Document management system and document management method using ai
The AI-driven document management system addresses the inefficiency in customer service operations by automating summary generation across various channels, centralizing workflows, and reducing after-call work time, thereby improving business efficiency.
Patent Information
- Application Number
- PCT/JP2024/019684
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-30
- Filing Date
- 2024-05-29
- Publication Date
- 2024-12-05
AI Technical Summary
Existing document management systems and methods do not effectively improve business efficiency in customer service operations across various channels like telephone, chat, and email using AI-driven summary generation.
A document management system and method utilizing AI for summary generation, which includes a document information acquisition module, a machine learning model for generating summaries, and an output processing module to streamline customer service workflows by centralizing operations and reducing after-call work time.
The system significantly reduces the time required for after-call work and enhances business efficiency by automatically generating summaries, making it applicable to multiple customer service channels beyond traditional call centers.
Smart Images

Figure JP2024019684_05122024_PF_FP_ABST
Abstract
Description
Document management system and document management method using AI
[0001] [RELATED APPLICATIONS] This application claims priority to Japanese Patent Application No. 2023-088817, filed on May 30, 2023, entitled "Document Management System and Document Management Method Using AI," the disclosure of which is incorporated herein by reference in its entirety. The present disclosure relates to a document management system and a document management method using AI.
[0002] Japanese Patent Application Laid-Open No. 2013-190991 (Patent Document 1) is a background technology in this technical field. This publication states that "in a support method for improving the efficiency of call center response monitoring (listening to and evaluating speech), the method is capable of improving the efficiency of monitoring work even for a set of responses that deviate from the dialogue flow of a talk script. The voice dialogue summarization device extracts multiple utterances from a set of dialogue texts, and extracts, from the extracted multiple utterances, those that are connected to each other by an implication relationship as utterance clusters" (see abstract).
[0003] JP 2013-190991 A
[0004] The aforementioned Patent Document 1 describes a mechanism for a voice dialogue summarization device that extracts multiple utterances from a collection of dialogue texts in a call center and extracts utterance clusters. However, this patent document does not consider a method for improving work efficiency using AI-based summary generation for various types of customer service work using telephone, chat, email, etc. Therefore, the present disclosure provides a mechanism for improving work efficiency using AI-based summary generation.
[0005] In order to solve the above problem, for example, the configuration described in the claims is adopted. The present application includes a plurality of means for solving the above problem, and one example thereof is characterized by having a document information acquisition means for acquiring first document information, a summary generation means for inputting the first document information acquired by the document information acquisition means into a trained machine learning model that has performed machine learning to generate a summary using document information as input, and thereby acquiring a summary for the first document information from the machine learning model, and an output processing means for outputting the summary acquired by the summary generation means.
[0006] According to the present disclosure, it is possible to provide a system for improving business efficiency by using AI-based summary generation. Problems, configurations, and effects other than those described above will become clear from the following description of the embodiment.
[0007] FIG. 1 is an example of a configuration diagram of the entire customer interaction management system 1. FIG. 2 is an example of a hardware configuration diagram of the management server 101. FIG. 3 is an example of a hardware configuration diagram of the user terminal 102. FIG. 4 is an example of a hardware configuration diagram of the administrator terminal 103. FIG. 5 is an example of a system configuration diagram of the summary generation module 214. FIG. 6 is an example of customer information 600. FIG. 7 is an example of category information 700. FIG. 8 is an example of document information 800. FIG. 9 is an example of summary information 900. FIG. 10 is an example of memo information 1000. FIG. 11 is an example of interaction record information 1100. FIG. 12 is an example of FAQ information 1200. FIG. 13 is an example of input assistance information 1300. FIG. 14 is an example of a customer interaction processing flow 1400. FIG. 15 is an example of a document information acquisition processing flow 1500. FIG. 16 is an example of a summary generation processing flow 1600. FIG. 17 shows an example of a detailed summary generation process flow 1700. FIG. 18 shows an example of a category-specific summary generation process flow 1800. FIG. 19 shows another example of a category-specific summary generation process flow 1900. FIG. 20 shows an example of a highlighting process flow 2000. FIG. 21 shows an example of a customer information search screen 2100. FIG. 22 shows an example of a customer information inquiry screen 2200. FIG. 23 shows an example of a category selection input screen 2300. FIG. 24 shows an example of a summary display screen 2400. FIG. 25 shows an example of another summary display screen 2500. FIG. 26 shows an example of a summary setting screen 2600. FIG. 27 shows an example of a reference information call setting screen 2700. FIG. 28 shows an example of a feedback reception screen 2800. FIG. 29 shows an example of a work status display screen 2900. FIG. 30 shows an example of a work status detail display screen 3000. FIG. 31 shows an example of another feedback reception screen 3100. Fig. 32 shows an example of a summary generation instruction screen 3200. Fig. 33 shows an example of a summary display screen 3300. Fig. 34 shows an example of a customer response processing flow 3400 using category information.
[0008] An embodiment will be described below with reference to the accompanying drawings. This embodiment facilitates customer service operations for users, such as operators working at call centers. For example, a system is provided that allows a call center operator to easily execute a series of processes after a customer interaction, entering information about the call content into a system and reflecting the information in a CRM (Customer Relationship Management) system. This type of post-customer interaction processing is called after-call work (ACW). Examples of ACW that can be handled by the system include entering and selecting various categories of information about the inquiry content, creating a summary of the call content, and storing the response status and response method in the CRM system in association with customer information determined during the call.
[0009] By using this system, it is possible to centralize customer service operations and dramatically reduce ACW time, thereby improving the efficiency of call center operations. This system is not limited to telephone response operations at call centers, but can also be applied to customer service operations using chat, email, etc., and can also be applied to any document management operations.
[0010] The customer response management system 1 includes multiple user terminals 102 and multiple administrator terminals 103, each connected to a management server 101 via a network. The network may be wired or wireless, and each terminal can send and receive information via the network. The user terminals 102 are terminals used by users such as operators who handle customer service at a call center. The administrator terminal 103 is a terminal that manages the multiple user terminals 102 and the work of numerous operators operating at the call center. The administrator terminal 103 may also have a function for managing the management server 101, such as changing the settings of the management server 101.
[0011] Each terminal and the management server 101 of the customer interaction management system 1 may be, for example, a mobile terminal such as a smartphone, tablet, mobile phone, or personal digital assistant (PDA), or a wearable terminal such as glasses / goggles, a wristwatch, or clothing. They may also be stationary or mobile computers, or servers located on the cloud or a network. In terms of functionality, they may also be virtual reality (VR) terminals, augmented reality (AR) terminals, or mixed reality (MR) terminals. Alternatively, they may be a combination of multiple of these terminals. For example, a combination of one smartphone and one wearable terminal may logically function as a single terminal. Other information processing terminals may also be used.
[0012] Each terminal and management server 101 of the customer response management system 1 includes a processor that executes an operating system, applications, programs, etc., a main storage device such as RAM (Random Access Memory), an auxiliary storage device such as an IC card, hard disk drive, SSD (Solid State Drive), or flash memory, a communication control unit such as a network card, wireless communication module, or mobile communication module, input devices such as a touch panel, keyboard, mouse, voice input, or input based on motion detection captured by a camera unit, and an output device such as a monitor or display. Note that the output device may also be a device or terminal that transmits information to be output to an external monitor, display, printer, or other device.
[0013] The main memory stores various programs and applications (modules), and the processor executes these programs and applications to realize the various functional elements of the overall system. These modules may be implemented in hardware, such as by integration. Each module may be an independent program or application, or may be implemented as a subprogram or function within a single integrated program or application.
[0014] In this specification, each module is described as the entity (subject) that performs processing, but in reality, the processing is performed by a processor that processes various programs, applications, etc. (modules). Various databases (DBs) are stored in the auxiliary storage device. A "database" is a functional element (storage unit) that stores data sets so that they can accommodate arbitrary data manipulation (e.g., extraction, addition, deletion, overwriting, etc.) from the processor or an external computer. The method of implementing the database is not limited, and may be, for example, a database management system, a file management system, spreadsheet software, or a text file such as XML or JSON.
[0015] 2 is an example of a hardware configuration diagram of the management server 101. The management server 101 is configured as a server deployed on the cloud, for example. However, the management server 101 may also be configured as an on-premise system at the same location as the user terminal 102 and / or the administrator terminal 103. Alternatively, the management server 101 may be configured as an integrated unit with the user terminal 102 and / or the administrator terminal 103. The main memory device 201 stores programs and applications such as a customer response processing module 211, a category selection module 212, a document information acquisition module 213, a summary generation module 214, a memo information acquisition module 215, a highlighting module 216, a feedback module 217, and a data search module 218. The processor 203 executes these programs and applications to realize the respective functional elements of the management server 101.
[0016] The customer response processing module 211 provides a function for managing the overall customer response processing performed by users (operators, etc.). While this embodiment will be described using telephone response in a call center as an example, the present invention is not limited to this, and can also be applied to customer requests received via chat, email, inquiry forms, message sending and receiving services, etc. Furthermore, the customer response processing module 211 can display document information acquired by the document information acquisition module 213 and a summary generated by the summary generation module 214 in a corresponding relationship on a single screen. In addition to displaying two panels on a single screen, the module can also display document information and summaries by switching between them using tabs, for example. In this case, the document information and the summary can also be displayed in a corresponding relationship. Alternatively, the module can simply display the contents of one panel.
