Knowledge accumulation support methods, knowledge accumulation support programs, and content

Abstract

To collect and accumulate highly practical and accurate knowledge information from experts. [Means for solving the problem] We propose a knowledge accumulation support method that uses a computer to perform the following steps: a question-related information input receiving step that receives input of information linked to the knowledge attributes to be accumulated; a question extraction step that extracts one or more questions related to the received information; a question output step that outputs the extracted questions; and a response recording step that records the responses to the output questions linked to those questions.

Description

Technical Field

[0001] The present invention relates to a method and the like for efficiently accumulating knowledge from persons having various types of skilled expertise in a business operator.

Background Art

[0002] Against the background of the fluidization of human resources, the aging of engineers, and the demand for personnel to be immediately deployable, there is a growing demand for the effective utilization of expertise accumulated in the past within business operators. By actively utilizing such expertise, which is also referred to as knowledge, the continuity of business activities can be ensured, business growth can be promoted, or the possibility of avoiding troubles can be increased.

[0003] Various techniques have been disclosed regarding initiatives related to the effective utilization of such knowledge. First, techniques for facilitating the extraction of accumulated knowledge in timely scenes of business activities are known. In addition, in a certain aspect, techniques for efficiently accumulating such expertise and facilitating preparation for knowledge utilization at a required timing are also known.

[0004] Specifically, Patent Document 1 discloses a technique for extracting or editing information well known by skilled persons as knowledge information, using a database in which work videos and uttered voices related to the work are associated with each other and stored.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of Invention

Problems to be Solved by the Invention

[0006] However, when using the technology described in Patent Document 1, the setup involves recording the opinions freely expressed by skilled workers while they watch specific work videos, making it difficult to judge the usefulness or extent of the recorded information as knowledge. Furthermore, even skilled workers who encountered this setup did not know what kind of information should be provided as knowledge, leaving challenges in the accuracy of the information provided. [Means for solving the problem]

[0007] To solve the above-mentioned problems, the present invention proposes a knowledge accumulation support method that includes a question-related information input receiving step for receiving question-related information which is information linked to the knowledge attributes to be accumulated; a question extraction step for extracting one or more questions related to the received question-related information; a question output step for outputting the extracted questions; and a response recording step for recording the responses to the output questions linked to those questions, all of which are executed by a computer.

[0008] Furthermore, in connection with the above invention, we also propose a knowledge accumulation support method in which the question output step further has an order control substep that controls the order in which questions are output according to the knowledge attributes associated with the question.

[0009] Furthermore, in relation to the above inventions, we also propose a knowledge accumulation method in which the question output step further has a related question substep that outputs questions related to the question in accordance with the response result to the output question.

[0010] Furthermore, in relation to the above inventions, we also propose a knowledge accumulation method in which the response recording step further includes a summary recording substep that records the summarized result of the response to the outputted question in association with the question.

[0011] Furthermore, we also propose inventions related to programs and systems associated with each of the above inventions. [Effects of the Invention]

[0012] The present invention, primarily based on the configuration described above, makes it possible to collect and accumulate highly practical and accurate knowledge information from experts. [Brief explanation of the drawing]

[0013] [Figure 1] Schematic diagram of the present invention [Figure 2] This figure shows an example of the display screen of a provider terminal when providing knowledge using the system of the present invention. [Figure 3] This figure shows an example of a functional block of the knowledge accumulation support system of Embodiment 1. [Figure 4] This schematic diagram shows an example of a configuration when the functional components of the knowledge accumulation support system of the present invention are combined and implemented as a single hardware component. [Figure 5] This diagram shows an example of the processing flow of the knowledge accumulation support system of Embodiment 1. [Figure 6] This figure shows an example of a functional block of the knowledge accumulation support system of Embodiment 2. [Figure 7] This diagram shows an example of the processing flow of the knowledge accumulation support system of Embodiment 2. [Figure 8] This figure shows an example of a functional block of the knowledge accumulation support system of Embodiment 3. [Figure 9] This diagram shows an example of the processing flow of the knowledge accumulation support system of Embodiment 3. [Figure 10] This figure shows an example of a functional block of the knowledge accumulation support system of Embodiment 4. [Figure 11] This diagram shows an example of the processing flow of the knowledge accumulation support system of Embodiment 4. [Modes for carrying out the invention]

