps S1205 and S1208.

[0251] (Step S1210) The representative determination unit 137 adds the distance to the total distance corresponding to the jth query information included in the i-th group. The initial value of the total distance is 0.

[0252] (Step S1211) The representative determination unit 137 increments counter k by 1. Return to step S1207.

[0253] (Step S1212) The representative determination unit 137 associates the total distance with the j-th query information included in the i-th group.

[0254] (Step S1213) The representative determination unit 137 increments counter j by 1. Return to step S1204.

[0255] (Step S1214) The representative determination unit 137 selects the query information with the minimum total distance for the i-th group as the representative query information. The representative determination unit 137 stores the representative query information in association with the group identifier of the i-th group. The storage location is, for example, the query management unit 112, but is not limited to that location.

[0256] (Step S1215) The representative determination unit 137 increments counter i by 1. Return to step S1202.

[0257] Next, an example of the statistical processing in step S809 will be explained using the flowchart in Figure 13.

[0258] (Step S1301) The statistical processing unit 139 assigns 1 to counter i.

[0259] (Step S1302) The statistical processing unit 139 determines whether or not the i-th channel exists. If the i-th channel exists, the process goes to step S1303; otherwise, the process goes to step S1305.

[0260] (Step S1303) The statistical processing unit 139 obtains the number of query information paired with the channel identifier of the i-th channel and stores the number paired with the channel identifier of the i-th channel.

[0261] (Step S1304) The statistical processing unit 139 increments counter i by 1. Return to step S1302.

[0262] (Step S1305) The statistical processing unit 139 assigns 1 to counter i.

[0263] (Step S1306) The statistical processing unit 139 determines whether or not the i-th organization exists. If the i-th organization exists, the process proceeds to step S1307; otherwise, it returns to the higher-level processing.

[0264] (Step S1307) The statistical processing unit 139 obtains the number of query information pairs with the organization identifier of the i-th organization, and stores the number paired with the organization identifier of the i-th organization.

[0265] (Step S1308) The statistical processing unit 139 assigns 1 to counter j.

[0266] (Step S1309) The statistical processing unit 139 determines whether or not the j-th reason exists. If the j-th reason exists, proceed to step S1310; otherwise, proceed to step S1312.

[0267] (Step S1310) The statistical processing unit 139 obtains the number of query information pairs with the organization identifier of the i-th organization and the j-th reason, and stores this number paired with the organization identifier of the i-th organization and the j-th reason.

[0268] (Step S1311) The statistical processing unit 139 increments counter j by 1. Return to step S1309.

[0269] (Step S1312) The statistical processing unit 139 increments counter i by 1. Return to step S1307.

[0270] In the flowchart of Figure 13, the statistical processing unit 139 may obtain the number of inquiry information for each department. Alternatively, the statistical processing unit 139 may obtain the number of inquiry information for each department and for each reason.

[0271] Next, an example of the good suggestion acquisition process in step S416 will be explained using the flowchart in Figure 14.

[0272] (Step S1401) The improvement acquisition unit 138 acquires two or more inquiry messages that are paired with the organization identifier of the target organization from the inquiry management unit 112.

[0273] (Step S1402) The improvement acquisition unit 138 refers to the page storage unit 114 and acquires one or more web pages that are paired with the organization identifier of the target organization.

[0274] (Step S1403) The improvement acquisition unit 138 acquires an improvement proposal prompt from the prompt management unit 113.

[0275] (Step S1404) The improvement acquisition unit 138 provides the two or more inquiry messages acquired in Step S1401, the one or more web pages acquired in Step S1402, and the improvement proposal prompt acquired in Step S1403 to the generation AI.

[0276] (Step S1405) The improvement acquisition unit 138 determines whether an answer has been acquired from the generation AI. If an answer has been acquired, it proceeds to Step S1406; if no answer has been acquired, it returns to Step S1405.