[0017] The category selection module 212 identifies the category to which the customer inquiry belongs. Examples of categories include the following: Target business categories: finance, insurance, e-commerce, public energy (electricity, gas, water, etc.), tourism, bridal, etc. Target product categories: Wedding venues, reception parties, honeymoons, etc. for the bridal category; Property insurance, life insurance, education insurance, automobile insurance, fire insurance, etc. for the insurance category; Content categories: application / procedures, questions (inquiries), consultations, requests, etc. Hierarchical category information design enables flexible response to various types of customer service processes. While category information is stored in the category information 700, it can be registered in advance or can be freely written and registered by users, etc. The category selection module 212 may also be configured to newly register character strings extracted from terms used in dialogue text as category information. Category information may also be meta information. Meta information for documents, etc. can be treated as category information.
[0018] The document information acquisition module 213 acquires the content of conversations with customers as text data and stores it as document information. For example, document information is generated by analyzing the voice of a telephone conversation with a customer and converting it into text. In the case of interactions with customers via chat, email, inquiry forms, message sending and receiving services, etc., the content of the text sent and received is stored as document information. In addition, any document information can be acquired, not limited to the content of conversations with customers.
[0019] The summary generation module 214 inputs document information and generates a summary of it. When AI is used, the summary generation module 214 generates a summary for the input document information by inputting the document information into a trained machine learning model that has been trained to generate summaries using the document information as input. At this time, by inputting category information along with the document information, the accuracy of the generated summary can be improved and the summary content can be adjusted. The memo information acquisition module 215 acquires and stores notes written by the user while serving a customer. Note that the notes may be a screen where blank text is entered, or may be configured to be entered into some predetermined format. Any document information may be input. Note that the notes may be acquired before or after serving a customer.
[0020] The highlighting module 216 makes customer service easier by highlighting portions of document information, summaries, and memo information that meet certain conditions. Highlighting can be achieved, for example, by bolding, underlining, changing the color, italicizing, shading, changing the font, blinking, highlighting by surrounding the text, changing the background color, or overlaying a color layer on top of the text. Highlighting can be achieved by highlighting text that stands out or is distinguishable from other text. Highlighting can also be achieved by displaying the highlighted text on a separate screen or window.
[0021] The feedback module 217 accepts user evaluations and corrections of the generated summary. The feedback module 217 may also accept corrections to document information and memo information. In this embodiment, these evaluations and corrections are collectively referred to as user feedback. The data search module 218 searches for customer information 600, FAQ information 1200, input assistance information 1300, etc. when responding to customers.
[0022] The auxiliary storage device 202 includes various databases for storing various types of information. Each database stores, for example, customer information 600, category information 700, document information 800, summary information 900, memo information 1000, and response record information 1100. Details of each type of information will be described later.
[0023] 3 is an example of a hardware configuration diagram of the user terminal 102. The user terminal 102 is configured as a terminal such as a smartphone, tablet, laptop PC, or desktop PC. The main memory device 301 stores programs and applications such as a management server cooperation module 310, and the processor 303 executes these programs and applications to realize the various functional elements of the user terminal 102.
[0024] The management server cooperation module 310 cooperates with the customer service processing module 211 of the management server 101 to execute the customer service processing provided by the management server 101. The management server cooperation module 310 displays various screens such as Figures 21 to 33 provided by the management server 101 on a display which is one of the output devices 205.
[0025] The customer response processing module 211 of the management server 101 provides, for example, a customer response function from the cloud, and the management server cooperation module 310 displays on a display screen information sent from the management server 101. However, the configuration may be such that all of the program and application functions provided by the management server 101 are implemented on the user terminal 102, and the user terminal 102 independently realizes the customer response function, or the configuration may be such that some of the program and application functions are implemented on the user terminal 102, and the user terminal 102 and the management server 101 cooperate to realize the customer response function.
[0026] The auxiliary storage device 302 stores, for example, user terminal management information 320. The user terminal management information 320 stores, for example, information that identifies an operator who is a user using the user terminal 102. Furthermore, in a configuration in which some of the functions of the management server 101 are implemented on the user terminal 102 side, the user terminal management information 320 may store information similar to the various information stored in the management server 101.
[0027] 4 is an example of a hardware configuration diagram of the administrator terminal 103. The administrator terminal 103 is configured as a terminal such as a smartphone, tablet, laptop PC, or desktop PC. The main memory device 401 stores programs and applications such as a user terminal management module 410 and a management server management module 411, and the processor 403 executes these programs and applications to realize the various functional elements of the administrator terminal 103.
[0028] The user terminal management module 410 collectively manages multiple user terminals 102 operating in the call center. The user terminal management module 410 displays screens that allow monitoring of the efficiency of ACW work, such as those shown in Figures 29 and 30. A list of each call (interaction) can be checked and filtered, and input for each interaction can be displayed. The information displayed on these screens is obtained by the user terminal management module 410 in cooperation with the customer interaction processing module 211 of the management server 101.
[0029] The user terminal management module 410 also calculates and displays distributions and statistics of the time spent in ACW for each important attribute, such as by time (month, week, etc.), by user, by category of customer service content, etc. The user terminal management module 410 also displays a log of ACW work for each user, as shown in FIG.
[0030] The management server management module 411 performs various settings for the management server 101. For example, the management server management module 411 can perform settings for summary generation using a summary setting screen 2600 in FIG. 26 . Note that it is also possible to implement a configuration in which all of the program and application functions provided by the management server 101 are implemented on the administrator terminal 103, and the administrator terminal 103 realizes the customer response function. It is also possible to implement a configuration in which some of the program and application functions are implemented on the administrator terminal 103, and the user terminal 102 and the administrator terminal 103 work together to realize the customer response function.
[0031] The auxiliary storage device 402 stores, for example, administrator terminal information 420. The administrator terminal information 420 stores, for example, information identifying the administrator who uses the administrator terminal 103, information identifying the user terminals 102 and each user that are being managed, etc. Furthermore, in a configuration in which some of the functions of the management server 101 are implemented on the administrator terminal 103 side, the administrator terminal information 420 may store information similar to the various information stored in the management server 101.
[0032] FIG. 5 is an example of a system configuration diagram of the summary generation module 214. There are various possible methods for implementing the summary generation module 214, but FIG. 5 shows an example of a specific implementation using a machine learning module. An API (Application Programming Interface) module 501 receives a summary generation request from the user terminal 102, acquires document information, and starts the summary generation process. The API module 501 also transmits the generated summary to the user terminal 102 via a web server module 505. The API module 501 may be implemented as an independent API server.
[0033] The message queuing module 502 temporarily stores document information in a queue and sequentially inputs it to the ML module 503. The ML (Machine Learning) module acquires the document information accumulated in the queue, generates a summary, and passes it to the queue. The ML module 503 may be implemented as an independent ML server. The worker module 504 associates the summary received from the message queuing module 502 with a document ID and stores it in the DB 511. The web server module 505 displays the summary results, etc., as a UI (User Interface). The storage device 512 stores multiple machine learning trained models, which can be read into the ML module 503 to change the content of the generated summary. A detailed summary generation process flow by the summary generation module 214 in FIG. 5 will be described later with reference to FIG. 17.
[0034] 6 to 13 show various types of information stored in the management server 101. FIG. 6 is an example of customer information 600. The customer information 600 stores information about customers who have called the call center. The customer information 600 includes information to identify the customer, such as a user ID 601, a user display name, a name, a telephone number, an email address, a date of birth, a gender, and an address, and values such as those shown in the sample value 620 are entered for each item 610. The user ID 601 is a unique ID, such as an automatically generated hash value, and is a primary key referenced by other information. The user display name (sometimes called a user title or user display name) is a user name displayed on the screen of the user terminal 102, the administrator terminal 103, etc.
[0035] 7 shows an example of category information 700. The category information 700 is information selected according to the content of a customer response, and stores information indicating the content of the customer response. The category information 700 includes a category ID 701, a subcategory ID 702, a subcategory ID2 703, a subcategory content 704, a subcategory content 705, and a subcategory content2 706, and values such as those shown in sample values 720 are entered for each item 710. The category ID 701, the subcategory ID 702, and the subcategory ID2 703 are IDs that, either alone or in combination, uniquely identify the content of a customer response.
[0036] 7, Bridal001 of category ID 701 indicates wedding-related category content 704, Ceremony001 of subcategory ID 702 indicates wedding venue-related subcategory content 705, and Inquiry002 of subcategory ID2 703 indicates inquiry-related subcategory content 2 706. In this example, for example, under the wedding category, in addition to the wedding venue-related subcategory, there are subcategories such as reception party and honeymoon. Furthermore, under the wedding venue-related subcategory, in addition to inquiries, there is a sub-subcategory called applications.
[0037] In another example, under the category "insurance," there are subcategories such as property insurance, life insurance, education insurance, automobile insurance, and fire insurance. Furthermore, under the subcategory of property insurance, there are sub-subcategories such as inquiries, applications, and insurance claims. In this way, by designing category information hierarchically, it becomes possible to flexibly respond to various types of customer service processes.
[0038] In the example of Figure 7, the categories have a three-level hierarchy, but they may have two levels, or four or more levels. It is also possible to have no hierarchy and manage all categories at only one level of category ID. In addition, while the hierarchy has been described with weddings, wedding venue-related, and inquiries in the descending order, this order may be reversed, for example, so that inquiries, weddings, and wedding venue-related descend in that order.
[0039] 8 shows an example of document information 800. The document information 800 stores, for example, information about a document acquired through customer service. The document information 800 includes a document ID 801, a document display 802, a recording date and time 803, and a document information storage URL 804, and values such as those shown in sample values 820 are input for each item 810.
