Information processing device and method

The information processing device addresses the issue of unresolved document flaws in automatic reply systems by extracting similar inquiries and providing timely document improvement suggestions, reducing maintenance costs through proactive engineer intervention.

JP2024168898AActive Publication Date: 2024-12-05HITACHI LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2023085934
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-25
Publication Date
2024-12-05
Estimated Expiration
2043-05-25

AI Technical Summary

Technical Problem

Existing automatic reply systems for customer inquiries in equipment maintenance fail to analyze individual inquiries in real time, leading to unresolved issues with document flaws and increased maintenance costs due to engineers' inability to respond promptly to document deficiencies.

Method used

An information processing device and method that stores past inquiries and their solutions, extracts similar cases, determines document improvement needs based on inquiry frequency, equipment operation years, and customer history, and provides engineers with timely improvement suggestions.

Benefits of technology

Reduces maintenance costs by allowing engineers to address document deficiencies proactively, thereby reducing the number of cases they need to handle.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024168898000001_ABST
    Figure 2024168898000001_ABST
Patent Text Reader

Abstract

To propose an information processing device and method capable of reducing maintenance cost.SOLUTION: Past cases being same as or similar to a new inquiry are extracted, whether the necessity of improvement of a document provided to a client due to an apparatus is determined if the past case same as or similar to the new inquiry can be extracted, a person in charge is requested to improve the document in the case of determining that the improvement of the document is required, and the determination of the necessity of the document improvement is performed on the basis of at least one piece of information of the number of times of receiving an inquiry same as or similar to the new inquiry, the number of inquiring times of an inquiry source of the new inquiry, the number of operation years of the apparatus of the inquiry source, and whether the inquiry source of the new inquiry is the client who made an inquiry requiring document improvement in the past.SELECTED DRAWING: Figure 6
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an information processing apparatus and method, and is suitable for application to an information processing apparatus that presents solutions to customer inquiries about equipment to the customer. [Background technology]

[0002] Traditionally, in maintenance work for computer equipment and the like, engineers would respond to inquiries from customers, such as "I don't know how to use the equipment" or "I don't know how to set environmental variables," by analyzing the content of the inquiry and replying to the customer with a solution.

[0003] At this time, emails between the customer and the engineer are exchanged repeatedly until the problem is resolved. After the problem is resolved, the engineer also considers whether or not to improve the contents of the materials (hereinafter referred to as "documents") provided to the customer along with the equipment, such as manuals and release notes, and improves such documents if necessary.

[0004] However, in recent years, the number of engineers working on software products with a long history, such as mainframes, has been decreasing, and as a result, the number of products that each engineer must maintain is increasing. For this reason, some kind of initiative to reduce the burden on engineers is urgently needed in the field of maintenance work.

[0005] In view of this situation, in recent years, a system (hereinafter referred to as an automatic reply system) has been proposed and put into practical use, which searches for similar past cases according to the content of a customer inquiry, and automatically replies to the inquiry source with a solution from the past case when there is a similar past case (for example, Patent Document 1). By introducing such an automatic reply system, it is possible to reduce the number of inquiries that engineers must deal with, and thus reduce the workload of engineers. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2001-282944 A Summary of the Invention [Problem to be solved by the invention]

[0007] However, when such an automatic reply system is introduced, there is a problem that the content of each inquiry cannot be analyzed in real time because engineers are not available to respond to many inquiries. Therefore, even if there are errors in the contents of documents such as manuals, or the descriptions are difficult to understand, the defects are left unaddressed for a long period of time, and there is a risk that engineers will be forced to respond to the inquiries.

[0008] Therefore, in an automatic reply system such as that described above, if the person in charge can be given an opportunity to improve the document at the appropriate time, the number of cases that the person in charge must deal with due to document deficiencies can be reduced, and as a result, maintenance costs can be reduced.

[0009] The present invention has been made in consideration of the above points, and aims to propose an information processing device and method capable of reducing maintenance costs. [Means for solving the problem]

[0010] In order to solve this problem, in the present invention, an information processing device stores the contents of past inquiries from customers about equipment and solutions to the inquiries as past cases, and, when a new inquiry is received, extracts the past case whose contents are identical or similar to the new inquiry, and presents the solution to the extracted past case as a solution to the new inquiry. The information processing device includes a similar case extraction unit that extracts the past case that is identical or similar to the new inquiry, and an improvement need determination unit that, when the similar case extraction unit is able to extract the past case that is identical or similar to the new inquiry, determines whether or not a document provided to the customer in association with the equipment needs to be improved, and, when it is determined that improvement of the document is necessary, requests a person in charge to improve the document. The improvement need determination unit determines whether or not the document needs to be improved based on at least one of information including the number of times an inquiry that is the same as or similar to the new inquiry has been received, the number of inquiries made by the source of the new inquiry and the number of years in which the equipment of the source of the inquiry has been in operation, and whether or not the source of the new inquiry is a customer who has previously made an inquiry that a document needs improvement.

[0011] In addition, the present invention provides an information processing method executed by an information processing device which stores the contents of past inquiries from customers regarding equipment and solutions to those inquiries as past cases, and when a new inquiry is received, extracts the past case whose contents are identical or similar to the new inquiry, and presents the solution to the extracted past case as a solution to the new inquiry, the information processing method including a first step of extracting the past case that is identical or similar to the new inquiry, and when a past case that is identical or similar to the new inquiry can be extracted, determining whether or not a document provided to the customer in association with the equipment needs to be improved, and when it is determined that improvement of the document is necessary, requesting a person in charge to improve the document, and in the second step, the information processing device determines whether or not the document needs to be improved based on at least one of information among the number of times inquiries that are the same as or similar to the new inquiry have been received, the number of inquiries made by the source of the new inquiry and the number of years in which the equipment of the source of the inquiry has been in operation, and whether or not the source of the new inquiry is a customer who has previously made an inquiry that requires improvement of a document.

[0012] According to the information processing device and method of the present invention, it is possible to give the person in charge an opportunity to improve such documents at an appropriate time, thereby reducing the number of cases that the person in charge must deal with due to document defects. Effect of the Invention

[0013] According to the present invention, it is possible to realize an information processing device and method that can reduce maintenance costs. [Brief description of the drawings]

[0014] [Figure 1] 1 is a block diagram showing the overall configuration of an automatic response system according to an embodiment of the present invention; [Diagram 2] 1 is a diagram showing an example of the configuration of an inquiry database. [Diagram 3] 1 is a diagram showing an example of a configuration of a system configuration database. [Figure 4]1 is a diagram illustrating an example of the configuration of a document database. [Diagram 5] 1 is a diagram showing an example of the configuration of an inquiry customer database. [Figure 6] FIG. 13 is a block diagram showing a sequence of steps in a document improvement support process. [Figure 7] 11 is a table showing an example of the configuration of similar customer list information. [Figure 8] 13 is a flowchart showing a procedure for document improvement necessity determination processing; [Figure 9A] 13 is a flowchart showing a processing procedure for document improvement proposal creation processing. [Figure 9B] 13 is a flowchart showing a processing procedure for document improvement proposal creation processing. [Figure 10A] 13 is a flowchart showing a processing procedure of a similar customer extraction process. [Figure 10B] 13 is a flowchart showing a processing procedure of a similar customer extraction process. [Figure 11A] 13 is a flowchart showing a processing procedure for customer-specific notification priority determination processing; [Figure 11B] 13 is a flowchart showing a processing procedure for customer-specific notification priority determination processing; [Figure 12A] 13 is a flowchart showing a processing procedure for customer-specific notification content creation processing. [Figure 12B] 13 is a flowchart showing a processing procedure for customer-specific notification content creation processing. [Figure 13A] 13 is a flowchart showing a processing procedure for an inquiry customer creation process. [Figure 13B] 13 is a flowchart showing a processing procedure for an inquiry customer creation process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] An embodiment of the present invention will now be described in detail with reference to the drawings.

[0016] (1) Configuration of the automatic response system according to this embodiment In Fig. 1, reference numeral 1 denotes an automatic answering system according to the present embodiment as a whole. This automatic answering system 1 is configured with first and second clients 3 and 4, an automatic answering server 5, and a file server 6, which are connected to each other via a network 2.

[0017] The first client 3 is a communication terminal device used by a customer who uses a device to be supported, and is configured from a general-purpose computer device such as a personal computer or a tablet capable of sending and receiving e-mail.

[0018] The second client 4 is a communication terminal device used by an engineer to respond to an inquiry on behalf of the automatic response server 5 when the automatic response server 5 cannot find a past case (hereinafter referred to as a similar case) whose content is identical or similar to an inquiry about a device to be supported (hereinafter referred to as a supported device) given by a customer as described below. The second client 4, like the first client 3, is composed of a general-purpose computer device capable of sending and receiving e-mail, such as a personal computer device or a tablet.

[0019] The automatic response server 5 is a server device having a function of searching for similar cases in response to the above-mentioned inquiry given from the first client 3, automatically responding to the inquiry source with a solution for the similar case for an inquiry for which a similar case was found, and notifying the engineer of an inquiry for which a similar case was not found. This automatic response server 5 is configured with a CPU (Central Processing Unit) 10, a memory 11, a storage device 12, a network adapter 13, an input device 14, and a display device 15.

[0020] The CPU 10 is a processor that controls the overall operation of the automatic response server 5. The memory 11 is composed of, for example, a nonvolatile semiconductor memory, and is used as a working memory for the CPU 10. The storage device 12 is composed of a large-capacity nonvolatile storage device such as a hard disk device or SSD (Solid State Drive), and is used to store and hold various programs and data that needs to be stored for a long period of time.

[0021] The programs stored in the storage device 12 are read from the storage device 12 to the memory 11 when the automatic response server 5 is started up or when needed, and the programs read to the memory 11 are executed by the CPU 10 to perform various processes of the automatic response server 5 as a whole, as described below.

[0022] The network adapter 13 is an adapter for connecting the automatic response server 5 to the network 2, and is composed of, for example, a NIC (Network Interface Card), a network card, or a LAN (Local Area Network) card.