[0277] (Step S1406) The improvement acquisition unit 138 acquires one or more improvement proposals from the answer acquired in Step S1405, and accumulates the one or more improvement proposals in pairs with the organization identifier of the target organization. It returns to the upper-level process.

[0278] Hereinafter, a specific operation example of the information system A in the present embodiment will be described.

[0279] The department management unit 111 of the inquiry processing device 1 stores the department management table shown in Figure 15. The department management table is a table that manages department information for each organization. The department management table manages one or more records that have "ID", "organization identifier", "department identifier", "notification flag", "contact information", and "department information". "Department information" has "keyword" and "department description". "ID" is information that identifies the record. "Notification flag" is information that indicates whether or not to immediately notify the department of the inquiry information. Notification flag "0" indicates that the department will not immediately notify the department of the inquiry information. Notification flag "1" indicates that the department will immediately notify the department of the inquiry information only if the inquiry attribute value (urgency) of the inquiry information indicates "urgent". Notification flag "2" indicates that the department will always immediately notify the department of the inquiry information. "Contact information" is information that indicates the contact information of the department. "Contact information" is information that indicates where the inquiry information will be notified. "Keyword" is a term that indicates the characteristics of the department, for example, the name of the product handled by the department. "Department description" is a description of the department.

[0280] The inquiry management unit 112 stores an inquiry management table having the structure shown in Figure 16. The inquiry management table is a table that manages inquiry information for each organization. The inquiry management table manages one or more records that have "ID", "Inquiry Information", "Organization Identifier", "Department Identifier", "Channel Identifier", "Reason", "Group Identifier", "Inquiry Vector", and "Date and Time". "Inquiry Information" is a string here. Inquiries received by telephone are processed by speech recognition and the string corresponding to the inquiry is stored. "Organization Identifier" is information that identifies the organization that is the subject of the inquiry. "Department Identifier" is information that identifies the department that is the subject of the inquiry. "Channel Identifier" is information that identifies the channel from which the inquiry was made. "Reason" is one or more reasons to which the inquiry information belongs. "Group Identifier" is the identifier of the group to which the inquiry information belongs. "Inquiry Vector" is information that vectorizes the inquiry information. "Date and Time" is the date and time the inquiry was made. "Date and Time" may be entered by the operator and does not need to be accurate.

[0281] The prompt management unit 113 stores the prompt management tables shown in Figures 17 and 18. A prompt management table is a table that manages prompts. The prompt management table manages one or more records that have a "prompt identifier" and a "prompt".

[0282] The page storage unit 114 stores one or more web pages for each organization, or URLs for each organization to access websites.

[0283] In light of the above situation, the following three specific examples will be explained. Specific example 1 is the processing of a single inquiry. Specific example 2 is the processing of a set of numerous inquiry pieces. The processing of the inquiry set includes reason determination, classification determination, organizational analysis, and improvement suggestion processing. Specific example 3 is the improvement suggestion processing using inquiry information and web pages.

[0284] (Specific example 1) The inquiry receiving unit 121 of the inquiry processing device 1 receives inquiry information from numerous call center servers and chat servers (not shown). The received inquiry information is associated with an organization identifier (e.g., company name) and a channel identifier. The following describes the processing of one inquiry (e.g., "I cannot change my address because I cannot log in...").

[0285] The processing unit 13 determines whether each received inquiry is audio information. If the inquiry is audio information, the processing unit 13 performs speech recognition processing on the inquiry to obtain the inquiry information as a string.

[0286] Next, the decision unit 130 performs the decision processing described using the flowchart in Figure 5 for the string query information. That is, the decision unit 130 obtains the query decision prompt from the prompt management table (Figure 17). Next, the decision unit 130 provides the generation AI with the query information and the obtained query decision prompt. Then, the decision unit 130 obtains a response from the generation AI for each query. Next, the decision unit 130 obtains a decision result ("yes" or "no") from the obtained response for each query. Here, the decision unit 130 obtains a decision result of "yes". Note that the following processing is performed only for query information that corresponds to a decision result of "yes".