[0040] The document ID 801 is identification information that uniquely identifies a stored document. The document display 802 (sometimes referred to as the document title 802 or document display name 802) stores a description, such as the title of the document, that is displayed on the user terminal 102 or the administrator terminal 103. The recording date and time 803 indicates the date and time the document was stored in the storage device. In the example of Figure 8, "20220401141201" indicates "April 1, 2022, 14:12:01." If the document is updated, the document update date and time and the last update date and time may also be stored. The document information storage URL 804 indicates the location where the actual document is stored. Note that a path to the data may be specified in the form of a URL, or a path to a file in internal storage on the server may be specified. Furthermore, if the data size is not large, the information may be written directly.
[0041] In this embodiment, the document information stored in the document information storage URL 804 of the document information 800 may also be referred to as "document information." For example, the text data stored in the document information storage URL 804 may also be referred to as "document information." Furthermore, the text data before being stored in the document information 800 may also be referred to as "document information."
[0042] The text data stored here may be, for example, a dialogue (dialogue text) that is a text version of the content of the conversation exchanged between a customer and an operator during customer service. In addition, in the case of interactions with customers via chat, email, an inquiry form, a message sending / receiving service, etc., the content of the text sent and received is stored as document information. In addition, any document information relating to the content of the customer service may be acquired and stored. Furthermore, when this embodiment is broadly applied to operations other than call centers, any document information may be acquired and stored. Customer service is referred to by various names, such as support, inquiries, customer success, CX (Customer Experience), and contact center.
[0043] 9 shows an example of summary information 900. The summary information 900 stores the summary generated by the summary generation module 214. The summary information 900 includes a summary ID 901, a summary display 902, a document ID 903, a recording date and time 904, a last update date and time 905, and a summary information storage URL 906, and values such as those shown in sample values 920 are input for each item 910.
[0044] The summary ID 901 is identification information that uniquely identifies the stored summary. The summary display 902 (sometimes called the summary title 902 or summary display name 902) stores a description such as the title of the summary that is displayed on the user terminal 102 or the administrator terminal 103. The document ID 903 stores the value of the document ID 801 of the document information 800, and identifies which document information 800 the summary corresponds to. The recording date and time 904 indicates the date and time the summary was stored in the storage device.
[0045] The last update date and time 905 indicates the date and time of the most recent update after the summary was modified. It is also possible to store not only the last update date and time but also the update date and time of the summary if it has been updated multiple times. The summary information storage URL 906 indicates the location where the actual summary is stored. It is also possible to specify a path to the data in the form of a URL, or to specify a path to a file in storage within the server. Furthermore, if the data size is not large, it is also possible to write the information directly. In this embodiment, if a summary identified by a single summary ID 901 has been updated multiple times, the update history of that summary or all versions of the summary can be stored in association with the same summary ID 901. Alternatively, a new summary ID can be assigned each time the summary is updated and stored.
[0046] FIG. 10 shows an example of memo information 1000. The memo information 1000 includes a memo ID 1001, a memo display 1002, a document ID 1003, a recording date and time 1004, a last update date and time 1005, and a memo information storage URL 1006. Values such as those shown in the sample value 1020 are entered for each item 1010. The memo ID 1001 is identification information that uniquely identifies the stored memo information. The memo display 1002 (also referred to as a memo title 1002 or a memo display name 1002) stores a description, such as the title of the memo information displayed on the user terminal 102 or the administrator terminal 103. The document ID 1003 stores the value of the document ID 802 of the document information 800, and identifies which document information 800 the memo information corresponds to. The recording date and time 1004 indicates the date and time the memo information was stored in the storage device.
[0047] The last update date and time 1005 indicates the date and time of the last update after the memo information was modified. It is also possible to store not only the last update date and time but also the update date and time of the memo information when it is updated multiple times. The memo information storage URL 1006 indicates the location where the actual memo information is stored. In this embodiment, when memo information identified by one memo ID 1001 is updated multiple times, the update history can be stored in association with the same memo ID 1001, and a new memo ID can also be assigned each time an update is made.
[0048] 11 is an example of interaction record information 1100. The interaction content when a user interacts with a customer is stored as an interaction record. The interaction record information 1100 includes an interaction ID 1101, an interaction start date and time 1102, an interaction end date and time 1103, a person in charge 1104, a customer ID 1105, a category ID 1106, a summary ID 1107, an evaluation score 1108, and a customer input data storage URL 1109. Values such as those shown as sample values 1120 are entered for each item 1110.
[0049] The response ID 1101 is identification information that uniquely identifies a customer response. The response start date / time 1102 and the response end date / time 1103 indicate the date / time when the customer response started and ended. The person in charge 1104 stores information that identifies the user (operator) who responded to the customer. The customer ID stores information that identifies the customer who was served.
[0050] The category ID 1106 is information indicating the content of the customer response, and stores the category ID 701 of the category information 700 in Fig. 7. The subcategory ID and subcategory ID2 below the category ID 1106 store the information of the subcategory ID 702 and subcategory ID2 703 in Fig. 7, respectively. The summary ID 1107 stores the information of the summary ID 901 in Fig. 9, and stores the type of summary generated based on the customer response in association with the information.
[0051] The evaluation score 1108 stores the evaluation score from the user or administrator for the generated summary. The evaluation score 1108 may be stored in the summary information 900 instead of the response record information 1100. The customer input data storage URL 1109 stores information that identifies or explains the content of the customer response and the content of the processing input as a result. For example, in the case of a response to an application from a customer, information about the resulting application is written in the customer input data storage URL 1109.
[0052] 12 is an example of FAQ information 1200. The FAQ information 1200 stores a list of frequently asked questions and their answers. The FAQ information 1200 includes an FAQ ID 1201, an overview, a category ID 1202, a subcategory ID, and an FAQ data storage URL 1203, and values such as those shown in sample values 1220 are entered for each item 1210.
[0053] FAQ ID 1201 is identification information that uniquely identifies an FAQ. The overview stores information indicating the content, such as the title of the FAQ. FAQ data storage URL 1203 indicates the location where the actual FAQ content is stored. In addition to specifying the path to the data in the form of a URL, it is also possible to specify a path to a file in storage within the server. Furthermore, if the data size is not large, the information may be written directly. For example, FAQ data can be stored as text information with questions and answers in separate columns.
[0054] Category ID 1202 is information indicating the content of the customer service, and stores category ID 701 of category information 700 in Fig. 7. Note that subcategory ID and subcategory ID2 below category ID 1202 store information on subcategory ID 702 and subcategory ID2 703 in Fig. 7, respectively. By specifying and storing customer service categories for FAQ information 1200 in this way, FAQ information related to the customer service category can be obtained and displayed on user terminal 102, etc.
[0055] 13 is an example of auxiliary input information 1300. The auxiliary input information 1300 stores auxiliary input information for retrieving input contents to be input for each customer when serving a customer. The auxiliary input information 1300 has an auxiliary input information ID 1301, an overview, a category ID 1302, a subcategory ID, and an input content storage URL 1303, and values such as those shown in sample values 1320 are entered for each item 1310. The auxiliary input information ID 1301 is identification information that uniquely identifies auxiliary input information such as a form to be filled in. The overview stores information indicating the content, such as the title of the auxiliary input information.
[0056] The category ID 1302 is information indicating the content of the customer service, and the category ID 701 of the category information 700 in Fig. 7 is stored in the category ID and subcategory ID2 below the category ID 1302. The information of the subcategory ID 702 and subcategory ID2 703 in Fig. 7 is stored in the subcategory ID and subcategory ID2 below the category ID 1302, respectively. By specifying and storing the customer service category for the auxiliary input information 1300 in this way, auxiliary input information related to the customer service category can be obtained and displayed on the user terminal 102, etc.
[0057] FIG. 14 shows an example of a customer response processing flow 1400. The customer response processing module 211 starts customer response (step 1410). Specifically, when an operator (user) receives a call and starts customer response, the operator may be notified of this, or the operator may input that the operator has started customer response. The document information acquisition module 213 converts the contents of the call into text during the call and acquires the document information (step 1420). The converted text of the call is sometimes called dialogue text. The document information acquisition process will be described later.
[0058] Next, the customer response processing module 211 identifies the customer on the other end of the call (step 1430). Specifically, the customer response processing module 211 calls the data search module 218, which then displays the contents of the customer search field 2140 on the customer information search screen 2100 of FIG. 21. The operator searches for customer information while viewing the contents of the call and the dialogue text, and if a matching customer is found, identifies that customer as the call partner. The case where a matching customer is not found and a new customer is registered will be described later.
[0059] 21 shows an example of a customer information search screen 2100. The dialogue text field 2110 displays the contents of the call converted into text by the document information acquisition process 1420 in near real time. The memo field 2120 is where the operator enters memos when responding to a customer. The memo information entered and acquired here is sometimes called second document information.
[0060] The customer search field 2140 is a display area where the operator can input keywords to search for a customer and search for customer information registered in the customer information 600. The customer response processing module 211 obtains information on the telephone number of the customer who is answering the call and displays it in the incoming call information 2141. Note that the configuration may also be such that the operator inputs the telephone number displayed on the telephone into the incoming call information 2141.
[0061] Possible methods for specifying customer information include, for example: the customer response processing module 211 automatically inputs customer information from a telephone number; the operator inputs customer information by searching and selecting; or the operator inputs the information manually.