[0023] The input device 14 is composed of, for example, a keyboard, a mouse, etc., and is used to input various information and operations to the automatic response server 5. The display device 15 is composed of, for example, a liquid crystal display, an organic EL (Electro-Luminescence) display, etc., and is used to display necessary information. Note that, instead of the input device 14 and the display device 15, a touch panel in which these are integrated may be applied.

[0024] The file server 6 is composed of a general-purpose server device equipped with a CPU, memory, storage device, and network adapter (not shown). In response to a request from the automatic response server 5, the file server 6 transmits the requested data to the automatic response server 5 via the network 2.

[0025] (2) Document Improvement Support Function According to the Present Embodiment Next, a document improvement support function installed in the automatic response server 5 according to this embodiment will be described. When a similar case of technical inquiry from a customer regarding the device to be supported can be detected as described above (i.e., when an automatic response is made), this document improvement support function is a function that judges whether or not the contents of documents such as manuals and release notes provided to the customer along with the device to be supported need to be improved, and if it is judged that improvement is necessary, notifies the engineer of that fact and a proposal for improving the document.

[0026] Technical inquiries from customers regarding the Supported Equipment shall be related to the software installed in the Supported Equipment, and such software shall be referred to below as "software related to the inquiry." Furthermore, documentation such as manuals and release notes shall be created for each piece of software, and documentation created for a piece of software shall be referred to below as "documentation corresponding to that software."

[0027] The automatic response server 5 also extracts other customers who use supported devices that have the same or similar configuration as the supported devices used by the customer who made the inquiry and that have software related to the inquiry installed as similar customers.The automatic response server 5 then sends warning notices to these similar customers, with contents according to the degree of similarity of the supported devices.At this time, the warning notices include at least the contents of the inquiry and the contents of the automatic response server's response to the inquiry, which makes it possible to prevent similar customers from making similar inquiries.

[0028] As means for realizing the document improvement support function of this embodiment, as shown in FIG. 1, the storage device 12 of the automatic response server 5 stores a control unit 20, an automatic application filing / registration unit 21, a similar case extraction unit 22, an automatic response unit 23, an inquiry information notification unit 24, and an inquiry customer database creation unit 25, and the file server 6 stores an inquiry database 26, a system configuration database 27, a document database 28, an inquiry customer database 29, and a definition file 30.

[0029] The control unit 20 is a program having a function of executing a part of the control process related to the document improvement support function.

[0030] The automatic form creation and registration unit 21 is a program that has the function of creating mail information of a prescribed format according to the contents of the inquiry mail or reply mail when a customer sends an email (hereinafter referred to as an inquiry mail) for a technical inquiry about a supported device to the automatic response server 5, or when the automatic response server 5 or an engineer sends an email (hereinafter referred to as a reply mail) in response to the inquiry to the customer, and registering the created mail information in the inquiry database 26 held by the file server 6.

[0031] The similar case extraction unit 22 is a program having a function of searching and extracting, from the inquiry database 26, past cases in which the content of the inquiry is identical or similar to that of a new inquiry email (hereinafter referred to as a new inquiry email), or past cases in which the content of the response to the inquiry is identical or similar, as similar cases to the inquiry.

[0032] In practice, the similar case extraction unit 22 uses techniques such as morphological analysis and TF-IDF (Term Frequency-Inverse Document Frequency) to extract all words with high importance as keywords from the text of a new inquiry email and from the email information in the text of past inquiry emails and reply emails registered in the inquiry database 26. The similar case extraction unit 22 then extracts, as keywords, the number of keywords extracted from the inquiry email or reply email that overlap with the keywords extracted from the new inquiry email, T, using the following equation:

number

[0033] The automatic response unit 23 is a program having a function of, when the similar case extraction unit 22 extracts a similar case, creating a reply email based on the email information of the similar case registered in the inquiry database 26, and sending the created reply email to the customer who made the inquiry. If the similar case extraction unit 22 is unable to extract a similar case, the control unit 20 sends a new inquiry email and a notice to that effect to the second client 4, requesting an engineer to respond to the inquiry via the new inquiry email.

[0034] On the other hand, the inquiry information notification unit 24 is configured with a document improvement necessity determination unit 40, a document improvement proposal creation unit 41, a similar customer extraction unit 42, a customer-specific notification priority determination unit 43, a customer-specific notification content creation unit 44, and an inquiry customer database creation unit 45.

[0035] The document improvement necessity determination unit 40 is a program having a function of determining, when the similar case extraction unit 22 has extracted a similar case of the new inquiry email, whether or not the contents of the document describing the software related to the inquiry should be improved.

[0036] In this embodiment, the document improvement necessity determination unit 40 determines that the content of the document corresponding to the software needs to be improved if the inquiry at that time satisfies any of the following determination conditions (A) to (C).

[0037] (A) If there have been α or more inquiries identical or similar to the inquiry in the past (B) The inquiry is from a customer whose operating years are β or more and who has made inquiries γ or more times. (C) The inquiry is from a customer who has previously made an inquiry requiring improvements to the document.

[0038] Here, (A) is because there is a possibility that the description in the document is incomprehensible to the customer, and (B) is because if it is difficult for an experienced staff member to understand, it may be difficult for other customers to understand as well. Furthermore, (C) is because it is worthwhile to determine whether or not the corresponding document needs to be improved for inquiries from customers who have made appropriate inquiries in the past. Note that "α" in (A) and "β" and "γ" in (B) are all predefined thresholds.

[0039] The document improvement proposal creation unit 41 is a program having a function of creating an improvement proposal for the document from the contents of the new inquiry email and the reply email to the new inquiry email, and the contents of the document, and proposing the proposal to the engineer, when the document improvement necessity determination unit 40 determines that the contents of the document should be improved. Note that the "improvement proposal" here refers to a proposal that is just advice on how to improve the document.

[0040] In practice, the document improvement proposal creation unit 41 extracts all important words (keywords) from the body of a new inquiry email and the body of a reply email to that new inquiry email as a keyword group, and calculates the degree of similarity between the extracted keyword group and the contents of each document corresponding to the inquiry from the new inquiry email (each document describing software related to the inquiry) as a similarity.

[0041] In this case, the similarity can be calculated, for example, by extracting the keywords contained in the body of a new inquiry email and the reply email to that new inquiry email as a first group of keywords, and extracting the keywords contained in each of these documents as a second group of keywords, and then calculating the ratio of the number of overlapping keywords between the first and second keyword groups to the total number of keywords contained in the first and second keyword groups.

[0042] Then, for documents with a calculated similarity of, for example, 70% or more, the document improvement proposal creation unit 41 determines that the document contains a description related to the inquiry but is written in a way that is difficult for the user to notice, and creates an improvement proposal to make the relevant part of the document more prominent, such as "changing the document structure by moving the description or changing the font, etc.", and notifies the engineer of this.

[0043] In addition, for documents with a similarity of, for example, 30% or more but less than 70%, the document improvement proposal creation unit 41 assumes that the document contains a description related to the inquiry but is written in a way that makes the meaning of the description difficult to understand, and creates an improvement proposal to rewrite the relevant part of the document to make it easier to understand, such as "rewriting the sentences to make them easier to understand or changing the writing style to bullet points, etc." and notifies the engineer of this.

[0044] Furthermore, for documents with a similarity of, for example, less than 30%, the document improvement proposal creation unit 41 determines that the document does not contain an explanation related to the inquiry, and creates an improvement proposal such as "add new explanation" to add the necessary explanation to the relevant part of the document and notifies the engineer.

[0045] The similar customer extraction unit 42 is a program having the function of extracting customers (similar customers) who are using supported equipment having the same or similar configuration as the supported equipment used by the customer who is the sender (inquiry source) of the new inquiry email (hereinafter referred to as the new inquiring customer).

[0046] In practice, the similar customer extraction unit 42 sequentially compares the configuration of the supported equipment used by the new inquiring customer with the configuration of the supported equipment used by each customer other than the new inquiring customer, and calculates a score (hereinafter referred to as a system score) for each customer other than the new inquiring customer, the score being defined so that the greater the amount of overlapping software and the closer the versions of the overlapping software are, the higher the score.

[0047] The similar customer extraction unit 42 extracts, as similar customers, customers whose system score calculated as described above is equal to or greater than a predefined threshold and who use a supported device in which software related to the inquiry from the new inquiring customer is installed. However, the similar customer extraction unit 42 may also extract, as similar customers, customers who use a supported device in which software related to the inquiry from the new inquiring customer is installed and who are in the top predetermined number of system scores.

[0048] The customer-specific notification priority determination unit 43 is a program having the function of determining, for each similar customer extracted by the similar customer extraction unit 42, the degree of similarity between the configuration of the supported equipment used by that similar customer and the configuration of the supported equipment used by the new inquiring customer, and determining the priority of sending a warning notification (hereinafter referred to as the notification priority) for each similar customer based on the determination result.

[0049] In practice, the customer-specific notification priority determination unit 43 compares the version of the software related to the new inquiry that is installed in the supported equipment used by the new inquiring customer and the setting values ​​of each parameter set for that software with the version of the software that is installed in the supported equipment used by each similar customer and the setting values ​​set for that software.

[0050] The customer-specific notification priority determination unit 43 determines the notification priority to be "important" for similar customers whose supported device versions and parameter settings match those of the new inquiring customer. The customer-specific notification priority determination unit also determines the notification priority to be "caution" for similar customers whose version and parameter settings match those of the new inquiring customer.

[0051] The customer-specific notification content creation unit 44 is a program having a function of creating a warning notification whose content corresponds to the notification priority of each similar customer and sending it to that similar customer.

[0052] In practice, for similar customers whose notification priority has been determined to be "important," the customer-specific notification content creation unit 44 creates a warning notification whose text is the inquiry content of the new inquiry email and the content of the reply email from the automatic response server 5 to that inquiry, and sends the warning notification to the similar customer.

[0053] In addition, for similar customers whose notification priority has been determined to be "caution," the customer-specific notification content creation unit 44 creates a notification that includes, in addition to the inquiry content of the new inquiry email and the content of the reply email from the automatic response server 5 to that inquiry, a message to the effect that changing the version or parameter settings of the software will cause problems, and sends the notification to that similar customer.

[0054] The inquiring customer database creation unit 45 is a program having a function of creating the inquiring customer database 29, which will be described later with reference to Fig. 5. The inquiring customer database creation unit 45 judges, for each customer, whether or not the customer satisfies the judgment condition (C) described above, and creates the inquiring customer database 29 based on the judgment result.