[0014] First, FIG. 1 is presented. FIG. 1 is a diagram showing an outline of the present invention, and as shown in the figure, the present invention can be implemented through transmission and reception of various types of information performed via a network between one or a plurality of server computers 0101 and 0102, and a terminal managed by a business operator (business operator terminal) 0111, a terminal managed by a knowledge provider in the business operator (provider terminal) 0121, and a terminal managed by a person who intends to utilize knowledge in the business operator (user terminal) 0131, respectively. A plurality of the business operator terminals, the provider terminals, and the user terminals can all provide and record knowledge, or utilize the knowledge at the same time.

[0015] Note that, in the present invention, it is also assumed that a knowledge provider may become a knowledge user in a certain aspect, and vice versa. That is, a user of the present system may use the present system to accumulate knowledge in a business operator, or may utilize the accumulated knowledge, and can use the system for each of these cases without using different IDs separately.

[0016] Further, a configuration is also conceivable in which the server computer constituting the present invention is connected via a network to one or a plurality of external servers 0141 managed by a business operator that operates an external service, and under this configuration, technical contrivances are also possible such that the knowledge modification support system using the present invention or the accumulated knowledge providing system is linked with the external service through an API or other modes. Specifically, it is possible to reduce the load of various processes in the present system by cooperating with an external server or the like for providing an external service that performs analysis processing of information acquired when accumulating knowledge or supports the accumulation processing.

[0017] Next, the various terminals connected to the server computer implementing the present invention are not particularly limited in type. Specifically, mobile terminals such as smartphones and tablets are conceivable, and wearable smart devices including smart watches are also conceivable, and of course terminals such as personal computers are also acceptable. If forced to specify, a terminal having a technical configuration that is equipped with a microphone and allows a knowledge provider to input information by voice is desirable. By enabling an input mode such as voice input that places a relatively low burden on the provider, it is possible to improve the efficiency of knowledge accumulation.

[0018] Next, Figure 2 is presented. Each of these drawings is a diagram showing an example of the display screen 0201 of a certain provider terminal 0200 (provider name: Masako Sato) when accepting the accumulation of knowledge using the system of the present invention. The provider terminal is a tablet terminal equipped with a camera 0202 and a microphone 0203, and transmits and receives voice, text, and image information via a network.

[0019] In the example shown in this figure, a display 0211 indicating that the knowledge attribute of "recruitment" has been selected in advance is presented, and further, as question-related information related to the knowledge attribute that has been input in advance, question-related information such as "1 Regarding human resources recruited via agents, what skill level groups should be specifically targeted?", "2 While startups rely on referrals and the like, how should they differentiate themselves from other companies?", and "3 Specifically, what potential groups should be appealed to?" is displayed in a predetermined area 0221 (the part after a portion of the second question-related information is displayed by screen transition). Then, when questions related to these pieces of question-related information are displayed in a predetermined area 0231, operations and information input processing from the provider who viewed the display are accepted as responses and can be displayed in a predetermined area 0241.

[0020] Furthermore, it is conceivable that these question-and-answer exchanges would be conducted by displaying the interviewer's avatar 0251 on the provider's terminal and controlling its display as appropriate. By adopting a configuration that provides the provider with a virtual communication environment with the interviewer, it is expected that the provider will find it easier to answer questions.