[0062] On the other hand, the customer response processing module 211 may be configured to assist in input based on the content of the dialogue text. For example, the customer response processing module 211 may input, into the name field of the search criteria 2142, name information 2111 that is likely to be a customer among information 2111 and 2112 about personal names in the dialogue displayed in the dialogue text field 2110. For example, the name information 2111 that appears first among the words uttered by the customer may be input into the name field. Note that the customer name heard by the operator during the dialogue may also be input into the name field of the search criteria 2142. Note that the customer response processing module 211 may also extract named entities from the dialogue text and input customer information by inference therefrom. For example, customer information may be inferred and input from the name, phone number, content, etc. displayed in the dialogue text using a rule-based or machine learning-based technique using a dictionary or the like.
[0063] In addition, the search criteria 2142 can also accept input of customer information such as name, date of birth, telephone number, and postal code. The highlighting module 216 of the management server 101 highlights parts of the dialogue text displayed in the dialogue text field 2110 that relate to attributes that are likely to be important for a customer search, such as name, date, time, telephone number, location, and contract number. Furthermore, if there is an item in the highlighted content that corresponds to the search criteria 2142, the highlighting module 216 inputs the name, telephone number, etc. that corresponds to the relevant part to complete the text. Conversely, if the operator enters information such as name and telephone number that he or she hears over the phone into the search criteria 2142, the highlighting module 216 can also highlight the corresponding part in the dialogue text.
[0064] When data search module 218 detects that the operator has selected search button 2143, it searches customer information 600 using incoming call information 2141 and the information entered in search conditions 2142 to check whether a corresponding record exists. If a corresponding record is found in customer information 600, data search module 218 displays customer information inquiry screen 2200 of FIG. 22 .
[0065] The customer information display field 2150 displays customer information registered in the customer information 600. If customer information is found by a search, the found customer information may be displayed and highlighted in the customer information display field 2150. If corresponding customer information is not found in the customer information 600, the customer response processing module 211 accepts selection of the new customer registration button 2160, inherits the information entered in the incoming call information 2141 and the search condition field 2142, and displays a new customer information registration screen. The new customer registration may also be provided with a function to complete the information from the dialogue text and a function to highlight the completed portion in the dialogue text. In this embodiment, the data search module 218 displays information obtained from the customer information 600 in the customer search field 2140. However, all of this display may be performed by the customer response processing module 211.
[0066] 22 shows an example of a customer information inquiry screen 2200. The data search module 218 displays customer information 2241 found by the search in a customer introduction field 2240. The highlighting module 216 highlights information corresponding to the customer information displayed in the displayed customer information 2241 in the dialogue text in the dialogue text field 2210. This configuration makes it easy to confirm where in the dialogue text the customer information found by the search or the customer information to be newly registered is written.
[0067] The customer response processing module 211 stores the customer ID corresponding to the customer on the other end of the call identified by search or new registration. Note that the data search module 218 may be configured to search the customer information 600 based on the incoming telephone number, the name extracted from the dialogue text, etc., to identify the customer, without the user having to perform a search. Alternatively, the data search module 218 may be configured to display candidate customer names, etc., and the customer may be identified by the user entering a confirmation.
[0068] 32 shows an example of a summary generation instruction screen 3200. When the customer response processing module 211 detects that the summary start button 3235 has been pressed, it transmits the dialogue text, which is document information, to the summary generation module 214, which then executes the summary generation process (step 1440). Specific details of the summary generation process will be described in detail with reference to FIG. 16 or 17.
[0069] FIG. 33 is an example of a summary display screen 3300. A summary generated by the summary generation module 214 is displayed. In the example of the displayed summary 3330, a customer name 3332 and a representative name 3331 are highlighted, and important matters 3333 in the generated summary are also highlighted. FIGS. 32 and 33 show examples in which the input document information and the summary generated from it are each displayed in a single panel, but these may also be displayed in two panels within a single window. Furthermore, FIGS. 32 and 33, each displaying only one panel, may also be displayed side by side or vertically within a single window.
[0070] The customer response processing module 211 associates the summary generated by the summary generation module 214 with the content of the customer response and stores them in the response record information 1100 (step 1450). Specifically, the customer response processing module 211 generates a response ID 1101 for each customer response, and associates the staff ID of the staff member who handled the response, the customer ID indicating the customer identified in step 1430, the summary ID indicating the summary generated in step 1440, the response start date and time, and the response end date and time in the response record information 1100.
[0071] When the feedback module 217 receives an evaluation of the summary, the evaluation information can also be stored. Furthermore, when an application or the like is generated as a result of the customer interaction, the input data for the procedure of the application or the like can be reflected in the CRM system or the like. In such a case, information that identifies or explains the content of the customer interaction and the content of the processing input as a result can be stored in the customer input data storage URL 1109 of the interaction record information 1100. When the operator makes corrections to the summary or updates the memo, the change history can be stored in the interaction record information 1100, the summary information 900, the memo information 1000, etc.
[0072] Figure 34 shows an example of a customer response processing flow 3400 using category information. Step 3440, which accepts the selection of category information, is added to the customer response processing flow 1400 in Figure 14. The processing of steps 3410 to 3430 by the customer response processing module 211 is similar to steps 1410 to 1430 in Figure 14, and therefore a description thereof will be omitted.
[0073] Next, the customer service processing module 211 displays a category selection input screen 2300 shown in FIG. 23 and accepts the operator's selection of a category for the service content (step 3440). Categories are classifications of the customer service content and are stored in the category information 700 shown in FIG. 7. Depending on the selected category, the customer service processing module 211 acquires and stores information identifying the category, such as the corresponding category ID, subcategory ID, and subcategory ID2. As described above, the category information can be freely entered by the user, or meta information can be registered, added, or used as category information.
[0074] FIG. 23 is an example of a category selection input screen 2300. An inquiry content input field 2350 displays fields for inputting or selecting categories, such as case classification, content, and response method. The caller classification is information for identifying the type of person the customer is, such as applicant, spouse, father, mother, policyholder, insured person, prospective customer, other party involved in the accident, dealer, financial institution, etc. The caller classification may also be treated as a category. Specifically, the customer response processing module 211 calls the category selection module 212, and the category selection module 212 displays the content of the inquiry content input 2350 on the category selection input screen 2300 of FIG. 23 .
[0075] 23, "bridal" is selected as the category, "ceremony hall" as the subcategory, and "inquiry" as the sub-subcategory according to the content of the customer service. When the set button 2352 is pressed after selecting a category, the customer service processing module 211 acquires and stores the category ID 701, subcategory ID 702, and subcategory ID2 703 shown in the example of FIG. 7 as IDs corresponding to these categories.
[0076] The display of "content" and "type of response method" in the inquiry content input 2350 in Fig. 23 changes depending on the selected "case category." For example, if "insurance" is selected as the case category, "automobile" and "fire" will be displayed, indicating automobile insurance and fire insurance, which are subcategories of this insurance.
[0077] The highlighting module 216 can highlight character strings such as keywords and sentences related to categories in the dialogue text in the dialogue text field 2310. For example, in the example of Fig. 23, keywords related to bridal, such as "bridal," "ceremony in Hawaii," and "bride and groom," are highlighted. By checking the highlighted character strings, the operator can easily determine which category the content of the customer interaction belongs to, which has the effect of facilitating category selection.
[0078] Furthermore, keywords and sentences related to categories registered in the category information 700 can be registered in advance, or related keywords and sentences can be acquired each time. The category selection module 212 analyzes keywords and sentences in the dialogue text to determine which category registered in the category information 700 the keywords and sentences are related to, and can automatically select input into the inquiry content input field 2350 according to the keywords and sentences displayed in the dialogue text. It is also possible to input dialogue text information and classify the entire text into categories on a rule-based basis using dictionary information or the like, or using a machine learning model.
[0079] Furthermore, by linking the category selection module 212 and the highlighting module 216, for example, if the operator selects "application" as the type of response method, keywords related to "application" are highlighted in the dialogue text, and if the dialogue method type is switched to "inquiry," keywords related to "inquiry" are highlighted in the dialogue text. Similarly, if the case category is changed from bridal to insurance, keywords related to insurance are highlighted.
[0080] The cooperation between the category selection module 212 and the highlighting module 216 enables various implementations, such as the following: The category selection module 212 can extract character strings, such as keywords and sentences, related to categories, which are classifications of customer service content, from the dialogue text, and automatically select a category based on this. It is also possible to input dialogue text information and classify the entire text into categories on a rule-based basis using dictionary information or the like, or using a machine learning model.
[0081] Furthermore, the highlighting module 216 can highlight character strings, such as keywords and sentences related to a category, from within the dialogue text, and the category selection module 212 can automatically select a related category based on the highlighted character strings. Furthermore, the highlighting module 216 can highlight character strings related to the selected category content in accordance with the selected category content. When the selected category content is changed, the highlighting module 216 can highlight character strings related to the changed category content. Furthermore, it is also possible to input dialogue text information and select a category from the entire text. Furthermore, it is also possible to input dialogue text information and select a category from the entire text based on rules using dictionary information or the like, or using a machine learning model.
[0082] In this embodiment, the category selection module 212 acquires the category information stored in the category information 700 and displays the content in the inquiry content input field 2350 based on this information, but this display may also be entirely performed by the customer response processing module 211.