[0287] First, the inquiry analysis unit 131 analyzes the inquiry information using the inquiry analysis process described using the flowchart in Figure 6 and obtains two or more inquiry attribute values. Specifically, the channel acquisition means 1311 obtains a channel identifier (e.g., "chat") that identifies the channel through which the inquiry information (e.g., "I cannot log in, so I cannot change my address...") was received. The department determination means 1312 obtains the organization identifier "Company A" that is paired with the inquiry information. Furthermore, the department determination means 1312 obtains a department identifier (e.g., "Department AA") that corresponds to the inquiry information using the department determination process described using the flowchart in Figure 7.

[0288] Furthermore, the inquiry attribute value acquisition means 1313 obtains the inquiry analysis prompt from the prompt management table (Figure 17). Next, the inquiry attribute value acquisition means 1313 provides the inquiry information and the inquiry analysis prompt to the generating AI. Then, the inquiry attribute value acquisition means 1313 obtains the answer from the generating AI. Next, the inquiry attribute value acquisition means 1313 obtains one or more inquiry attribute values ​​from the answer. Here, let's assume that the obtained inquiry attribute values ​​are, for example, emotion "neutral", text type "request", urgency "medium", and final result "pending". Also, let's assume that the inquiry attribute value acquisition means 1313 vectorizes the inquiry information and obtains an inquiry vector (x1, x2, ...). Furthermore, the inquiry attribute value acquisition means 1313 obtains the date and time "2024 / 11 / 28 13:11" from a clock (not shown).

[0289] Next, the inquiry storage unit 133 associates two or more inquiry attribute values ​​with inquiry information and stores them in the inquiry management table (Figure 16). Such records are those among the "ID=1" records in Figure 16 that do not have a reason or group identifier.

[0290] Next, the department inquiry processing unit 132 determines whether the inquiry information matches the department notification conditions. Here, the department identifier paired with the inquiry information is "Department AA," and the notification flag "2" paired with the department identifier is obtained. Since notification flag "2" indicates that all inquiry information will be sent, the department inquiry processing unit 132 determines that the inquiry information matches the department notification conditions. The department inquiry processing unit 132 obtains the department identifier "Department AA." The department inquiry processing unit 132 obtains the contact information "aa@x.jp" paired with the department identifier "Department AA." The department inquiry processing unit 132 sends the inquiry information, etc., via email to the contact indicated by the contact information. Through the above process, the person in charge of "Department AA" can immediately view the inquiry information in their mail client.

[0291] (Specific example 2) The processing unit 13 determines that it is time to analyze the inquiry information. Then, the processing unit 13 processes the inquiry set for each organization as follows. Here, we will explain using the organization identifier "Company A" as an example.

[0292] Processing unit 13 obtains the organization identifier "Company A". Next, processing unit 13 uses all query information that is paired with the organization identifier "Company A" to perform the organization analysis process described using the flowchart in Figure 8.

[0293] First, the processing unit 13 retrieves all inquiry information that corresponds to the organization identifier "Company A" from the inquiry information management table (Figure 16).

[0294] Next, the classification determination unit 134 determines two or more reasons for the inquiry information of the organization identifier "Company A" through the reason determination process described using the flowchart of FIG. 9. That is, the classification determination unit 134 acquires the classification determination prompt from the prompt management table (FIG. 18). Next, the classification determination unit 134 substitutes a number of inquiry information acquired in the variable <inquiry set> of the classification determination prompt and constructs a classification determination prompt to be passed to the generation AI. Next, the classification determination unit 134 passes the classification determination prompt to the generation AI. Next, the classification determination unit 134 acquires an answer from the generation AI. Next, the classification determination unit 134 acquires two or more reasons from the answer. Here, assume that the two or more reasons are "card loss", "login related", "address change", "payment delay", etc. Next, the classification determination unit 134 accumulates the two or more acquired reasons in association with the organization identifier "Company A".