[0021] The provider will efficiently provide knowledge related to the relevant knowledge attribute by sequentially answering questions 0231 that are output in relation to the question-related information displayed in the designated area 0221. For businesses, this also makes it possible to prevent errors that can occur when receiving knowledge from experts, such as conversations going off-topic or interviews failing to properly articulate the interview content, which can happen when interviews are conducted between people.

[0022] The embodiments of the present invention will now be described with reference to the drawings. First, the relationship between the embodiments and the claims is as follows. Embodiment 1 mainly corresponds to claims 1, 5, and 6. Embodiment 2 mainly corresponds to claim 2. Embodiment 3 mainly corresponds to claim 3. Embodiment 4 mainly corresponds to claim 4. However, the technical features described in each embodiment can also be used in combination with the technical features described in other embodiments.

[0023] It should be noted that the present invention is not limited in any way to these embodiments, and can be implemented in various ways in accordance with common technical knowledge, having the technical idea described in each claim, without departing from the spirit thereof.

[0024] <<Embodiment 1>> <Overview> Figure 3 is a diagram showing an example of the functional blocks of the knowledge accumulation support system of this embodiment. As shown in the figure, the "knowledge accumulation support system" 0300 of this embodiment includes a "question-related information input receiving unit" 0301, a "question extraction unit" 0302, a "question output unit" 0303, and an "answer recording unit" 0304.

[0025] Furthermore, the knowledge accumulation support system described in detail below can be configured to implement one or more of its functions using multiple devices, and each of its functional blocks can be implemented as hardware or software. Taking a computer as an example, hardware components such as a CPU, main memory, GPU, TPU, image memory, bus, secondary storage device (hard disk or non-volatile memory), keyboard, microphone, touch panel, electronic pen for touching the touch panel, various input devices such as various terminals, speakers, displays and other various output devices, and other external peripheral devices can be included, as well as interfaces for those external peripheral devices, communication interfaces, driver programs and other application programs for controlling that hardware.

[0026] Then, through arithmetic processing according to the program deployed in main memory, data input from input devices and other interfaces and held in memory and hardware is processed and stored, and instructions for controlling the aforementioned hardware and software are created. Here, the above program may be implemented as multiple modularized programs, or two or more programs may be combined into a single program by cloud computing or other methods.

[0027] <Functional configuration> The "Question-Related Information Input Reception Unit" 0301 is configured to accept input of question-related information, which is information linked to the knowledge attributes to be collected. The types of knowledge attributes to be collected can be set as appropriate, and it is conceivable to divide the information attributes related to the business operator's knowledge into multiple levels, such as major, medium, and minor categories, and collect and record each level. For example, if a business operator wants to collect knowledge related to recruitment, knowledge attributes such as "A Recruitment," "B Document Screening," and "C First Interview" can be set in the major category, and knowledge attributes such as "A-1 Recruitment Media," "A-2 Recruitment Period," and "A-3 Job Titles" can be set in the medium category. Furthermore, depending on the case, knowledge attributes of minor categories such as "A-1-a Company Media," "A-1-b Other Companies' Media," and "A-1-c Hello Work" may also be set (of course, a configuration in which only the knowledge attribute "Recruitment" is set, as shown in Figure 2, is also within the scope of the present invention).

[0028] Furthermore, the information to be collected may be assigned either one knowledge attribute as described above, or multiple knowledge attributes may be assigned. By adopting this configuration, it becomes possible to efficiently collect knowledge that is useful across multiple information fields and to utilize it later.

[0029] Incidentally, knowledge attributes can be selected or specified by each business operator. A business operator can log in to this system using a specific ID and individually specify knowledge attributes with appropriate content to organize the information they want to accumulate as their own knowledge. Alternatively, a configuration can be considered in which users can specify or select attributes that are similar to the information they want to accumulate as knowledge from tag information related to knowledge attributes that is already set up in this system. By adopting the latter configuration in particular, it becomes easier for users to understand what kind of knowledge attribute information is being collected and utilized through this system.