[0083] When the customer response processing module 211 detects that the summary start button 2335 has been pressed after the category selection, it transmits the dialogue text, which is document information, to the summary generation module 214, and the summary generation module 214 executes the summary generation process (step 3450). Specific details of the summary generation process will be described in detail with reference to FIG. 16 or 17. Note that category selection is not essential, and the summary generation process may be executed even if no category is selected. Alternatively, the category selection module 212 may automatically select a category based on the content of the document information in the background, without displaying the category selection input screen 2300 as shown in FIG. 23.
[0084] The customer response processing module 211 associates the summary generated by the summary generation module 214 with the content of the customer response and stores them in the response record information 1100 (step 3460). Specifically, the customer response processing module 211 generates a response ID 1101 for each customer response, and associates the staff ID of the staff member who handled the response, the customer ID indicating the customer identified in step 3430, the category ID indicating the category selected in step 3440, the summary ID indicating the summary generated in step 3450, the response start date and time, and the response end date and time in the response record information 1100.
[0085] When the feedback module 217 receives an evaluation of the summary, the evaluation information can also be stored. Furthermore, when an application or the like is generated as a result of the customer interaction, the input data for the procedure of the application or the like can be reflected in the CRM system or the like. In such a case, information that identifies or explains the content of the customer interaction and the content of the processing input as a result can be stored in the customer input data storage URL 1109 of the interaction record information 1100. When the operator makes corrections to the summary or updates the memo, the change history can be stored in the interaction record information 1100, the summary information 900, the memo information 1000, etc.
[0086] FIG. 15 shows an example of a document information acquisition process flow 1500. While this example assumes telephone customer service, the document information acquisition process is applicable to any situation in which document information can be acquired. The document information acquisition module 213 receives telephone voice input during telephone service (step 1510). The document information acquisition module 213 analyzes the voice information and converts it into text information (step 1520). The text version of the content of a conversation with a customer is called a conversation text. The document information acquisition module 213 displays the text information as a conversation text, for example, in the conversation text field 2110 of FIG. 21 (step 1530).
[0087] If there is no need to store the text information, the text information is discarded as is (No in step 1540); if there is a need to store it (Yes in step 1540), the document information acquisition module 213 generates a document ID, associates it with the text information, and stores it in the document information 800 (step 1550).
[0088] The text information stored in the document information 800 is not limited to dialogue text, but may include text content sent and received in response to interactions with customers via chat, email, inquiry forms, message sending and receiving services, etc. Any other document information related to the content of interactions with customers may be acquired and stored. Document information may also be acquired by linking with another service, by being copied and brought after a call, etc. Document information may also be acquired by linking with another service, by being copied and brought after a call, etc. When this embodiment is broadly applied to operations other than call centers, any document information may be acquired and stored.
[0089] 16 shows an example of a summary generation process flow 1600. The summary generation module 214 acquires document information (step 1610). Specifically, the summary generation module 214 acquires text information (referred to as document information) from the location referenced by the document information storage URL 804 of the document information 800. Alternatively, the summary generation module 214 may acquire the dialogue text acquired by the document information acquisition module 213, or may acquire some other document information.
[0090] The summary generation module 214 inputs the acquired document information (text information) into the trained machine learning model to generate a summary (step 1620). The summary generation module 214 associates the generated summary with the generated document ID and stores them in the summary information 900 (step 1630). Note that although the input for generating the summary is assumed to be the dialogue text or memos, it may also be configured to reference other FAQs referenced during the call or logs of other services used.
[0091] Figure 17 shows an example of a detailed summary generation process flow 1700. There are various ways to implement the summary generation module 214, but Figure 5 shows an example of a specific implementation using a machine learning module, and the summary generation process using this configuration will be described below. The API module 501 acquires text information (step 1705). The API module 501 generates a document ID and stores the document ID and text information in the document information 800 (step 1710). The API module sends a message including the document ID and text information to the message queuing module 502 (step 1715).
[0092] The ML module 503 inputs the message obtained from the message queuing module 502 into the machine learning model obtained from the storage device 512 to generate a summary (step 1720). The ML module 503 sends a pair of the document ID and the summary result to the message queuing module 502 (step 1725). The worker module 504 stores the document ID and the summary result in the summary information 900 (step 1735).
[0093] The API module 501 acquires the summary result (referred to as "summary information") stored in the summary information 900 (step 1740). The web server module 505 acquires the summary information from the API module 501 (step 1745). The web server module 505 outputs the summary information (step 1750). Note that a large-scale language model can be used as the machine learning model, which is a highly accurate generative model, but is not limited to this. For example, a machine learning model that generates summaries by character string extraction may also be used.
[0094] FIG. 18 shows an example of a category-specific summary generation processing flow 1800. In the category-specific summary generation processing flow 1800, when generating a summary, category information can be added as input to generate a summary appropriate for the category. The summary generation module 214 acquires document information (step 1810). The summary generation module 214 acquires category information acquired by the category selection module 212 (step 1820). The summary generation module 214 inputs the acquired document information (text information) and category information into a trained machine learning model to generate a summary related to the category information for the input document information (step 1830). The summary generation module 214 associates the generated summary with the generated document ID and stores it in the summary information 900 (step 1840).
[0095] The trained machine learning model can be a trained machine learning model that has been trained using category information indicating the category to which the document information belongs and multiple pieces of document information belonging to this category as training data. In the example of selecting category information in step 3440 of the customer service processing flow 3400 using category information in Figure 34, a category related to the dialogue text has already been selected. Therefore, by inputting the document information of the dialogue text along with the category information into this machine learning model, the summary generation module 214 can generate a summary optimized for this document information. In this processing flow, a machine learning model that has been trained on a set of category information and document information is used, and a single machine learning model can output a summary corresponding to the input category.
[0096] As mentioned above, pre-selection of a category is not essential, and the summary generation process may be executed even if no category is selected. Alternatively, a category may be automatically selected in the background by the category selection module 212 based on the content of the document information, without displaying the category selection input screen 2300 shown in Figure 23. Alternatively, instead of using the category information itself, the document information may be input into a machine learning model to determine the document category, and then a summary corresponding to that category may be generated.
[0097] 19 shows an example of another category-specific summary generation process flow 1900. In this process flow 1900, a summary appropriate for the category can be generated by switching the machine learning model to be used depending on the category information. The summary generation module 214 acquires document information (step 1910). The summary generation module 214 then acquires the category information acquired by the category selection module 212 (step 1920).
[0098] The summary generation module 214 selects a machine learning model corresponding to the category information (step 1930). For example, if a separate server is provided for each machine learning model, the summary generation module 214 transmits the document information to the machine learning server corresponding to the category information.
[0099] The summary generation module 214 generates a summary for the acquired document information (text information) by inputting the acquired document information to the machine learning model selected corresponding to the category information (step 1940). The summary generation module 214 associates the generated summary with the generated document ID and stores them in the summary information 900 (step 1950).
[0100] The trained machine learning model can be a trained machine learning model obtained by performing machine learning using multiple pieces of document information belonging to a specific category as training data, and a trained machine learning model can be prepared for each category. These trained machine learning models are stored in the storage device 512 in Figure 5, and depending on the selection of a machine learning model, the selected machine learning model is read from the storage device 512 to the ML module 503. Note that an ML module 503 may be prepared for each machine learning model optimized for a category, and the ML module 503 may be switched.
[0101] As explained in the selection of category information in step 3440 of the customer service processing flow 3400 in Figure 34, a category related to the dialogue text has already been selected. Therefore, by inputting the document information of the dialogue text into a machine learning model optimized for each category selected based on the category information, the summary generation module 214 can generate a summary optimized for this document information. In this processing flow, by switching between machine learning models trained for each category based on the category information, a summary corresponding to the input category can be output.
[0102] Note that the selection of a category in step 3440 of the customer response processing flow 3400 in Figure 34 is not essential, and the summary generation process may be executed even if no category is selected. Also, a configuration may be adopted in which a category is automatically selected in the background by the category selection module 212 based on the content of the document information, without displaying the category selection input screen 2300 as shown in Figure 23. In other words, this embodiment provides a mechanism for explicitly switching summaries based on category information of the call content.
[0103] 20 shows an example of a highlighting process flow 2000. The highlighting module 216 acquires text information (referred to as document information) from the document information 800 (step 2010). The highlighting module 216 acquires a summary (referred to as summary information) from the summary information 900 (step 2020). The highlighting module 216 acquires written memo information from the memo information 1000 (step 2030).
[0104] The highlighting module 216 performs analysis, such as natural language processing, on each piece of acquired information to extract one or more characteristic strings (step 2040). The highlighting module 216 highlights characteristic strings that satisfy various conditions described below, or strings that include these characteristic strings (step 2050). Note that a characteristic string refers to a word or a string that includes a word in the text data, and does not generally include strings that simply consist of particles. However, it is not limited to this and may include a concatenation of words that include particles, a string that includes particles, or any other broad string or sentence. Keywords are also one type of characteristic string. However, it is also possible to highlight only particles in parts that are likely to contain particle errors or parts where grammatical errors should be emphasized.
[0105] Highlighting can be achieved, for example, by highlighting portions of a document that meet certain conditions in bold, underlining, changing the color, italicizing, shading, changing the font, blinking, surrounding the text with a line, changing the background color, or overlaying a color layer on top of the text. Highlighting can be achieved by highlighting text in a way that makes it stand out or distinguishable from other text. In addition to highlighting text in some way, text can also be highlighted by displaying it in a separate frame. For example, a space containing only the date information you want to highlight can be displayed on a separate screen or window.