[0055] On the other hand, the inquiry database 26 is a database that stores mail information of all e-mails (inquiry mails and response mails) exchanged between the customer and the automatic response server 5 or engineers.

[0056] As shown in Fig. 2, the inquiry database 26 has a table structure including a registration number column 26A, a management number column 26B, a customer column 26C, a subject column 26D, a sender address column 26E, a destination address column 26F, a date column 26G, and a text column 26H. In the inquiry database 26, one record (row) in Fig. 2 corresponds to the email information of one inquiry email or reply email (hereinafter, these will be referred to as emails as appropriate) exchanged between the customer and the automatic response server 5 or an engineer.

[0057] The registration number field 26A stores a registration number that is assigned to the corresponding e-mail and is unique to that e-mail within the inquiry database 26. As this registration number, a consecutive number starting from "1" can be used.

[0058] In addition, in the management number column 26B, if the corresponding e-mail is an inquiry mail, a management number assigned to the inquiry mail and unique to the inquiry mail is stored, and if the corresponding e-mail is a reply mail, the management number of the inquiry mail corresponding to the reply mail is stored.

[0059] In addition, if multiple inquiry emails and response emails are exchanged derived from a single inquiry email, the same management number assigned to the first inquiry email will be assigned to each of the subsequent response emails and inquiry emails.

[0060] The customer column 26C stores the name of the sender if the corresponding e-mail is an inquiry e-mail, or the name of the destination customer (a company or government agency) if the e-mail is a reply e-mail, and the subject column 26D stores the subject of the e-mail.

[0061] Furthermore, the sender address field 26E stores the email address (sender address) of the sender of the email (customer's person in charge, the automatic response server 5, or an engineer), and the destination address field 26F stores the email address (destination address) of the destination of the email. The date field 26G stores the date and time when the email was sent, and the body field 26H stores the body of the email.

[0062] Therefore, in the example in Figure 2, the email assigned the registration number "1" and the management number "0001" is an inquiry email sent by a person in charge of "Company A" on "2019 / 4 / 1 9:41", with the subject "Confirming environmental variables", the sender address "sample.xxx@a.com", and the destination address "sample.zzz@hatachi.com", and the content of the body of the email was "This is xx from Company A. Environmental variables for software 1...".

[0063] The system configuration database 27 is a database that stores configuration information of supported devices used by each customer. As shown in Fig. 3, the system configuration database 27 has a table structure including a registration number column 27A, a customer column 27B, a contact column 27C, a number of years in operation column 27D, a configuration software group column 27E, a version column 27F, and a parameter column 27G. In the system configuration database 27, a record (row) corresponding to one registration number column 27A corresponds to one supported device.

[0064] The registration number field 27A stores a registration number that is assigned to the corresponding supported device and is unique to that supported device within the system configuration database 27. As this registration number, a consecutive number starting from "1" can be used.

[0065] The customer column 27B stores the name of the customer who uses the supported device. The contact column 27C stores the email address of the customer's contact (for example, the email address of the person in charge at the customer), and the years of operation column 27D stores the number of years of operation of the supported device.

[0066] The configuration software group column 27E, the version column 27F, and the parameter column 27G are each divided into sub-columns 27H corresponding to each piece of software installed in the corresponding supported device.

[0067] Each sub-column 27H in the configuration software group column 27E stores the name of the corresponding software installed in the corresponding supported device. Each sub-column 27H in the version column 27F stores the version of the corresponding software, and each sub-column 27H in the parameter column 27G stores the setting values ​​of all parameters set for the corresponding software.

[0068] Therefore, in the example of Figure 3, for example, the supported equipment assigned the equipment number "1" is an equipment that has been in operation for "15 years" and is used by "Company A," and the contact information for the person in charge of that supported equipment at "Company A" is "xx@A.com."

[0069] Figure 3 also shows that the supported device is equipped with at least the following software: "OS1" version "7.6", "Software 1" version "12.6", "Software 2" version "12.0", and "Software 3" version "12.0".

[0070] Furthermore, Figure 3 shows that for one piece of software called "Software 1," configuration values ​​have been set for a parameter called "prm," a maximum number of retries parameter called "rec," a retry interval parameter called "rei," and an environment variable parameter called "env."

[0071] The document database 28 is a database that stores information about documents such as manuals and release notes provided to customers in association with each supported device. As shown in Fig. 4, the document database 28 has a table structure with a registration number column 28A, a document name column 28B, a type column 28C, a software name column 28D, a version column 28E, a text column 28F, and a date column 28G. In the document database 28, one record (row) in Fig. 4 corresponds to one document provided to a customer in association with one of the supported devices.

[0072] The registration number column 28A stores a registration number that is assigned to the corresponding document and is unique to that document in the document database 28. A consecutive number starting from "1" can be used as this registration number. The document name column 28B stores the name of the corresponding document, and the type column 28C stores the type of the corresponding document.

[0073] Furthermore, the software name column 28D stores the software name of the software whose description is described in the corresponding document, and the version column 28E stores a numerical value (the value of the corresponding version) indicating which version of the corresponding software the document corresponds to. The text column 28F stores the data of the text of the corresponding document, and the date column 28G stores the date that the document was issued.

[0074] Therefore, in the example of Figure 4, the document with the registration number "1" is a "manual" describing version "1.0" of software called "Software 1," its name is "Software 1 Basic Guide," and it was published on "4 / 1 / 2000."

[0075] The inquiry customer database 29 is a database used to manage, for each customer staff member, whether or not that staff member has made an inquiry in the past that requires document improvement (an inquiry that triggers document improvement). As shown in Fig. 5, this inquiry customer database 29 has a table structure with a registration number column 29A, a contact column 29B, and an improvement flag column 29C. In the inquiry customer database 29, one record (row) in Fig. 5 corresponds to one staff member for one customer.

[0076] The registration number field 29A stores a registration number that is assigned to the corresponding person in charge and is unique to that person in the inquiry customer database 29. As this registration number, a consecutive number starting from "1" can be used.

[0077] Furthermore, the contact field 29B stores the email address of the corresponding person in charge, and the improvement flag field 29C stores a flag (improvement flag) indicating whether or not the person in charge has made an inquiry in the past that requires document improvement. In the example of Fig. 5, if the person in charge has made an inquiry in the past that requires document improvement, the improvement flag is set to "True", and if not, the improvement flag is set to "False".

[0078] Therefore, in the example in Figure 5, the contact information for the person in charge of registration number "1" is "sample.xxx@a.com", and it is shown that this person has previously made an inquiry that required the document to be improved (the improvement flag is "True").

[0079] The definition file 30 is a file in which various predefined threshold values ​​are stored.

[0080] (3) Flow of various processes (3-1) Overall processing flow Next, the flow of various processes executed by the automatic response server 5 in relation to the document improvement support function described above will be described.

[0081] Fig. 6 shows the flow of a series of processes (hereinafter, referred to as document improvement support processes) executed in the automatic response server 5 when the automatic response server 5 receives a new inquiry email from the first client 3. As shown in Fig. 6, when the automatic response server 5 receives a new inquiry email from the first client 3, first, the control unit 20 stores the new inquiry email in the memory 11 (Fig. 1) and passes it to the automatic application creation and registration unit 21 (S1).

[0082] When a new inquiry email is handed over from the control unit 20, the automatic application creation / registration unit 21 registers the email information of the new inquiry email in the inquiry database 26 (S2), and then calls the similar case extraction unit 22 (S3).

[0083] When the similar case extraction unit 22 is called by the automatic application issuing / registration unit 21, it searches the inquiry database 26 for past cases (similar cases) whose contents are identical or similar to that of the new inquiry email (S4). If the similar case extraction unit 22 cannot find such a similar case, it sends the new inquiry email together with a notice to that effect to the engineer. As a result, a response to the new inquiry email is made by that engineer.

[0084] On the other hand, if the similar case extraction unit 22 detects such a similar case, it extracts the email information of the inquiry email and reply email of the similar case from the inquiry database 26 (S5), and stores this extracted email information in the memory 11 as similar case information 50 (S6). Note that the configuration of similar case information 50 is the same as that of the inquiry database 26, so a detailed description of the configuration of similar case information 50 will be omitted. The similar case extraction unit 22 then calls the automatic response unit 23 (S7).

[0085] When called by the similar case extraction unit 22, the automatic response unit 23 reads out similar case information 50 stored in the memory 11 (S8), creates a reply email by referring to the read out similar case information 50, and sends the created reply email to the sender of the new inquiry email (inquiry source).

[0086] The automatic response unit 23 also stores the content of the body of the reply mail at this time in the memory 11 as response content information 51 (S9), and then calls the document improvement necessity determination unit 40 of the inquiry information notification unit 24 (S10). Note that this call may be made by the control unit 20.

[0087] When called by the automatic response unit 23, the document improvement necessity determination unit 40 collects necessary information from the inquiry database 26, the system configuration database 27, the document database 28, the inquiry customer database 29, and the memory 11 (S11).

[0088] Specifically, the document improvement necessity determination unit 40 acquires email information of the new inquiry email from the inquiry database 26, and acquires configuration information of the support target device used by the new inquiring customer from the system configuration database 27.

[0089] The document improvement necessity determination unit 40 also obtains documents related to the software inquired about in the new inquiry email from the document database 28, and obtains the improvement flag value of the new inquiring customer from the inquiring customer database 29. Furthermore, the document improvement necessity determination unit 40 obtains similar case information 50 stored in the memory 11 by the similar case extraction unit 22 in step S6.

[0090] Then, the document improvement necessity determination unit 40 determines whether or not the document acquired in step S11 (hereinafter, referred to as the target document) needs to be improved based on each piece of information collected in step S11. If the document improvement necessity determination unit 40 determines that the target document needs to be improved, it calls the document improvement proposal creation unit 41 (S12).

[0091] When called by the document improvement necessity determination unit 40, the document improvement proposal creation unit 41 acquires the response content information 51 stored in the memory 11 by the automatic response unit 23 in step S9, and reads the data of the target document from the document database 28 (S13).