[0030] Question-related information is entered as information that triggers the acquisition of knowledge related to the selected knowledge attributes, after pre-selecting the knowledge attributes of the information to be collected from the above-mentioned business operator. In the example shown in Figure 2, text related to specific questions such as "1. What kind of skills should be targeted for personnel recruited through an agent?" is entered as question-related information. Alternatively, the system could be configured to acquire only keywords related to the question, such as "through an agent," "target," and "what skills."

[0031] Other possible approaches include acquiring information about a business's operations, track record, and corporate policies, rather than using a question-and-answer format, and linking this information to knowledge attributes. For example, information such as a company's emphasis on hiring individuals with certain national qualifications, or its strength in serving clients in specific niche industries, could be acquired as question-related information. Specific acquisition methods could include obtaining links from the business's website and crawling the text information from those links. Alternatively, the system could read brochures, corporate PR articles, and other documents created by the business and extract the content of that data. This could involve directly reading text data, or reading PDF data and then extracting text data through OCR processing. Adopting such a configuration would facilitate the extraction of questions that take into account the specific characteristics of each business.

[0032] Furthermore, it is desirable to accept a certain number of inputs for each knowledge attribute related to questions. The more question-related information that is linked, the richer the content of the question output described later becomes, contributing to meaningful knowledge accumulation. For this reason, it is conceivable to set a minimum number of question-related information that should be linked to a single knowledge attribute in advance, and to implement control so that the system cannot proceed to the next processing step unless it accepts input of question-related information equal to or greater than that minimum number. Moreover, it is conceivable to implement a configuration in which the minimum number of question-related information that should be linked decreases as the classification of knowledge attributes is subdivided from broad categories to minor categories. By adopting this configuration, it becomes possible to obtain highly accurate question-related information according to the granularity of the knowledge attributes, and consequently, to realize effective knowledge accumulation.

[0033] Incidentally, as mentioned above, if knowledge attributes are set up in multiple layers, or if the system accepts input of question-related information spanning multiple knowledge attributes, it is also possible to adopt a configuration that links multiple knowledge attributes at once and accepts input of multiple question-related information for each.

[0034] The "question extraction unit" 0302 is configured to extract one or more questions related to the question-related information received. The extraction method can vary depending on the format in which the question-related information is input. For example, if the question-related information is input as text, it is conceivable to perform linguistic analysis on the content of the text and generate text data in question format according to the analysis results. It is also conceivable to perform a search process to extract pre-held question data (which may be question data generated during past knowledge accumulation or question data stored in advance).

[0035] Furthermore, if question-related information is entered via voice, it is conceivable to analyze the voice, convert it to text, and then perform the extraction process described above. By enabling voice input, conceptual knowledge that is difficult to express in text alone can be stored as question data in advance, and then extracted after determining its relevance to the question data.

[0036] It should be noted that, regardless of the number of question-related pieces of information received as input, it is possible to extract multiple questions. In order to effectively accumulate knowledge from a single piece of question-related information, it may be necessary to output multiple questions. When extracting multiple questions in this way, it may be necessary to output questions from different perspectives, or to output further questions based on the answers to the questions, and it is desirable to be able to handle all of these configurations.

[0037] Incidentally, the question generation process may be performed within this system, or it may be performed by outputting question-related information to an external server and extracting the results of the processing on that external server. This generation process can be implemented by executing expert systems suited to various industries and business types, or by running programs using artificial intelligence such as Akinator, but the specific processing method can be selected as appropriate.

[0038] Furthermore, if multiple questions are extracted, the system can accept user instructions to select which questions to output, or to specify or modify the order in which the questions should be output (details will be explained in Embodiments 2 and 3). For example, it can display multiple questions and accept the user's selection to rearrange them in the order in which they should be asked, or it can display a flowchart-like screen to branch the output of the next question depending on the answer to the previous question. By adopting this configuration, flexible question output tailored to the individual circumstances of the business operator is made possible, and consequently, highly accurate knowledge accumulation is made possible.