[0106] Although the highlighting module 216 has been described as acquiring information from the document information 800, summary information 900, and memo information 1000 stored in the storage device and performing highlighting, this is not limiting. For example, the document information acquisition module 213, the summary generation module 214, and the memo information acquisition module 215 may acquire information from the document information 800, summary information 900, and memo information 1000, respectively, and the highlighting module may perform the highlighting process. Alternatively, the customer service processing module 211 may acquire various information, and the highlighting module may perform the highlighting process.
[0107] Furthermore, the highlighting module 216 may be configured to highlight text information (dialogue text) acquired by the document information acquisition module 213, summary information generated by the summary generation module 214, and memo information acquired by the memo information acquisition module 215, instead of the document information 800, summary information 900, and memo information 1000 already stored in the storage device. Also, the customer response processing module 211 may be configured to perform highlighting in cooperation with the highlighting module 216, or the customer response processing module 211 itself may have the function of the highlighting module 216.
[0108] Fig. 24 shows an example of a summary display screen 2400. After selecting a category in Fig. 23, a summary generated by pressing the summary start button 2335 is displayed in the summary text field 2430. The summary displayed here was generated by inputting the document information displayed in the dialogue text field 2410 into a machine learning model that generates a summary corresponding to the selected category.
[0109] The customer response processing module 211 displays three types of information—first document information (dialogue text) acquired by the document information acquisition module 213, the summary acquired by the summary generation module 214, and second document information (memo information) acquired by the memo information acquisition module 215—in a dialogue text field 2410, a summary field 2430, and a memo field 2420, respectively, and outputs them in a corresponding manner using a three-panel display. Because the dialogue text 2410 is a text version of the audio information from a telephone conversation, there are some parts that are not accurately transcribed. However, even when dialogue text containing such errors is used as input, the summary generation module 214 can generate a correct summary 2430.
[0110] Note that not all three panels are necessary, and a configuration in which no dialogue, no summary, or no memo section is displayed is also possible. In this case, any two of the dialogue text section 2410, the summary section 2430, and the memo section 2420 can be displayed in association with each other. It is also possible to display only the dialogue, the summary, or the memo section independently. Furthermore, a three-panel or two-panel display is not necessarily required. For example, even if three types of information or any two types of information are displayed by switching between them using tabs or the like, this can also be considered a configuration in which each piece of information is displayed in association with each other. Note information may be used as the first document information, and dialogue text, chat, email, an inquiry form, a message sending / receiving service, or the like may be used as the second document information.
[0111] The highlighting module 216 highlights the name "Patent Jiro" shared between the dialogue text and the summary with an underline in 2411 and 2431. Furthermore, even though the names are completely different, the corresponding names "Eliza Hanako" 2412 in the dialogue text are highlighted with an underline in "Eliza Hanako" 2432 in the summary and "Eliza Hanako" 2422 in the memo section. Furthermore, the statement 2413 in the dialogue text that reads "Airfare and hotel costs for other persons are not included" and the corresponding statement 2433 that reads "Hotel costs for other persons are not included" are both highlighted in color. Furthermore, the date February 2, 2022, is shared between the dialogue text and the memo and is highlighted with an underline.
[0112] In this example, keywords in the dialogue text that are common to the summary are highlighted. In other words, the highlighting module 216 performs highlighting when a characteristic string corresponding to the first document information (dialogue text) and the summary exists, as a condition. The highlighting module 216 highlights a characteristic string that corresponds to the summary from among one or more characteristic strings included in the first document information (dialogue text). In addition, it is also possible to highlight a characteristic string that exists in common in the dialogue text, the summary, or a memo.
[0113] It is also possible to highlight corresponding keywords between the summary and the memo. That is, the highlighting module 216 sets the condition that a characteristic string corresponding to the second document information (memo information) and the summary exists as a condition, and performs highlighting when this condition is met. The highlighting module 216 highlights characteristic strings that correspond to the second document information (memo information) and the summary from among one or more characteristic strings included in at least one of the first document information (dialogue text), the second document information (memo information), and the summary. Furthermore, although the dialogue text and the summary may be expressed differently, it is possible to highlight those that correspond.
[0114] Pressing the settings button 2436 will take you to the summary generation settings screen of Fig. 26. By changing the settings here, such as changing the category selection, and then pressing the summary start button 2435 again, a summary will be generated again based on the changed settings. The summary generation settings screen may be configured to display a separate screen as shown in Fig. 26, or may be displayed on the summary output screen as shown in Figs. 23, 24, and 25.
[0115] 25 is an example of another summary display screen 2500. Different portions from those in FIG. 24 are highlighted. In this example, a statement 2511 in the dialogue text 2310, "Airfare for the bride and groom is included, but airfare and hotel costs for other persons are not included," and a corresponding statement 2531 in the summary sentence 2530, "Airfare for the bride and groom is included, but hotel costs for other persons are not included," are highlighted with an underline.
[0116] Keywords that are present in the memo but not in the summary can also be highlighted in the dialogue information or memo. That is, a condition is set that a characteristic string exists in the second document information (memo information) but does not exist in the summary, and the highlighting module 216 highlights characteristic strings that exist in the second document information (memo information) but do not exist in the summary, among one or more characteristic strings included in at least one of the first document information (dialogue text) and the second document information (memo information).
[0117] 25, a statement 2522 written in the memo field 2520, "I would like to add two more people," exists in the memo, but does not exist in the summary 2530. Therefore, the highlighting module 216 highlights the corresponding statements 2512 and 2522, "I would like to add two more people," in the memo and the dialogue text.
[0118] Keywords that are present in the summary but not in the dialogue information or notes can also be highlighted in the summary. That is, a feature string is present in the summary but not in the first document information (dialogue text) or the second document information (note information), and highlighting is performed when this condition is met. The highlighting module 216 highlights one or more feature strings contained in the summary that are present in the summary but not in the first document information (dialogue text) or the second document information (note information). In the example of FIG. 25 , the statement 2538 "decided to change the wedding plans" is present in the summary but not in the dialogue text or notes. Some machine learning models may generate sentences that do not exist in the input dialogue text. However, since sentences containing such strings may be inaccurate, highlighting sentences containing such strings can provide the user with an opportunity to correct them.
[0119] In the example of Figure 25, the statement 2533 "future policy direction" written in the summary statement column 2530 is present in the summary statement but not in the memo column 2520. Therefore, the highlighting module 216 highlights the statement 2513 "future policy direction" corresponding to this statement in the dialogue text. It is also possible to highlight character strings that are present in the summary and memo but not in the dialogue text. It is also possible to highlight character strings that are present in the summary and memo but not in the dialogue text. It is also possible to highlight other areas where the AI is not confident.
[0120] In other words, the highlighting module 216 sets a condition that the summaries generated by the summary generation module 214 may not be highly accurate, and highlights the summaries if this condition is met. For example, the determination can be made according to the probability distribution of the model that outputs the summaries. For example, highlighting is made when the probability distribution is constant and there is no significant difference in vocabulary selection. It is also possible to make the determination by creating a separate learning model for discrimination. The highlighting module 216 highlights the summaries generated by the summary generation module 214 that may not be highly accurate, from among the first document information (dialogue text), the second document information (memo information), or one or more characteristic character strings contained in the summaries.
[0121] Furthermore, the summary selection tag 2537 is a tab for comparing and displaying multiple summaries output by machine learning models. The display method may be to display at least two or more summaries in parallel, or to switch between them by selecting a button or in a pull-down menu. The display method may be to display at least two or more summaries in parallel, or to switch between them by selecting a button or in a pull-down menu. Each tab may display summaries generated by multiple machine learning models, or may display summaries generated by different categories.
[0122] The highlighting module 216 provides a mechanism for highlighting input and output to make it easier for people to review. The following are examples of mechanisms for making it easier for people to review input and output individually: Highlighting important points (dates, names, contract numbers, etc.). Highlighting points where the AI is unsure. Providing a mechanism for mutual reference between input and output to make review easier. Highlighting points that are in the summary but not similar to the input (dialogue text, notes, etc.). Note that similarity can be determined using the following methods: - Rule-based determination - Determining the degree of match or similarity in some unit, such as strings or words - Determining using dictionaries, etc. - Determining using machine learning technology - Similarity of distributed representations (vector representations) - Determining using machine learning models - Highlighting which parts of the input were used in the summary - Highlighting parts that are in the notes but not in the summary - Highlighting parts of the dialogue text that may be good to include in the summary Note that the following methods can be used to determine whether something is good to include. - Generate multiple summaries and determine the strings contained in those summaries as candidates. - Use a machine learning model / rule to determine the importance separately.
[0123] The following features are available for suggesting corrections after a review: - Suggest corrections at word or phrase level using a thesaurus or machine learning calculation of similarities between words and phrases. For example, the word "meeting" is highlighted in a summary, and correction suggestions such as "conference," "meeting," "MTG," and "meeting" are displayed as selectable corrections. Possible display formats include pop-up display of correction suggestions, displaying them in a separate tab or screen, or superimposing correction suggestions when the mouse cursor is moved over the highlighted text. Correction suggestions may also be displayed for sections where the AI is unsure. Alternatively, correction suggestions may be displayed when the user clicks or specifies a location. - Create conversion rules for commonly mistyped expressions at word or phrase level using past data or manually, and suggest them. - Suggest other candidates that the AI could have output at various levels, such as words, phrases, and sentences. For example, suggest the next most likely candidate. For example, among the candidates generated by the machine learning model, display the next most likely candidate as a correction suggestion. The correction point may be, for example, a point where the AI is unsure, or may be an arbitrary point designated by the user by clicking, etc. For example, in the summary example of FIG. 24 , for the expression "A quote for Hawaii was provided," a suggestion such as "A quote for Hawaii was mailed the other day" is presented. This correction suggestion may have a function for accepting feedback if the suggestion is incorrect or not very good. For example, a "check mark" indicating acceptance of the suggestion or a "cross mark" indicating rejection may be displayed below the correction suggestion, allowing the user to select a point. Alternatively, instead of accepting or rejecting the correction suggestion, a "like mark" or "dislike mark" may be displayed to accept a positive or negative evaluation from the user, allowing the user to evaluate the suggestion.