[0092] Furthermore, the document improvement proposal creating unit 41 calculates the similarity between the contents of the body of the corresponding inquiry email and reply email and the contents of the target document for each document based on the response content information 51 acquired in step S13 and the data of the target document. Then, the document improvement proposal creating unit 41 creates the reason for improving the target document and an improvement proposal based on the calculation result, and stores the created improvement proposal in the memory 11 as document improvement proposal information 52 (S14).

[0093] The document improvement proposal creating unit 41 also sends this document improvement proposal information 52 and data on the document with the highest similarity among the target documents to the engineer.

[0094] Furthermore, if the reason for improving the target document created in step S14 is "not explained" (the similarity is 0 to 30% or less), the document improvement proposal creation unit 41 calls the similar customer extraction unit 42 (S15).

[0095] When the similar customer extraction unit 42 is called by the document improvement proposal creation unit 41, it reads out the configuration information of the support target device used by the new inquiring customer from the system configuration database 27 (S16).

[0096] The similar customer extraction unit 42 also searches the system configuration database 27 for supported devices having the same or similar configuration as the supported device used by the new inquiring customer whose configuration information was read out in step S16, and creates similar customer list information 53 as shown in Figure 7, in which the configuration information of each detected supported device and customer information of each customer (similar customer) who uses each of these supported devices are registered.

[0097] 7, the similar customer list information 53 has a table structure including a registration number column 53A, a customer column 53B, a contact column 53C, a system score ("system_score") column 53D, a configuration software group column 53E, a version column 53F, a parameter column 53G, a notification priority column 53H, and a difference column 53I. In the similar customer list information 53, one record (row) corresponding to a registration number described later corresponds to one similar customer.

[0098] The registration number field 53A stores a registration number that is assigned to the corresponding similar customer and is unique to that similar customer in the similar customer list information 53, and the customer field 53B stores the customer name of the corresponding similar customer.

[0099] In addition, the contact information field 53C stores the email address of the person in charge of the corresponding similar customer, and the system score field 53D stores the calculated similarity between the supported equipment used by the similar customer and the supported equipment used by the new inquiring customer.

[0100] The configuration software group column 53E, version column 53F, parameter column 53G, notification priority column 53H and difference column 53I are divided into sub-columns 53J corresponding to each software implemented in the supported equipment used by the corresponding similar customer.

[0101] Each small column 53J in the configuration software group column 53E stores the software name of the corresponding software, and each small column 53J in the version column 53F stores the version of that software. Each small column 53J in the parameter column 53G stores the setting values ​​of various parameters set for the corresponding software, and each small column 53J in the notification priority column 53H stores the notification priority when a notification priority is set for the corresponding software.

[0102] Furthermore, each sub-column 53J of the difference column 53I stores the difference between the corresponding software installed in the supported device used by the new inquiring customer and the corresponding software installed in the supported device used by the corresponding similar customer.

[0103] Such differences include the version and the parameter setting value. If the difference is the version, information "version" is stored in the corresponding sub-column 53J of difference column 53I, and if the difference is the parameter setting value, information "parameter" is stored in the corresponding sub-column 53J of difference column 53I.

[0104] The similar customer extraction unit 42 then stores the created similar customer list information 53 in the memory 11 (S17), and thereafter calls the customer-specific notification priority determination unit 43 (S18).

[0105] When called by the similar customer extraction unit 42, the customer-specific notification priority determination unit 43 reads the document improvement proposal information 52 and the similar customer list information 53 stored in the memory 11. Then, the customer-specific notification priority determination unit 43 determines the notification priority for each similar customer registered in the similar customer list information 53 based on the read document improvement proposal information 52, and stores the determination results in the similar customer list information 53 in sequence (S19).

[0106] Specifically, for a similar customer who is using a supported device that is equipped with the same software as that installed in the supported device used by the new inquiring customer and has completely matching software versions and parameter settings, the customer-specific notification priority determination unit 43 determines the notification priority to be “important” and stores “important” in the sub-column 53J of the notification priority column 53H corresponding to that software of that similar customer in the similar customer list information 53.

[0107] In addition, for similar customers who are using supported equipment that is equipped with the same software as that installed in the supported equipment used by the new inquiring customer, but in which the versions of the software and the settings of various parameters do not completely match, the customer-specific notification priority determination unit 43 determines the notification priority to be ``Caution'' and stores ``Caution'' in the sub-column 53J of the notification priority column 53H corresponding to the software of that similar customer in the similar customer list information 53.

[0108] Furthermore, the customer-specific notification priority determination unit 43 does not set a notification priority for software whose notification priority does not correspond to "important" or "caution." Therefore, in this case, the customer-specific notification priority determination unit 43 does not store anything in the sub-column 53J of the notification priority column 53H corresponding to that software in the similar customer list information 53.

[0109] Then, when the customer-specific notification priority determination unit 43 has completed determining the notification priority for the necessary software installed in the supported equipment used by each similar customer registered in the similar customer list information 53 as described above, it calls the customer-specific notification content creation unit 44 (S20).

[0110] When called by the customer-specific notification priority determination unit 43, the customer-specific notification content creation unit 44 reads the similar customer list information 53 (S21A), creates the notification content (text) of a warning notification for similar customers who have the notification priority of any of the software implemented in the supported device they are using set to "important" or "caution" (S21B), and sends the warning notification including the created notification content to the corresponding similar customers as customer-specific notification content information 54 (S22).

[0111] For example, for a warning notification for a similar customer who has the notification priority of any of the software implemented in the supported device being used set to "important," the customer-specific notification content creation unit 44 creates notification content that includes the content of the new inquiry email and the content of the reply email that the automatic response server 5 automatically sends to the new inquiry email, and sends a warning notification with the created notification content to the similar customer.

[0112] In addition, for a warning notification to a similar customer for which the notification priority of any of the software implemented in the supported equipment they use is set to "Caution," the customer-specific notification content creation unit 44 creates, in addition to the above-mentioned text, a text at the beginning of the text that includes a warning based on the difference between the supported equipment used by the similar customer and the supported equipment used by the new inquiring customer (equipment used by the new inquiring customer), such as "If parameter ○○ is changed to XX, the following event may occur," and sends a warning notification with the created notification content to the similar customer.

[0113] On the other hand, the control unit 20 calls the inquiry customer database creation unit 45 in accordance with an instruction for creating the inquiry customer database 29 given by the engineer at any time (S23).

[0114] When called by the control unit 20, the inquiring customer database creation unit 45 collects necessary information on each customer who has made an inquiry in the past (hereinafter, these customers are referred to as inquiring customers) from the inquiry database 26 (S24).

[0115] Specifically, the inquiring customer database creation unit 45 reads customer information such as the customer name and sender address of each inquiring customer, and the contents of the body of inquiry emails sent by those inquiring customers in the past, from the inquiry database 26. The inquiring customer database creation unit 45 also reads the software names, versions, and parameter settings of all software implemented in the supported devices used by those inquiring customers, from the system configuration database 27. The inquiring customer database creation unit 45 also reads documents such as manuals and release notes corresponding to each of those software from the document database 28.

[0116] The inquiring customer database creation unit 45 then stores the contact information of each inquiring customer collected in step S24 in the contact information column 29B (FIG. 5) of the inquiring customer database 29. The inquiring customer database creation unit 45 also identifies all of the inquiring customers who have made inquiries in the past that require document improvement, based on the information collected in step S24. The inquiring customer database creation unit 45 further sets the improvement flags stored in the improvement flag column 29C (FIG. 5) of each of these identified customers in the inquiring customer database 29 to "True," and sets the improvement flags stored in the improvement flag column 29C of the other inquiring customers to "False" (S25).

[0117] (3-2) Document Improvement Needs Determination Process Fig. 8 shows the flow of the document improvement necessity determination process executed by the document improvement necessity determination unit 40 called by the automatic response unit 23 in step S10 of the document improvement support process described above with reference to Fig. 6. The document improvement necessity determination unit 40 determines whether the document described about the software inquired about in the new inquiry email needs to be improved, according to the processing procedure shown in Fig. 8.

[0118] In practice, when called by the automatic response unit 23, the document improvement necessity determination unit 40 starts the document improvement necessity determination process shown in Fig. 8 and first reads out similar case information 50 stored in memory 11 (Fig. 1) by the similar case extraction unit 22 as described above (S30). The document improvement necessity determination unit 40 also reads out the predefined values ​​of thresholds α, β, and γ from the definition file 30 stored in the file server 6 (Fig. 1) (S31).

[0119] Next, the document improvement necessity determination unit 40 judges whether the number of similar cases extracted by the similar case extraction unit 22 is less than α (S32). If the document improvement necessity determination unit 40 obtains a negative result in this judgment, it judges that the document described in the software inquired about in the new inquiry email needs to be improved, and calls the document improvement proposal creation unit 41 (S33). The document improvement necessity determination unit 40 then ends this document necessity determination process.

[0120] In contrast, if the document improvement necessity determination unit 40 obtains a positive result in the determination in step S32, it reads the new inquiry email from the memory 11 and extracts the inquiry source email address (the email address of the new inquiring customer) from the read new inquiry email (S34).

[0121] Next, the document improvement necessity determination unit 40 acquires the number of years of operation of the supported equipment used by the new inquiring customer (S35). Specifically, the document improvement necessity determination unit 40 searches for records in the system configuration database 27 (FIG. 3) in which the email address extracted in step S34 is stored in the contact information column 27C (FIG. 3), and acquires the number of years of operation stored in the number of years of operation column 27D (FIG. 3) of the hit record.

[0122] Next, the document improvement necessity determination unit 40 counts the number of inquiries made in the past by new inquiring customers in the inquiry database 26 (S36). Specifically, the document improvement necessity determination unit 40 searches for all records in the inquiry database 26 whose sender email address stored in the sender address field 26E (FIG. 2) matches the email address of the new inquiring customer extracted in step S34, and counts the total number of such records.

[0123] Furthermore, the document improvement necessity determination unit 40 judges whether the number of years of operation acquired in step S35 is equal to or greater than a threshold value β and the number of past inquiries counted in step S36 is equal to or greater than a threshold value γ (S37). If the document improvement necessity determination unit 40 obtains a positive result in this judgment, it processes step S33 in the same manner as described above, and then ends this document improvement necessity determination process.