[0295] Next, the classification unit 135 acquires two or more reasons such as "card loss", "login related", "address change", "payment delay", etc. acquired by the classification determination unit 134. Also, the classification unit 135 acquires the classification prompt from the prompt management table (FIG. 18). Next, the classification unit 135 acquires the inquiry information to be classified that pairs with the organization identifier "Company A". Next, the classification unit 135 substitutes the acquired inquiry information (for example, "Unable to login, so unable to change address...") into the variable <inquiry information> of the classification prompt and substitutes the two or more acquired reasons into the variable <reason> to construct a classification prompt to be passed to the generation AI. Next, the classification unit 135 passes the classification prompt to the generation AI. The classification unit 135 acquires an answer from the generation AI. The classification unit 135 acquires reasons (for example, "login related", "address change") from the answer. Next, the classification unit 135 associates the reasons with the inquiry information and accumulates them. Such information is the "reason" information in the record of "ID=1" in FIG. 16. Perform the above process for all the inquiry information to be classified of "Company A", and assume that one or more reasons are associated with all the inquiry information.

[0296] Next, the group determination unit 136 performs the grouping process described using the flowchart in Figure 11, and groups the inquiry information for analysis for "Company A".

[0297] Next, the representative determination unit 137 determines the representative inquiry information for each of the two or more groups of inquiry information to be analyzed for "Company A" using the representative determination process explained using the flowchart in Figure 12.

[0298] Furthermore, the statistical processing unit 139 performs statistical processing on all inquiry information of "Company A" that is subject to analysis, using the process described with reference to the flowchart in Figure 13.

[0299] Through the above processing, each inquiry piece of data analyzed for "Company A" was associated with a reason and a group identifier. Furthermore, the statistical processing results of the inquiry data were associated with the organizational identifier "Company A" or the individual department identifiers of Company A.

[0300] Furthermore, representative inquiry information from each of the two or more groups may be notified, for example, to the person in charge or representative of Company A.

[0301] (Specific example 3) The improvement acquisition unit 138 acquires two or more inquiry pieces from the inquiry management table (Figure 16) that correspond to the organization identifier of the organization of interest (in this case, "Company A"). The improvement acquisition unit 138 also refers to the page storage unit 114 and acquires one or more web pages that correspond to the organization identifier "Company A".

[0302] Next, the improvement acquisition unit 138 acquires an improvement suggestion prompt from the prompt management table (Figure 18). Then, the improvement acquisition unit 138 places two or more acquired inquiry pieces into the variable <Inquiry Set> of the improvement suggestion prompt (Figure 18). Also, the improvement acquisition unit 138 places one or more acquired web pages into the variable <Web Page> of the improvement suggestion prompt (Figure 18). Then, the improvement acquisition unit 138 constructs an improvement suggestion prompt to be given to the generating AI. Next, the improvement acquisition unit 138 gives the said improvement suggestion prompt to the generating AI. Then, the improvement acquisition unit 138 acquires a response from the generating AI. Next, the improvement acquisition unit 138 acquires one or more improvement suggestions from the response and stores the said one or more improvement suggestions paired with the organizational identifier "Company A" of the organization of interest. An example of such improvement suggestions is shown in Figure 19.

[0303] As described above, according to this embodiment, the department to which a user's inquiry pertains can be automatically determined.

[0304] Furthermore, according to this embodiment, statistical processing results for each query channel can be obtained.

[0305] Furthermore, according to this embodiment, two or more queries can be used to automatically determine two or more reasons for a query, and two or more queries can be used to classify two or more queries.

[0306] Furthermore, according to this embodiment, two or more queries can be grouped into two or more groups, and a representative query can be determined from each of the two or more groups.