[0124] The following features are available for suggesting corrections after a review: - Suggest corrections at word or phrase level using a thesaurus or machine learning calculation of similarities between words and phrases. For example, the word "meeting" is highlighted in a summary, and correction suggestions such as "conference," "meeting," "MTG," and "meeting" are displayed as selectable corrections. Possible display formats include pop-up display of correction suggestions, displaying them in a separate tab or screen, or superimposing correction suggestions when the mouse cursor is moved over the highlighted text. Correction suggestions may also be displayed for sections where the AI is unsure. Alternatively, correction suggestions may be displayed when the user clicks or specifies a location. - Create conversion rules for commonly mistyped expressions at word or phrase level using past data or manually, and suggest them. - Suggest other candidates that the AI could have output at various levels, such as words, phrases, and sentences. For example, suggest the next most likely candidate. For example, among the candidates generated by the machine learning model, display the next most likely candidate as a correction suggestion. The correction portion may be, for example, a portion where the AI is unsure, or may be a portion designated by the user by clicking, for example. For example, in the summary example of FIG. 24 , for the expression “A quote for Hawaii was provided,” a suggestion such as “The quote for Hawaii was mailed the other day” may be presented. This correction suggestion may have a function for accepting feedback if the suggestion is incorrect or not very good. For example, a “check mark” indicating acceptance of the suggestion or a “cross mark” indicating rejection may be displayed below the correction suggestion, allowing the user to select a choice. Alternatively, instead of accepting or rejecting the correction suggestion, a “like mark” or a “dislike mark” may be displayed to accept a positive or negative evaluation of the suggestion, allowing the user to evaluate it. When a character string such as a word is corrected, the corrected character string may be re-input into the machine learning model, and the entire summary or the portion following the corrected character string may be re-generated by the machine learning model.
[0125] Examples of UIs (User Interfaces) that make it easy for humans to edit a single summary include: - Providing a UI that allows users to compare and edit the summary output by AI with the text of the edited summary. The feedback module 217 accepts edits to the summary generated by the summary generation module 214. The summaries before and after the edit can be displayed in a form that allows them to be compared. For example, the summaries before and after the edit can be displayed by switching between tabs or side by side. It is also possible to highlight edited sections by changing the color, etc. - Providing a function to use templates to edit the summary generated by the summary generation module 214. - Providing a function to compare and select the outputs of multiple summaries. Note that a configuration that allows users to compare and select the outputs of multiple machine learning models may also be used. Note that a configuration that allows users to compare and select the outputs of multiple machine learning models may also be used.
[0126] FIG. 26 is an example of a summary setting screen 2600. This screen is used to set up the output of summaries. The summary setting screen is displayed by selecting the summary setting button 2436 in FIG. 24 or the summary setting button 2536 in FIG. 25. The summary setting screen 2600 may be displayed as a separate screen, in parallel with other UIs, or in an overlay (full or partial) format. The example in FIG. 26 is a summary setting screen that is displayed when "insurance" has already been selected as the case category. Insurance type 2601 is displayed as a subcategory, and automobile insurance has been selected. Response method type 2602 is displayed as a sub-subcategory, and procedures have been selected.
[0127] Keyword specification 2603 specifies the terms to be used when generating the summary. The summary generation module 214 generates a summary that includes the terms specified here. Length adjustment 2604 adjusts the length of the summary to be generated; in this example, a summary of up to 180 characters is generated. Note that in this example, an upper limit for the length of the summary is specified, but a lower limit or a rough guideline for the length of the summary may also be specified. Selecting bullet points 2605 generates a summary in bullet points rather than long sentences.
[0128] When the date and time expression variation correction function 2606 is selected, a summary is generated in which the date and time are expressed in a consistent format, even if various formats for date and time are used in the dialogue text. Other functions for correcting variations in expression may also be provided. For example, a setting can be made to correct all occurrences of "today" to "today." The summary location specification 2607 allows the user to specify the location of the dialogue text, memos, etc. to be summarized. For example, clicking on the summary location specification 2607 displays a screen that accepts the specification of a portion of the dialogue text or memos, and the specified portion can be used as the summary target. After making these settings, by pressing the summary start button 2435, the summary generation module 214 generates a summary based on the specified content from the dialogue text.
[0129] As mentioned above, it provides various functions to change the summary output, as listed below. - Ability to control summary output by specifying category information for the inquiry content. - Ability to specify keywords and have those words reflected in the output. - Ability to adjust the length of the summary. - Ability to output the summary in bullet point format. - Ability to correct variations in the spelling of expressions. - Ability to specify the parts to summarize from input such as dialogue text or notes.
[0130] 27 is an example of a reference information call setting screen 2700. When the customer response processing module 211 accepts selection of the dialogue text button 2701 in the upper left of the screen, it displays the contents of the dialogue text. When the customer response processing module 211 accepts selection of the FAQ button 2702, it calls the data search module 218. The data search module 218 acquires the category information that has already been selected, and acquires the FAQ data specified by the corresponding category ID, etc. 1202 from the FAQ information 1200 in FIG. 12, from the location specified by the FAQ data storage URL 1203. In addition, it can also display FAQs used during the customer response, FAQs searched based on the dialogue text, etc.
[0131] As mentioned above, the category information can be specified by the operator, but it can also be automatically selected from the contents of the dialogue text. Therefore, when the operator selects the FAQ button 2702, related FAQ data associated with the dialogue text via the category information is obtained and displayed.
[0132] The external link 2703 is a button for calling other external systems. In addition to the dialogue text, FAQs referenced during the call, logs of other services used, such as web systems, and the like, can also be used as input for generating a summary. FIG. 28 shows an example of a feedback reception screen 2800. The feedback module 217 receives an evaluation 2801 for the generated summary. The feedback module 217 is sometimes referred to as an evaluation acquisition module. The feedback module 217 can also receive edits and corrections from the user for the summary displayed in the summary text field 2830. Note that instead of the feedback module 217, a separate summary editing module that receives edits for the summary may be provided.
[0133] The feedback module 217 accumulates summaries that have received high ratings, document information (dialogue text) that is the basis for the summaries that have received high ratings, and summaries corrected by the user, and can use these to retrain the machine learning model and improve it.In addition, it can also store summaries generated by the AI, summaries corrected by the user, evaluation scores, indicators of the difference between the user's summary and the AI's summary, etc. in a DB (database), for example.
[0134] When the user modifies text and saves it in a CRM or other system using an API, the text is reflected when the user performs that operation. When pasting the text separately using copy and paste, the text can be reflected when the user presses the copy button. Instead of a copy button, a button labeled "Update" or "Reflect" may be used. Updates may also be performed automatically in the background, without the need to explicitly press a button. In this way, the accuracy of the model can be improved by using a "human-in-the-loop" mechanism in which the user repeatedly evaluates and modifies the generated summary. Furthermore, for retraining, the difference score between the AI-generated summary and the manually modified summary, as well as information such as the extent of the changes made by the modifications, can also be used.
[0135] Model retraining can be done manually by a person on a regular basis, or a pipeline can be built for periodic retraining. Furthermore, as for the data to be trained, it is possible to retrain using all data in the database, or to prioritize training on data in the database that did not have very good scores. It is also possible to provide a system in which correction work for the results of model generation is carried out through BPO (Business Process Outsourcing), and the results are provided after improvement.
[0136] FIG. 31 shows another example of a feedback acceptance screen 3100. The user can output a template and start writing a new summary from there. For example, when the feedback module 217 receives a selection of the "Output Template" button 3131 displayed next to the summary 3130, it displays a template for optional input into the summary. The feedback module 217 can change the content of the template to be output based on the content of the category (procedural, automobile insurance, etc.) previously input.
[0137] Furthermore, if category information has not been input in advance, the feedback module 217 may display a template selection panel 3132 in accordance with the selection of template output and accept the selection of template selection within this panel. Also, categories may be structured hierarchically, and templates may be automatically selected based on the content of the dialogue, or highly likely candidates may be recommended. Alternatively, the summary generation module 214 may be configured to output templates.
[0138] FIG. 29 shows an example of a work status display screen 2900 displayed on the administrator terminal 103. The user terminal management module 410 of the administrator terminal 103 provides a function for monitoring the efficiency of ACW work for multiple users using each user terminal. The user terminal management module 410 displays the distribution and statistics of ACW time for each important attribute. For example, the ACW work log for each user can be displayed by sorting and filtering by attribute, such as time (month, week, etc.), user (operator), or customer interaction content category. In the example of FIG. 29, a list of each call (interaction) can be confirmed and filtered. Furthermore, when each item is selected, the user terminal management module 410 displays the work status details display screen 3000 shown in FIG. 30.
[0139] 30 is an example of a work status details display screen 3000. On the work status details display screen 3000, the dialogue text field 3010 displays text information about the dialogue text, and the memo field 3020 displays memos entered by the user. The inquiry content input field 3050 displays information about the selected category, and the selected category is confirmed by pressing a set button 3051. The summary field 3030 displays a summary generated from the dialogue text. If the category selection is corrected, for example, a summary can be generated again by pressing the start summary button 3031 again.