[0124] On the other hand, if the document improvement necessity determination unit 40 obtains a negative result in the determination at step S37, it checks the value of the improvement flag of the new inquiring customer stored in the inquiring customer database 29 (S38).

[0125] Specifically, the document improvement necessity determination unit 40 searches through the records in the inquiring customer database 29 for records in which the email address of the new inquiring customer acquired in step S34 is stored in the contact column 29B (Figure 5), and checks the value of the improvement flag stored in the improvement flag column 29C (Figure 5) of the found record.

[0126] Then, the document improvement necessity determination unit 40 judges whether the confirmed value of the improvement flag is "True" (S39). If the document improvement necessity determination unit 40 obtains a negative result in this judgment, it ends the document improvement necessity determination process.

[0127] On the other hand, if the document improvement necessity determination section 40 obtains a positive result in the determination at step S39, it processes step S33 in the same manner as described above, and thereafter ends the document improvement necessity determination process.

[0128] (3-3) Document improvement proposal creation process Figures 9A and 9B show the flow of the document improvement proposal creation process executed by the document improvement proposal creation unit 41 called by the document improvement necessity determination unit 40 in step S12 of the document improvement support process described above with reference to Figure 6. The document improvement proposal creation unit 41 creates improvement proposals for each document described in the software inquired about in the new inquiry email according to the processing procedure shown in Figures 9A and 9B.

[0129] In practice, when the document improvement proposal creation unit 41 is called by the document improvement necessity determination unit 40, it starts the document improvement proposal creation process shown in Figures 9A and 9B, and first resets (initializes) a parameter called "document_name" and a parameter called "score" (S40).

[0130] Next, the document improvement proposal creation unit 41 reads out the new inquiry email and the response content information 51 (FIG. 6) corresponding to this new inquiry email from the memory 11 (FIG. 1) (S41), and extracts all important keywords from the body of the new inquiry email and the response content information 51 read out using techniques such as morphological analysis and TF-IDF as an important keyword group MK (S42).

[0131] Next, the document improvement proposal creating unit 41 selects one document that has not been processed after step S44 from among the documents that describe the software inquired about in the new inquiry mail (S43).

[0132] The document improvement proposal creation unit 41 also reads out data of the selected document (hereinafter referred to as the selected document) from the document database, and based on the read out data, extracts all important keywords from the selected document as an important keyword group DK, similar to step S42 (S44).

[0133] Thereafter, the document improvement proposal creating unit 41 judges whether the number of overlapping keywords between the important keyword group MK extracted in step S42 and the important keyword group DK extracted in step S44 is the largest ever (S45). If the document improvement proposal creating unit 41 obtains a negative result in this judgment, it proceeds to step S48.

[0134] In response to this, if the document improvement proposal creating unit 41 obtains a positive result in the determination in step S45, it sets the value of the parameter "document_name" to the document name of the currently selected document (S46). Also, the document improvement proposal creating unit 41 sets the value of the parameter "score" to the ratio of the number of overlapping important keywords between the important keyword group MK and the important keyword group DK to the total number of important keywords in the important keyword group MK and the important keyword group DK extracted in step S42 (S47).

[0135] Specifically, the document improvement proposal creating unit 41 determines the number of overlapping keywords between the important keyword group MK extracted in step S42 and the important keyword group DK extracted in step S44 as the "overlap number" and calculates the following formula:

number

[0136] Next, the document improvement proposal creating unit 41 judges whether or not the processes of steps S44 to S47 have been executed for all documents describing the software inquired about in the new inquiry mail (S48).

[0137] If the document improvement proposal creating unit 41 obtains a negative result in this determination, it returns to step S43, and thereafter repeats the processes of steps S43 to S48 while sequentially switching the document selected in step S43 to other documents that have not been processed in steps S44 to S47.

[0138] Through this repetitive process, the document name of the document with the largest number of overlapping keywords between the important keyword group MK extracted in step S42 and the important keyword group DK extracted in step S44 from among all documents describing the software inquired about in the new inquiry mail is finally set as the value of the parameter "document_name." Also, the "score" value calculated in step S47 for that document is set as the final "score" value.

[0139] When the document improvement proposal creating unit 41 obtains a positive result in step S48 by completing the processes of steps S44 to S47 for all documents that describe the software inquired about in the new inquiry mail, it judges whether the current value of the parameter "score" is equal to or less than "0.3" (S49). If the document improvement proposal creating unit 41 obtains a negative result in this judgment, it proceeds to step S54.

[0140] In response to this, if the document improvement proposal creation unit 41 obtains a positive result in the judgment in step S49, it sets the value of the parameter "reason", which indicates the reason why the document should be improved, to "not explained" (S50), and sets the value of the parameter "handle", which indicates the content of the document improvement proposal to be proposed to the engineer, to "add new explanation" (S51).

[0141] Then, the document improvement proposal creating unit 41 writes the current parameter values ​​of "document_name", "score", "reason", and "handle" as document improvement proposal information 52 (FIG. 6) in the memory 11, and notifies the engineer of this document improvement proposal information 52 (S52). Furthermore, the document improvement proposal creating unit 41 calls the similar customer extraction unit 42 (S53), and then ends this document improvement proposal creating process.

[0142] On the other hand, if the document improvement proposal creating unit 41 obtains a negative result in the determination in step S49, it determines whether or not the current value of the parameter "score" is less than "0.7" (S54). If the document improvement proposal creating unit 41 obtains a positive result in this determination, it sets the value of the parameter "reason" to "the meaning is difficult to understand" (S55), and sets the value of the parameter "handle" to "the sentence needs to be rewritten or changed to bullet points" (S56), and then proceeds to step S59.

[0143] Furthermore, if the document improvement proposal creation unit 41 obtains a negative result in the judgment in step S54, it sets the value of the parameter "reason" to "hard to notice" (S57), and sets the value of the parameter "handle" to "make it stand out, such as by emphasizing the display" (S58).

[0144] Then, the document improvement proposal creation unit 41 writes the current parameter values ​​of "document_name", "score", "reason", and "handle" into the memory 11 as document improvement proposal information 52 (Figure 6), and notifies the engineer of this document improvement proposal information 52 (S59), and then terminates the document improvement proposal creation process.

[0145] (3-4) Similar customer extraction process 10A and 10B show the flow of a similar customer extraction process executed by the similar customer extraction unit 42 called by the document improvement plan creation unit 41 in step S15 of the document improvement support process described above with reference to Fig. 6. The similar customer extraction unit 42 extracts, according to the processing procedure shown in Fig. 10A and 10B, customers who use support target devices having the same or similar configuration as the support target device used by the new inquiring customer as similar customers.

[0146] In practice, when the similar customer extraction unit 42 is called by the document improvement proposal creation unit 41, it starts the similar customer extraction process shown in Figures 10A and 10B, and first obtains the email address of the new inquiring customer who is the sender of the new inquiry email from the data of the new inquiry email (S60).

[0147] Next, the similar customer extraction unit 42 reads out the predefined value of the threshold value δ from the definition file 30 (FIG. 1) stored in the file server 6 (FIG. 1) (S61). Furthermore, the similar customer extraction unit 42 extracts the software names and versions of all software installed in the supported devices used by the new inquiring customer from the system configuration database 27 (FIG. 3) (S62).

[0148] Specifically, the similar customer extraction unit 42 searches for records in which the email address of the new inquiring customer acquired in step S60 is stored in the contact column 27C (FIG. 3) from among the records in the system configuration database 27. Then, the similar customer extraction unit 42 reads out the software names of all software stored in the configuration software group column 27E of the records found by this search, and the versions of these software stored in the version column 27F (FIG. 3).

[0149] Next, the similar customer extraction unit 42 selects one record from the records of the system configuration database 27 other than the record corresponding to the new inquiring customer and which has not been processed after step S64 (S63), and reads out the data of that record (hereinafter referred to as the selected record) (S64).

[0150] Furthermore, based on the data read in step S64, the similar customer extraction unit 42 counts the number of pieces of software (the number of overlapping pieces of software) installed in both the supported device corresponding to the selected record and the supported device used by the new inquiring customer (S65).

[0151] Specifically, the similar customer extraction unit 42 counts the number of overlaps of such software by comparing each piece of software stored in the constituent software group column 27E (Figure 3) of the selected record with the software group acquired in step S62.

[0152] Thereafter, the similar customer extraction unit 42 sets the value of a parameter called a system score ("system_score") to the number of overlaps counted in step S65 (S66).

[0153] Next, the similar customer extraction unit 42 selects one piece of software from among the software implemented in both the supported device corresponding to the selected record and the supported device used by the new inquiring customer (hereinafter referred to as duplicate software) (S67).

[0154] In addition, for the duplicate software selected in step S67 (hereinafter referred to as selected duplicate software), the similar customer extraction unit 42 extracts the version of the selected duplicate software installed in the supported equipment used by the new inquiring customer and the version of the selected duplicate software installed in the supported equipment corresponding to the selected record (S68).

[0155] Specifically, for the version of the selected duplicate software installed in the supported device used by the new inquiring customer, the similar customer extraction unit 42 extracts the version of the selected duplicate software from the versions of each software acquired in step S62. Also, for the version of the selected duplicate software installed in the selected supported device, the similar customer extraction unit 42 extracts the version of the selected duplicate software from the data of the selected record read in step S64.

[0156] Next, the similar customer extraction unit 42 counts the number of versions of the selected overlapping software registered in the system configuration database 27 (FIG. 3) (S69) and calculates the number of versions by the following formula:

number

[0157] Specifically, the similar customer extraction unit 42 calculates a version score ("ver_score") by dividing the absolute value of the result obtained by subtracting the version of the selected duplicate software installed in the supported equipment corresponding to the selected record from the version of the selected duplicate software installed in the supported equipment used by the new inquiring customer by the total number of versions of the selected duplicate software counted in step S69 (S70).

[0158] Furthermore, the similar customer extraction unit 42 uses the following formula

number

[0159] If the similar customer extraction unit 42 obtains a negative result in this determination, it returns to step S67, and thereafter repeats the processes of steps S67 to S72 while sequentially switching the overlapping software selected in step S67 to other overlapping software that has not been processed in steps S68 to S71. Through this repeated process, the final system score of each overlapping software is calculated.