[0307] Furthermore, according to this embodiment, it is possible to suggest improvements to web pages using inquiries from users regarding the subject.

[0308] Furthermore, according to this embodiment, it is possible to make suggestions for improving web pages using user inquiries and responses regarding the subject.

[0309] Furthermore, according to this embodiment, the department can be determined only for inquiry information that requires processing.

[0310] In the above embodiment, the representative determination unit 137 may determine one or more representative inquiry information, and then the group determination unit 136 may determine that each inquiry information other than the representative inquiry information belongs to the group to which the closest (most similar) representative inquiry information belongs from among the two or more determined representative inquiry information.

[0311] The processing in this embodiment may be implemented by software. This software may be distributed by software download or the like. Alternatively, this software may be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments in this specification. The software that implements the inquiry processing device 1 in this embodiment is the following program. In other words, this program is a program that causes a computer to function as an inquiry processing unit that receives inquiry information, which is information about user inquiries from one or more channels, a department determination means that uses the department information of two or more departments in a department management unit that stores department information for two or more departments to determine the department corresponding to the inquiry information received by the inquiry receiving unit and obtain the department identifier of that department, and a department inquiry processing unit that processes the inquiry information using the department identifier obtained by the department determination means.

[0312] Figure 20 is a block diagram of a computer system 300 that executes the program described herein to realize the inquiry processing device 1 and other devices of the various embodiments described above.

[0313] In Figure 20, the computer system 300 includes a computer 301 with a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

[0314] In Figure 20, the computer 301 includes, in addition to the CD-ROM drive 3012, an MPU 3013, a bus 3014 connected to the CD-ROM drive 3012, a ROM 3015 for storing programs such as boot-up programs, a RAM 3016 connected to the MPU 3013 for temporarily storing instructions for application programs and providing temporary storage space, and a hard disk 3017 for storing application programs, system programs, and data. Although not shown here, the computer 301 may further include a network card that provides connectivity to a LAN.

[0315] The program that causes the computer system 300 to execute functions such as the inquiry processing device 1 of the above-described embodiment may be stored on the CD-ROM 3101, inserted into the CD-ROM drive 3012, and then transferred to the hard disk 3017. Alternatively, the program may be transmitted to the computer 301 via a network (not shown) and stored on the hard disk 3017. The program is loaded into the RAM 3016 during execution. The program may also be loaded directly from the CD-ROM 3101 or the network.

[0316] The program does not necessarily have to include an operating system (OS) or third-party program that causes the computer 301 to execute functions such as the query processing device 1 of the above-described embodiment. The program only needs to include the instruction portion that calls appropriate functions (modules) in a controlled manner and obtains the desired result. How the computer system 300 operates is well known, so a detailed explanation is omitted.

[0317] In the above program, steps such as sending information and receiving information do not include hardware-based processing, such as processing performed by a modem or interface card in the transmission step (processing that can only be performed by hardware).

[0318] Furthermore, the computer running the above program may be a single computer or multiple computers. In other words, it may perform centralized processing or distributed processing.

[0319] Furthermore, it goes without saying that in each of the above embodiments, two or more communication means present in a single device may be physically implemented in a single medium.

[0320] Furthermore, in each of the above embodiments, each process may be implemented by centralized processing by a single device, or by distributed processing by multiple devices.

[0321] It goes without saying that the present invention is not limited to the embodiments described above, and various modifications are possible, all of which are also included within the scope of the present invention. [Industrial applicability]

[0322] As described above, the inquiry processing device 1 according to the present invention has the effect of enabling the use of user inquiries for business improvements, etc., as a result of being unable to properly process user inquiries about the subject, and is useful as a server, etc., for processing inquiry information sent from one or more call centers. [Explanation of Symbols]