[0140] When the "Reflect in CRM" button 3060 is pressed, the entire content displayed here is finally stored in the CRM system. This embodiment makes it possible to streamline ACW work, which requires a large amount of man-hours. High-precision summaries are also automatically generated, allowing them to be instantly displayed on the Web. The generated summaries are displayed in an easy-to-edit UI, allowing the entire ACW process to be carried out using a single, simple customer interaction management system 1.
[0141] The present invention is not limited to the above-described embodiments, but includes various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.
[0142] Furthermore, the above-described configurations, functions, processing units, processing means, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. The above-described configurations, functions, etc. may also be implemented in software, with a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.
[0143] Furthermore, the control lines and information lines shown are those considered necessary for the explanation, and do not necessarily show all the control lines and information lines in the product. In reality, it can be considered that almost all components are interconnected. The above-mentioned embodiments disclose at least the configurations described in the claims.
[0144] 1...customer response management system, 102...user terminal, 103...administrator terminal, 211...customer response processing module, 212...category selection module, 213...document information acquisition module, 214...summary generation module, 215...memo information acquisition module, 216...highlighting module, 217...feedback module, 218...data search module
Claims
1. A document management system having: a document information acquisition means for acquiring first document information; a summary generation means for inputting the first document information acquired by the document information acquisition means into a trained machine learning model that has performed machine learning to generate a summary using document information as input, and acquiring a summary for the first document information from the machine learning model; and an output processing means for outputting the summary acquired by the summary generation means.
2. The document management system according to claim 1, further comprising a category specifying means for specifying a category to which the first document information belongs.
3. The document management system of claim 2, wherein the summary generation means selects a second machine learning model corresponding to the category and inputs the first document information into the second machine learning model, thereby obtaining a summary for the first document information from the second machine learning model.
4. The document management system according to claim 3, wherein the second machine learning model is a trained machine learning model that has undergone machine learning using multiple pieces of document information belonging to the category as training data.
5. The document management system described in claim 2, wherein the summary generation means inputs information indicating the category and the first document information, and obtains a summary related to the category for the first document information from the machine learning model.
6. The document management system described in claim 5, wherein the machine learning model is a trained machine learning model that has been subjected to machine learning using information indicating a category to which document information belongs and multiple pieces of document information belonging to the category as training data.
7. The document management system according to claim 2, wherein said summary generating means outputs a template of a summary related to said category.
8. The document management system according to any one of claims 1 to 7, further comprising summary editing means for accepting edits to said summary.
9. The document management system according to claim 8, wherein said summary generating means inputs a part of the edited summary and regenerates at least a part of the summary.
10. The document management system according to claim 8, wherein said summary editing means outputs a template for modifying said summary.
11. The document management system according to any one of claims 1 to 7, further comprising evaluation acquisition means for receiving an evaluation of the summary.
12. The document management system according to claim 11, wherein the machine learning model is retrained using document information from which the received highly-rated summaries have been generated.
13. A document management system according to any one of claims 1 to 7, wherein said summary generating means receives input of setting values for the summary to be generated, and changes the output content of the generated summary in accordance with said setting values.
14. A document management system according to any one of claims 1 to 7, wherein the summary generation means inputs second document information related to the first document acquired by the second information acquisition means and the first document information, and generates a summary.
15. The document management system according to any one of claims 1 to 7, further comprising highlighting means for highlighting one or more characteristic character strings included in the summary that satisfy a predetermined condition.
16. A document management system as described in claim 15, wherein the specified condition is the presence of information relating to attributes including at least one of date, time, name, location, telephone number, subscriber number, customer number, address, quantity, and address, and the highlighting means highlights a character string including information relating to the attribute among one or more of the characteristic character strings included in the summary.
17. A document management system as described in claim 15, wherein the specified condition is that the characteristic character string is present in the summary but not present in the first document information, and the highlighting means highlights, of the one or more characteristic character strings included in the summary, the characteristic character string that is present in the summary but not present in the first document information.
18. A document management system as described in claim 15, wherein the specified condition is that a corresponding characteristic string exists between the summary and the first document information, and the highlighting means highlights the characteristic string that corresponds between the summary and the first document information, out of one or more characteristic strings included in the summary.
19. A document management system as described in claim 15, wherein the specified condition is the presence of a character string in the summary generated by the summary generating means that may result in a low-accuracy summary, and the highlighting means highlights a character string from among one or more of the characteristic character strings included in the summary that may result in a low-accuracy summary in the summary generated by the summary generating means.
20. A document management system according to any one of claims 1 to 7, wherein the output processing means outputs the summary obtained by the summary generation means and the first document information obtained by the document information acquisition means in association with each other.
21. A document management system according to claim 20, further comprising highlighting means for highlighting one or more characteristic character strings contained in at least one of said summary and said first document information, said characteristic character strings satisfying a predetermined condition.
22. A document management system as described in claim 21, wherein the specified condition is the presence of information relating to attributes including at least one of date, time, name, location, address, quantity, and address, and the highlighting means highlights a character string including information relating to the attribute among one or more of the characteristic character strings included in at least one of the summary and the first document information.
23. A document management system as described in claim 21, wherein the specified condition is that the characteristic character string is present in the summary but not present in the first document information, and the highlighting means highlights, of the one or more characteristic character strings included in the summary, the characteristic character string that is present in the summary but not present in the first document information.
24. A document management system as described in claim 21, wherein the specified condition is that a corresponding characteristic character string exists between the summary and the first document information, and the highlighting means highlights the characteristic character string that corresponds between the summary and the first document information among one or more characteristic character strings contained in at least one of the summary and the first document information.
25. A document management system as described in claim 21, wherein the specified condition is the presence of a character string in the summary generated by the summary generating means that may result in a low-accuracy summary, and the highlighting means highlights a character string from among one or more of the characteristic character strings contained in at least one of the summary and the first document information that may result in a low-accuracy summary in the summary generated by the summary generating means.
26. A document management system according to any one of claims 1 to 7, wherein the output processing means outputs the summary obtained by the summary generation means and the second document information obtained by the second information acquisition means in association with each other.
27. A document management system according to claim 26, further comprising highlighting means for highlighting one or more characteristic character strings contained in at least one of said summary and said second document information, said characteristic character strings satisfying a predetermined condition.
28. A document management system as described in claim 27, wherein the specified condition is the presence of information relating to attributes including at least one of date, time, name, location, address, quantity, and address, and the highlighting means highlights a character string including information relating to the attribute among one or more of the characteristic character strings included in at least one of the summary and the second document information.
29. A document management system as described in claim 27, wherein the specified condition is that the characteristic character string is present in the second document information but not in the summary, and the highlighting means highlights, of the one or more characteristic character strings contained in the second document information, the characteristic character string that is present in the second document information but not in the summary.
30. A document management system as described in claim 27, wherein the specified condition is that a corresponding characteristic character string exists between the summary and the second document information, and the highlighting means highlights the characteristic character string that corresponds between the summary and the second document information, out of one or more characteristic character strings contained in at least one of the summary and the second document information.
31. A document management system as described in claim 27, wherein the specified condition is the presence of a character string in the summary generated by the summary generating means that may result in a low-accuracy summary, and the highlighting means highlights a character string from among one or more of the characteristic character strings contained in at least one of the summary and the second document information that may result in a low-accuracy summary in the summary generated by the summary generating means.
32. A document management system according to any one of claims 1 to 7, wherein the output processing means outputs the first document information acquired by the document information acquisition means, the summary acquired by the summary generation means, and the second document information acquired by the second information acquisition means in association with each other.
33. A document management system as described in claim 32, further comprising: a second information acquisition means for acquiring second document information; and a highlighting means for highlighting one or more characteristic character strings contained in at least one of the summary, the first document information, and the second document information, the characteristic character strings satisfying a predetermined condition.
34. A document management system as described in claim 33, wherein the specified condition is that a characteristic character string corresponding to the second document information and the summary exists, and the highlighting means highlights the characteristic character string corresponding to the second document information and the summary from among one or more characteristic character strings contained in at least one of the first document information, the second document information, and the summary.
35. A document management system as described in claim 33, wherein the specified condition is that the characteristic character string is present in the second document information but not in the summary, and the highlighting means highlights, among one or more of the characteristic character strings contained in at least one of the first document information or the second document information, the characteristic character string that is present in the second document information but not in the summary.
36. A document management system as described in claim 33, wherein the specified condition is that the characteristic character string is present in the summary but not present in the first document information or the second document information, and the highlighting means highlights, of the one or more characteristic character strings included in the summary, the characteristic character string that is present in the summary but not present in the first document information or the second document information.
37. A document management system according to any one of claims 1 to 7, wherein the document information acquisition means receives input of voice information, analyzes the voice information and converts it into text, thereby acquiring the first document information.
38. A program for causing a management server to function as each of the means of a document management system according to any one of claims 1 to 37.
39. A document management method comprising: acquiring first document information; inputting the first document information into a trained machine learning model that has performed machine learning to generate a summary using document information as input; acquiring a summary for the first document information from the machine learning model; and outputting the summary.
Citation Information
Patent Citations
Summary document creation support system and program
JP2010224639A
Interactive text summarization apparatus and method
JP2017111190A
Correspondence presentation device and correspondence method
JP2020098322A
Oncology workflow for clinical decision support
WO2022155607A1