[0160] Then, when the similar customer extraction unit 42 obtains a positive result in step S72 by completing the processing of steps S68 to S71 for all overlapping software, it judges whether the final system score value is equal to or greater than the threshold value δ obtained in step S61 (S73).If the similar customer extraction unit 42 obtains a negative result in this judgment, it proceeds to step S75.

[0161] On the other hand, if the similar customer extraction unit 42 obtains a positive result in the judgment in step S73, it adds and registers the information of the selected record at that time and the final system score calculated for the selected record in step S71 to the similar customer list information 53 (Figure 7) (S74).

[0162] Specifically, the similar customer extraction unit 42 adds a new record to the similar customer list information 53, and stores the information stored in the customer column 27B, contact column 27C, component software group column 27E, version column 27F, or parameter column 27G of the selected record described above with reference to Fig. 3 in the customer column 53B, contact column 53C, component software group column 53E, version column 53F, and parameter column 53G of the record. In addition, the similar customer extraction unit 42 stores the final system score calculated in step S71 in the system score column 53D of the new record in the similar customer list information 53.

[0163] Thereafter, the similar customer extraction unit 42 judges whether or not the processes of steps S64 to S74 have been executed for all records in the system configuration database 27 other than the records corresponding to the new inquiring customers (S75).

[0164] If the similar customer extraction unit 42 obtains a negative result in this determination, it returns to step S63, and thereafter repeats the processes of steps S63 to S74 while sequentially switching the customer selected in step S63 to records that have not been processed in steps S64 to S74 and that satisfy the above-mentioned conditions.

[0165] Then, when the similar customer extraction unit 42 eventually obtains a positive result in step S75 by completing the processing of steps S64 to S74 for all records other than the record corresponding to the new inquiring customer, it ends this similar customer extraction processing.

[0166] (3-5) Customer-specific notification priority determination process Figures 11A and 11B show the flow of the customer-specific notification priority determination process executed by the customer-specific notification priority determination unit 43 called by the similar customer extraction unit 42 in step S18 of the document improvement support process described above with reference to Figure 6. The customer-specific notification priority determination unit 43 determines the notification priority for each similar customer according to the processing procedure shown in Figures 11A and 11B.

[0167] In practice, when the customer-specific notification priority determination unit 43 is called by the similar customer extraction unit 42, it starts the customer-specific notification priority determination process shown in Figures 11A and 11B, and first reads out the new inquiry email stored in memory 11 and obtains the email address of the new inquiring customer that is the sender of the new inquiry email (S80).

[0168] Next, the customer-specific notification priority determination unit 43 reads (S81) the document name ("document_name") included in the document improvement proposal information 52 created by the document improvement proposal creation unit 41 and stored in the memory 11 in step S14 of the document improvement support process described above with reference to Fig. 6. The document with this document name is the document that is the subject of improvement this time.

[0169] Furthermore, the customer-specific notification priority determination unit 43 searches through the records in the document database 28 (FIG. 4) for records in which the document name read in step S81 is stored in the document name column 28B (FIG. 4), and extracts the software name stored in the software name column 28D of the hit record (S82). The software with this software name is the software related to the document that is the subject of this improvement (hereinafter, this will be referred to as the target software).

[0170] Then, the customer-specific notification priority determination unit 43 extracts the version of the target software implemented in the supported device used by the new inquiring customer and the setting values ​​of various parameters set for that target software from the system configuration database 27 (Figure 3) (S83).

[0171] Specifically, the customer-specific notification priority determination unit 43 searches in the system configuration database 27 for records in which the software name extracted in step S82 is stored in sub-column 27H (FIG. 3) of the configuration software group column 27E (FIG. 3) and the email address of the new inquiring customer obtained in step S80 is stored in sub-column 27H of the contact column 27C (FIG. 3).

[0172] Then, the customer-specific notification priority determination unit 43 extracts the version of the target software from the corresponding sub-column 27H of the version column 27F (FIG. 3) in the record hit by this search, and obtains the setting values ​​of each parameter stored in the corresponding sub-column 27H of the parameter column 27G of that record. The version and setting values ​​of each parameter obtained in this way are the version of the target software installed in the supported device used by the new inquiring customer, and the setting values ​​of each parameter set for that target software.

[0173] Next, the customer-specific notification priority determination unit 43 reads information for one record that has not been processed after step S85 from the similar customer list information 53 (FIG. 7) created by the similar customer extraction unit 42 and stored in the memory 11 (S84).

[0174] Furthermore, the customer-specific notification priority determination unit 43 determines whether or not the software name of the target software extracted in step S82 is stored in the constituent software group column 53E (FIG. 7) in the read record (S85).

[0175] A negative result in this determination means that the target software is not installed in the supported device used by the similar customer corresponding to the record read in step S84. Thus, at this time, the customer-specific notification priority determination unit 43 proceeds to step S93.

[0176] On the other hand, if an affirmative result is obtained in the determination in step S85, it means that the target software is installed in the supported device used by the similar customer corresponding to the record read in step S84.

[0177] Thus, at this time, the customer-specific notification priority determination unit 43 determines whether the version of the target software installed in the supported device used by the corresponding similar customer and the setting values ​​of each parameter set for that target software match with the version of the target software installed in the supported device used by the new inquiring customer and the setting values ​​of each parameter set for that target software (S86).

[0178] Specifically, the customer-specific notification priority determination unit 43 determines whether the version of the target software stored in the version column 53F (Figure 7) of the record whose information was read in step S84 and the setting values ​​of each parameter set for the target software stored in the parameter column 53G (Figure 7) of that record match the version and the setting values ​​of each parameter obtained in step S83, respectively.

[0179] If the customer-specific notification priority determination unit 43 obtains a positive result in this determination, it stores the information "important" in the notification priority column 53H (Figure 7) of the record whose information was read in step S84 in the similar customer list information 53 stored in the memory 11 (S87), and then proceeds to step S93.

[0180] On the other hand, if the customer-specific notification priority determination unit 43 obtains a negative result in the determination in step S86, it determines whether or not the version of the target software stored in the version column 53F of the record whose information was read in step S84 does not match the version of the target software installed in the supported equipment used by the new inquiring customer obtained in step S83 (S88).

[0181] If the customer-specific notification priority determination unit 43 obtains a negative result in this determination, it proceeds to step S90. If the customer-specific notification priority determination unit 43 obtains a positive result in the determination in step S88, it stores information called "version" in the difference column 53I (FIG. 7) of the record whose information was read in step S84 in the similar customer list information 53 stored in the memory 11 (S89).

[0182] Next, the customer-specific notification priority determination unit 43 determines whether or not at least one of the setting values ​​of each parameter set for the target software stored in the parameter column 53G of the record whose information was read in step S84 is inconsistent with the setting value of each parameter set for the target software in the supported device used by the new inquiring customer obtained in step S83 (S90).

[0183] If the customer-specific notification priority determination unit 43 obtains a negative result in this determination, it proceeds to step S92. If the customer-specific notification priority determination unit 43 obtains a positive result in the determination in step S90, it stores information "parameter" in the difference column 53I of the record whose information was read in step S84 in the similar customer list information 53 stored in memory 11 (S91), and stores information "Caution" in the notification priority column 53H of that record (S92).

[0184] Furthermore, the customer-specific notification priority determination unit 43 judges (S93) whether or not the processing from step S85 onwards has been completed for all records in the similar customer list information 53. If the customer-specific notification priority determination unit 43 obtains a negative result in this judgment, it returns to step S84, and thereafter repeats the processing from step S84 to step S93 while sequentially switching the record selected in step S84 to other records that have not been processed from step S85 onwards.

[0185] Then, when the customer-specific notification priority determination unit 43 eventually obtains a positive result in step S93 by completing the processing from step S85 onwards for all records in the similar customer list information 53, it ends this customer-specific notification priority determination processing.

[0186] (3-6) Creating notification content for each customer Figures 12A and 12B show the flow of the customer-specific notification content creation process executed by the customer-specific notification content creation unit 44 called by the customer-specific notification priority determination unit 43 in step S20 of the document improvement support process described above with reference to Figure 6. The customer-specific notification content creation unit 44 creates notification contents to be sent to target similar customers for each notification priority according to the processing procedure shown in Figures 12A and 12B.

[0187] In practice, when the customer-specific notification content creation unit 44 is called by the customer-specific notification priority determination unit 43, it starts the customer-specific notification content creation process shown in Figures 12A and 12B, and first reads from the memory 11 the content of the body of the new inquiry email, the email address of the new inquiring customer, and response content information 51 (Figure 6), which is the content of the body of the reply email from the automatic response unit 23 to that new inquiry email (S100).

[0188] Next, the customer-specific notification content creation unit 44 reads information for one record that has not been processed after step S102 from the similar customer list information 53 (FIG. 7) stored in the memory 11 (S101), and further extracts information related to one piece of software that has not been processed after step S103 (information on the version, parameter settings, notification priority, and differences) from the information of the record read in step S101 (S102).

[0189] Next, the customer-specific notification content creation unit 44 determines whether or not the notification priority is "important" for the software about which related information was extracted in step S102 (hereinafter, this will be referred to as "extracted software") (S103).

[0190] If the customer-specific notification content creation unit 44 obtains a positive result in this determination, it sends the contents of the body of the new inquiry email read in step S100 and the contents of the response content information 51 to the email address of the corresponding similar customer stored in the contact information field 53C (FIG. 7) of the record read in step S101 (S104).The customer-specific notification content creation unit 44 then proceeds to step S113.

[0191] On the other hand, if the determination in step S103 is negative, the customer-specific notification content creation unit 44 determines whether the notification priority of the extracted software is "Caution" based on the information extracted in step S102 (S105). If the determination in step S103 is negative, the customer-specific notification content creation unit 44 proceeds to step S113.

[0192] On the other hand, if the determination in step S105 is positive, the customer-specific notification content creation unit 44 determines whether the difference is a "version" based on the information extracted in step S102 (S106). If the determination in step S105 is negative, the customer-specific notification content creation unit 44 proceeds to step S109.

[0193] On the other hand, if the customer-specific notification content creating unit 44 obtains a positive result in the determination in step S106, it acquires the version of the extracted software installed in the device to be supported of the new inquiring customer (S107).