[0323] A Information Systems 1. Inquiry Processing Device 2 Terminal devices 3 Generation AI device 11 Storage Unit 12 Reception Department 13 Processing Unit 14 Output section 21 Terminal storage section 22 Terminal Reception Section 23 Terminal Processing Unit 24 Terminal transmission unit 25 Receiving part of the terminal 26 Terminal output section 111 Department Management Department 112 Inquiry Management Department 113 Prompt Management Department Page 114 storage section 121 Inquiry Reception Department 130 Judgment Department 131 Inquiry Analysis Department 132 Department Inquiry Processing Section 133 Inquiry Storage Department 134 Classification determination unit 135 Classification Department 136 Group Determination Section 137 Representative Determination Department 138 Improvement Acquisition Department 139 Statistical Processing Department 141 Representative Output Section 142 Improved Output Section 143 Statistical Output Section 1311 Channel acquisition method 1312 Department determination means 1313 Inquiry attribute value acquisition method

Claims

1. An inquiry reception department that receives inquiry information, which is information about user inquiries from one or more channels, A department determination means that uses the department information of two or more departments stored in the department management unit to determine the department corresponding to the inquiry information received by the inquiry reception unit and obtains the department identifier of that department, An inquiry processing device comprising a department inquiry processing unit that processes the inquiry information using the department identifier acquired by the department determination means.

2. The aforementioned inquiry reception department, We accept inquiry information from two or more channels. An inquiry storage unit stores the inquiry information received by the inquiry receiving unit in an inquiry management unit where one or more inquiry pieces are stored, paired with a channel identifier that identifies the channel of the inquiry. A statistical processing unit that obtains statistical processing results for the inquiry information for each of the two or more channel identifiers, The inquiry processing apparatus according to claim 1, further comprising a statistical output unit that outputs the statistical processing results acquired by the statistical processing unit.

3. The aforementioned inquiry reception department, We receive two or more of the above inquiry information, A classification determination unit that uses some or all of the two or more inquiry pieces received by the inquiry reception unit to obtain two or more reasons for the inquiry piece, A classification unit that determines one or more reasons from among the two or more reasons acquired by the classification determination unit, to which each of the two or more inquiry pieces received by the inquiry reception unit belongs, The inquiry processing apparatus according to claim 1, further comprising: an inquiry management unit that stores one or more inquiry pieces in association with one or more reasons determined by the classification unit; and an inquiry storage unit that stores the two or more inquiry pieces received by the inquiry reception unit.

4. The aforementioned inquiry reception department, We accept inquiry information from two or more channels. A group determination unit that groups the two or more inquiry pieces received by the inquiry reception unit and determines two or more groups, For each of the two or more groups mentioned above, there is a representative determination unit that determines representative inquiry information, which is inquiry information that represents the group, The inquiry processing apparatus according to claim 1, further comprising: a representative output unit that outputs the representative inquiry information determined by the representative determination unit.

5. An improvement acquisition unit that refers to a page storage unit that stores one or more web pages related to the subject of the inquiry, and uses the inquiry information received by the inquiry reception unit to acquire improvement suggestions for the one or more web pages, The inquiry processing apparatus according to any one of claims 1 to 4, further comprising an improvement output unit that outputs the aforementioned improvement proposal.

6. The aforementioned inquiry reception department, We also accept response information corresponding to the aforementioned inquiry information. The aforementioned improvement acquisition unit is, The inquiry processing device according to claim 5, which obtains the improvement proposal using the inquiry information and the response information received by the inquiry receiving department.

7. The system further comprises a determination unit that determines whether the inquiry information received by the inquiry reception unit satisfies the processing conditions, The aforementioned means for determining the department is, The inquiry processing apparatus according to any one of claims 1 to 6, wherein the determination unit determines that the inquiry information satisfies the processing conditions and obtains a department identifier only for that inquiry information.

8. A query processing method comprising all the steps performed by the query processing device described in any one of claims 1 to 7.

9. Computers, A program for causing a query processing device to function as described in any one of claims 1 to 7.

Images