[0194] Specifically, the customer-specific notification content creation unit 44 refers to the system configuration database 27 (Figure 3), identifies a record in which the email address of the new inquiring customer read in step S100 is stored in the contact column 27C (Figure 3) and the software name of the extracted software is entered in the configuration software group column 27E, and obtains the version by reading out the version stored in the version column 27F of that record.

[0195] Next, the customer-specific notification content creation unit 44 adds a sentence to the notification content for the corresponding similar customer (the similar customer corresponding to the record whose information was read in step S101) to warn about the change in the version of the extracted software (S108).

[0196] Specifically, the customer-specific notification content creation unit 44 adds a sentence such as "If the version of XX is changed to XX, the following event may occur" to the notification content for the similar customer. Note that "XX" here is the software name of the extraction software, and "XX" is the version of the extraction software installed in the supported device of the new inquiring customer.

[0197] Next, based on the information extracted in step S102, the customer-specific notification content creation unit 44 judges whether or not a "parameter" is stored in the difference column 53I corresponding to the extracted software in the similar customer list information 53 (S109). If the customer-specific notification content creation unit 44 obtains a negative result in this judgment, it proceeds to step S111.

[0198] On the other hand, if the customer-specific notification content creation unit 44 obtains a positive result in the determination in step S109, it acquires the setting values ​​of each parameter of the extraction software set in the supported device used by the new inquiring customer (S110).

[0199] Specifically, the customer-specific notification content creation unit 44 refers to the system configuration database 27 (Figure 3), identifies a record in which the email address of the new inquiring customer read in step S100 is stored in the contact column 27C (Figure 3) and the software name of the extracted software is entered in the configuration software group column 27E, and obtains the setting values ​​of each parameter stored in the parameter column 27G of that record.

[0200] Next, the customer-specific notification content creation unit 44 adds a sentence to the notification content for the corresponding similar customer (the similar customer corresponding to the record whose information was read in step S101) to warn about changing the parameter setting values ​​of the extraction software (S111).

[0201] Specifically, the customer-specific notification content creation unit 44 adds the following sentence to the notification content for the similar customer: "If the setting value of the XX parameter of XX is changed to △△, the following event may occur." Here, "XX" is the software name of the extraction software, and "XX" is the name of the corresponding parameter among the parameters of the extraction software implemented in the support target device of the new inquiring customer. Also, "△△" is the setting value of that parameter in the support target device used by the new inquiring customer.

[0202] Next, the customer-specific notification content creation unit 44 sends the contents of the body of the new inquiry email and the contents of the response content information 51 read in step S100, and the text added in step S108 and / or step S111, to the email address of the corresponding similar customer stored in the contact field 53C (Figure 7) of the record read in step S101 (S112).

[0203] Thereafter, the customer-specific notification content creation unit 44 determines whether or not the processing of steps S103 to S112 has been completed for all software whose software names are registered in the configuration software group column 53E (FIG. 7) of the record whose information was read in step S101 (S113).

[0204] If the customer-specific notification content creation unit 44 obtains a negative result in this determination, it returns to step S102, and thereafter repeats the processes of steps S102 to S113 while sequentially switching the software selected in step S102 to other software that has not been processed in steps S103 and after.

[0205] Then, when the customer-specific notification content creation unit 44 obtains a positive result in step S113 by completing the processing of steps S103 to S112 for all software whose software names are registered in the configuration software group column 53E (FIG. 7) of the record whose information was read in step S101, it determines whether or not the processing of step S102 and subsequent steps has been completed for all records in the similar customer list information 53 (S114).

[0206] If the customer-specific notification content creation unit 44 obtains a negative result in this determination, it returns to step S101, and thereafter repeats the processing of steps S101 to S114 while sequentially switching the record of the similar customer list information 53 selected in step S101 to other records that have not been processed in steps S102 and onwards.

[0207] Then, when the customer-specific notification content creation unit 44 eventually obtains a positive result in step S114 by completing the processing in steps S102 and subsequent steps for all records in the similar customer list information 53, it ends the customer-specific notification content creation processing.

[0208] (3-7) Customer inquiry database creation process 13A and 13B show the flow of the inquiring customer database creation process executed by the inquiring customer database creation unit 45 (FIG. 1) among various processes executed by the automatic response server 5 in relation to the document improvement support function. The inquiring customer database creation unit 45 creates the latest inquiring customer database 29 (FIG. 5) according to the processing procedure shown in FIG. 13A and FIG. 13B.

[0209] In practice, in this embodiment, the control unit 20 (FIG. 1) periodically calls the inquiring customer database creation unit 45. When called by the control unit 20, the inquiring customer database creation unit 45 starts the inquiring customer database creation process shown in FIGS. 13A and 13B.

[0210] Then, the inquiry customer database creation unit 45 first reads the threshold value ε and the threshold value ζ from the definition file 30 (FIG. 1) stored in the file server 6 (FIG. 1) (S120), and then reads all the sender addresses stored in the sender address column 26E (FIG. 2) of each record in the inquiry database 26 (FIG. 2) (S121).

[0211] Next, the inquiry customer database creation unit 45 executes a duplicate elimination process to eliminate duplicate sender addresses from all sender addresses read in step S121 (S122). In addition, the inquiry customer database creation unit 45 selects one sender address that has not been processed in step S124 or later from the sender addresses that have been subjected to the duplicate elimination process in step S122 (S123).

[0212] Then, the inquiry customer database creation unit 45 searches the system configuration database 27 (FIG. 3) using the sender address selected in step S123 (hereinafter referred to as the selected sender address) as a search key to extract the configuration software group of the supported device used by the customer corresponding to the selected sender address (S124).

[0213] Specifically, the inquiry customer database creation unit 45 searches for records in the system configuration database 27 in which the selected sender address is stored in the contact information column 27C (Figure 3), and extracts the software names of all software stored in the configuration software group column 27E (Figure 3) of that record.

[0214] Next, the inquiry customer database creation unit 45 acquires all documents related to each of the software whose software names have been extracted in step S124 from the document database 28 (FIG. 4) (S125).

[0215] Furthermore, the inquiring customer database creation unit 45 extracts, from the inquiry database 26, emails (inquiry mails or reply mails) in which the selected sender address is the sender address or the destination address (S126). Here, the inquiring customer database creation unit 45 extracts, from the records in the inquiry database 26, records in which the selected sender address is stored in the sender address column 26E (FIG. 2) or the destination address column 26F (FIG. 2).

[0216] Next, the inquiry customer database creation unit 45 resets (sets to "0") a counter (hereinafter referred to as the counter for counting the number of duplicates) to be used in step S134 described below (S127), and then selects one document that has not been processed in steps S129 and after from the documents obtained in step S125 (S128).

[0217] Next, the inquiry customer database creation unit 45 reads the text data of the document selected in step S128 (hereinafter, referred to as the selected document) from the document database 28 (S129). Based on the text data of the read selected document, the inquiry customer database creation unit 45 also extracts all important keywords from the selected document as an important keyword group RK using techniques such as morphological analysis and TF-IDF (S130).

[0218] Furthermore, the inquiry customer database creation unit 45 selects one email (record) that has not been processed after step S132 from among the emails (records) extracted in step S126 (S131), and extracts a group of important keywords MK from the body of the selected email (hereinafter referred to as the selected email) (S132).

[0219] Specifically, the inquiry customer database creation unit 45 extracts all important keywords from the body of the selected email stored in the body column 26H (Figure 2) of the record corresponding to the selected email in the inquiry database 26, using techniques such as morphological analysis and TF-IDF, as an important keyword group MK.

[0220] Next, the inquiry customer database creation unit 45 compares the important keyword group RK extracted in step S130 with the important keyword group MK extracted in step S132 to determine the number of overlapping important keywords (important keywords that exist in both the important keyword group RK and the important keyword group MK), and determines whether this number (hereinafter referred to as the overlap number) is greater than or equal to the threshold value ε obtained in step S120 (S133).

[0221] If the inquiry customer database creation unit 45 obtains a negative result in this determination, it proceeds to step S135. If the inquiry customer database creation unit 45 obtains a positive result in the determination in step S133, it increments the count value of the above-mentioned duplicate number counting counter by 1 (S134).

[0222] Furthermore, the inquiring customer database creation unit 45 judges whether or not the processes of steps S127 to S134 have been executed for all the emails (records) extracted in step S126 (S135). If the inquiring customer database creation unit 45 obtains a negative result in this judgment, it returns to step S131, and thereafter repeats the processes of steps S131 to S135 while sequentially switching the emails (records) selected in step S131 to other emails (records) not yet processed in steps S132 to S134.

[0223] Then, when the inquiry customer database creation unit 45 obtains a positive result in step S135 by completing the processing of steps S132 to S134 for all emails (records) extracted in step S126, it determines (S136) whether or not it has completed the processing of steps S129 to S135 for all documents obtained in step S125.

[0224] If the inquiry customer database creation unit 45 obtains a negative result in this determination, it returns to step S128, and thereafter repeats the processes of steps S128 to S136 while successively switching the document selected in step S128 to other documents not yet processed in steps S129 to S135.

[0225] Then, when the inquiry customer database creation unit 45 obtains a positive result in step S136 by completing the processing of steps S129 to S135 for all documents obtained in step S125, it determines whether the count value of the counter for counting the number of duplicates is equal to or greater than the threshold value ζ read out in step S120 (S137).

[0226] If the inquiry customer database creation unit 45 obtains a negative result in this determination, it registers the selected sender address at that time in the inquiry customer database 29 and sets the improvement flag to "False" (S138). More precisely, the inquiry customer database creation unit 45 secures an unused record in the inquiry customer database 29, stores the selected sender address in the contact information field 29B (Fig. 5) of that record, and sets the improvement flag stored in the improvement flag field 29C (Fig. 5) of that record to "False."

[0227] On the other hand, if the inquiry customer database creation unit 45 obtains a positive result in the determination in step S137, it registers the selected sender address at that time in the inquiry customer database 29 and sets the improvement flag to "True" (S139). More precisely, the inquiry customer database creation unit 45 secures an unused record in the inquiry customer database 29, stores the selected sender address in the contact information field 29B of the record, and sets the improvement flag stored in the improvement flag field 29C of the record to "True."

[0228] Next, the inquiring customer database creation unit 45 judges whether or not the processes of steps S123 to S139 have been executed for all sender addresses acquired in step S121 and after the duplicate elimination in step S122 have been completed (S140). If the inquiring customer database creation unit 45 obtains a negative result in this judgment, it returns to step S123, and thereafter repeats the processes of steps S123 to S140 while sequentially switching the sender address selected in step S123 to other sender addresses that have not been processed in the processes of steps S124 and after.

[0229] Then, when the inquiry customer database creation unit 45 eventually completes the processing of steps S124 to S139 for all sender addresses acquired in step S121 and after duplicate elimination in step S122, thereby obtaining a positive result in step S140, it terminates the inquiry customer database creation process.

[0230] (4) Effects of this embodiment As described above, in the automatic response system 1 of this embodiment, when the automatic response server 5 automatically responds to a new inquiry from a customer, if there have been α or more inquiries in the past that are identical or similar to the new inquiry, if the new inquiry is from a customer that has been in operation for β or more years and has made γ or more inquiries, or if the new inquiry is from a customer that has made an inquiry in the past that required document improvement, the automatic response server 5 determines that the corresponding document needs to be improved and presents the engineer with the reasons for this and a proposal for improving the document.

[0231] Therefore, according to the present automatic response system 1, it is possible to give engineers an opportunity to improve documents at the appropriate time, thereby reducing the number of cases that staff must deal with due to document defects, thereby reducing the maintenance costs of the automatic response system 1.

[0232] (5) Other embodiments In the above-mentioned embodiment, the control unit 20, the automatic application filing / registration unit 21, the similar case extraction unit 22, the automatic response unit 23, the inquiring customer database creation unit 25, and the document improvement necessity determination unit 40, the document improvement proposal creation unit 41, the similar customer extraction unit 42, the customer-specific notification priority determination unit 43, the customer-specific notification content creation unit 44, and the inquiring customer database creation unit 45 constituting the inquiry information notification unit 24 are arranged in one automatic response server 5. However, the present invention is not limited to this, and these may be arranged in a distributed manner in a plurality of computer devices constituting a distributed computing system.

[0233] In the above embodiment, the document improvement necessity determination unit 40 determines that the corresponding document needs improvement when any one of the following determination conditions is satisfied: (A) there have been α or more inquiries in the past that are identical or similar to the new inquiry, (B) the new inquiry is from a customer who has been in operation for β or more years and has made γ or more inquiries, or (C) the new inquiry is from a customer who has made an inquiry in the past that needs document improvement. However, the present invention is not limited to this, and the determination condition (C) may be omitted, and the corresponding document may be determined to need improvement when the determination condition (A) or (B) is satisfied. Also, other determination conditions may be applied in addition to or instead of the determination condition (C).

[0234] Furthermore, in the above embodiment, the document improvement proposal creation unit 41 has been described as switching between three levels of improvement proposal content to be presented to an engineer based on the value of the "score" calculated by the above-mentioned formula (2). However, the present invention is not limited to this, and the content of the improvement proposal may be switched between two levels or four or more levels based on the value of the "score". [Industrial Applicability]

[0235] The present invention can be widely applied to information processing devices of various configurations that automatically respond with a solution to a customer's inquiry about the device. [Explanation of symbols]

[0236] 1...document improvement support system, 3, 4...client, 5...automatic response server, 6...file server, 10...CPU, 20...control unit, 21...automatic invoicing and registration unit, 22...similar case extraction unit, 23...automatic response unit, 24...inquiry information notification unit, 25...inquiry customer database creation unit, 26...inquiry database, 27...system configuration database, 28...document database, 29...inquiry customer database, 30...definition file, 40...document improvement necessity determination unit, 41...document improvement proposal creation unit, 42...similar customer extraction unit, 43...customer-specific notification priority determination unit, 44...customer-specific notification content creation unit, 45...inquiry customer database creation unit, 50...similar case information, 51...response content information, 52...document improvement proposal information, 53...similar customer list information, 54...customer-specific notification content information.

Claims

1. 1. An information processing device which stores the contents of past inquiries from customers regarding a device and solutions to the inquiries as past cases, extracts the past case having the same or similar contents as the new inquiry when a new inquiry is received, and presents the solution of the extracted past case as a solution to the new inquiry, a similar case extraction unit that extracts the past cases that are identical or similar to the new inquiry; an improvement necessity determination unit that, when the similar case extraction unit is able to extract the past case that is identical or similar to the new inquiry, determines whether or not a document provided to the customer in association with the device needs to be improved, and, when it is determined that the document needs to be improved, requests a person in charge to improve the document; Equipped with The improvement necessity determination unit is The need for improving the document is determined based on at least one of the following information: the number of inquiries that are the same as or similar to the new inquiry, the number of inquiries made by the inquiry source of the new inquiry and the number of years in operation of the equipment of the inquiry source, and whether or not the inquiry source of the new inquiry is a customer who has made an inquiry in the past that requires improvement of the document.

23. An information processing apparatus comprising:

2. an improvement plan creation unit that creates an improvement plan for the document when the improvement necessity determination unit determines that the document needs improvement; The improvement plan creation unit Creating the improvement plan according to the degree of similarity between the contents of the new inquiry and the solution to the new inquiry and the contents of the document.

2. The information processing apparatus according to claim 1,

3. The improvement plan creation unit Calculating a degree of similarity between each content of the new inquiry and the solution to the new inquiry and the content of the document as a similarity; If the calculated similarity is equal to or greater than a first threshold, the improvement plan is created to highlight the content of the corresponding portion of the document; If the calculated similarity is less than the first threshold and equal to or greater than the second threshold, creating the improvement proposal to rewrite the relevant portion of the document to make it easier to understand; If the calculated similarity is less than the second threshold, the improvement plan is created to add a necessary explanation to the corresponding portion of the document.

3. The information processing apparatus according to claim 2.

4. a similar customer extraction unit that extracts other customers who use equipment having the same or similar configuration as the equipment used by the customer who has made the new inquiry as similar customers; a customer-specific notification content creation unit that creates and transmits a warning notification corresponding to the configuration of the device used by each of the similar customers extracted by the similar customer extraction unit; The information processing apparatus according to claim 1 , further comprising:

5. The similar customer extraction unit, For each of the customers other than the customer who made the new inquiry, a score is calculated that is defined so that the greater the number of overlapping software installed in the device used by the customer and the device used by the customer who made the new inquiry and the closer the versions of the overlapping software are, the higher the score is, and similar customers are extracted from among the customers other than the customer who made the new inquiry based on the calculated score.

5. The information processing apparatus according to claim 4.

6. a customer-specific notification priority determination unit that determines, for each of the similar customers extracted by the similar customer extraction unit, a similarity between a configuration of the device used by the similar customer and a configuration of the device used by the customer who has made the new inquiry, and determines a priority for sending the warning notification based on a result of the determination; The customer-specific notification content creation unit Based on the result of the determination by the customer-specific notification priority determination unit, the warning notification having contents according to the priority is generated for each of the similar customers.

5. The information processing apparatus according to claim 4.

7. An information processing method executed by an information processing device, which stores contents of past inquiries from customers regarding a device and solutions to the inquiries as past cases, and when a new inquiry is received, extracts the past case having contents identical or similar to the new inquiry, and presents the solution of the extracted past case as a solution to the new inquiry, A first step of extracting the past cases that are identical or similar to the new inquiry; a second step of determining whether or not a document provided to the customer in association with the device needs to be improved when the past case that is the same as or similar to the new inquiry can be extracted, and requesting a person in charge to improve the document when it is determined that the document needs to be improved; Equipped with In the second step, the information processing device The need for improving the document is determined based on at least one of the following information: the number of inquiries that are the same as or similar to the new inquiry, the number of inquiries made by the inquiry source of the new inquiry and the number of years in operation of the equipment of the inquiry source, and whether or not the inquiry source of the new inquiry is a customer who has made an inquiry in the past that requires improvement of the document.

23. An information processing method comprising:

8. a third step of generating an improvement plan for the document when the information processing device determines in the second step that the document needs to be improved; In the third step, the information processing device Creating the improvement plan according to the degree of similarity between the contents of the new inquiry and the solution to the new inquiry and the contents of the document.

8. The information processing method according to claim 7,

9. In the third step, the information processing device Calculating a degree of similarity between each content of the new inquiry and the solution to the new inquiry and the content of the document as a similarity; When the calculated similarity is equal to or greater than a first threshold, the improvement plan is created to make the description of the relevant portion in the document more prominent. When the calculated similarity is less than the first threshold and equal to or greater than a second threshold, the improvement plan is created to rewrite the relevant portion in the document to make it easier to understand. When the calculated similarity is less than the second threshold, the improvement plan is created to add a necessary explanation to the relevant portion in the document.

9. The information processing method according to claim 8,

10. a third step of extracting, as similar customers, other customers who use equipment having the same or similar configuration as the equipment used by the customer who made the new inquiry; a fourth step of creating and transmitting a warning notice to each of the extracted similar customers according to the configuration of the device used by the similar customer; 8. The information processing method according to claim 7, further comprising:

11. In the third step, the information processing device For each of the customers other than the customer who made the new inquiry, a score is calculated that is defined so that the greater the number of overlapping software installed in the device used by the customer and the device used by the customer who made the new inquiry and the closer the versions of the overlapping software are, the higher the score is, and similar customers are extracted from among the customers other than the customer who made the new inquiry based on the calculated score.

11. The information processing method according to claim 10.

12. In the fourth step, the information processing device determining a degree of similarity between a configuration of the device used by each of the extracted similar customers and a configuration of the device used by the customer who made the new inquiry, and determining a priority for sending the warning notification based on the result of the determination; Based on the result of the determination, the warning notice having contents according to the priority is generated for each of the similar customers.

11. The information processing method according to claim 10.

Citation Information

Patent Citations

  • Document preparing device

    JP1998049529A

  • Contact center system

    JP2004192521A

  • Improvement situation feedback system

    JP2004310321A

  • Information service provision method

    JP2007149114A

  • Information search system and method

    JP2019139746A