[0039] The "question output unit" 0303 is configured to output the extracted questions. When multiple questions should be extracted and output as described above, control is necessary so that multiple questions are not output at once, but rather the next question is output only after an answer has been obtained for the previous question. The output may be in the form of text, or it may be output by reading the text aloud. In particular, adopting a configuration that outputs by voice makes it easier for experts who are not accustomed to verbalizing knowledge to answer questions using conceptual expressions, thereby enabling effective knowledge accumulation.

[0040] Incidentally, questions may be output to provider terminals in various forms, but the provider terminal referred to here is not necessarily a terminal managed by the provider itself; it may be output to any terminal capable of sending and receiving information with this system. In the case of output to a terminal that cannot be managed as a provider terminal, it may be considered to prompt the terminal to input additional information as part of the response to identify that the responding party is the knowledge provider.

[0041] Incidentally, when displaying multiple questions as text, it is advisable to display subsequent questions on the same screen as the current question. This configuration essentially serves as a preview of the questions to the respondent, and in response to this preview, it is possible to expect related answers.

[0042] Furthermore, when outputting questions, it is conceivable to configure the system to also output other questions and answers related to the knowledge attributes associated with the question. Specifically, this could involve outputting related questions for a given question, similar questions received in the past, and even past answers to those questions. When outputting past answers, it is desirable to output them along with the date and time of the answer and the respondent (identified by an individual respondent ID, etc.). Outputting this information together encourages respondents to provide more accurate answers by being aware of the relevance to previously asked questions and answers.

[0043] The "response recording unit" 0304 is configured to record the response to the outputted question, linking it to the question. When recording, it is desirable that the date and time of the response and the provider ID are also linked, so that the response can be used for learning or reference in subsequent knowledge accumulation opportunities or opportunities to utilize the knowledge.

[0044] Regarding responses, while providers may adhere to a pre-specified response format, they may also be allowed to respond in any manner they choose. In this case, for example, one could allow respondents to answer freely for a set period of time (e.g., 30 seconds), and then acquire and record the audio data as the response. By recording the audio data as is, along with other information, without forcing any analysis or processing, it becomes possible to accumulate knowledge even for conceptual responses that are difficult to transcribe.

[0045] Furthermore, it is conceivable that the system could be structured to encourage knowledge providers to also provide information that identifies cases they have previously worked on in relation to the response. Specifically, this could involve obtaining information such as the case ID, project name, and client name within the company. By obtaining and recording this information as part of the response, it becomes easier to extract and utilize suitable knowledge from the perspective of similarity to past cases when later attempting to utilize the accumulated knowledge, thereby promoting knowledge utilization within the company.

[0046] <Specific Structure> Figure 4 is shown here. This figure is a schematic diagram showing an example of a configuration when the functional components of the knowledge accumulation support system of this embodiment are combined and implemented as a single hardware component. Each device is equipped with a "CPU" 0401 for performing various calculation processes, a "storage device (storage medium)" 0402, a "main memory" 0403, an "input / output interface" 0404, and a "network interface" 0405, and transmits and receives information with external peripheral devices such as a "microphone" 0406 via the input / output interface.

[0047] Furthermore, the knowledge accumulation support system of this embodiment can send and receive information with one or more "operator terminals" 0407, "provider (user) terminals" 0408, and "external servers" 0409 managed by partner businesses via a network interface.

[0048] The memory device stores various programs as described below, and the CPU reads these programs into the work area of ​​main memory, expands them, and executes them. These components are interconnected by data communication paths such as the "system bus" 0499, which transmit and receive information and perform processing (the basic configuration of the above is the same for all other devices described below).

[0049] (Specific configuration of the section for receiving input of question-related information) The question-related information input receiving unit consists of a computer program and computer hardware. Specifically, the CPU reads the "question-related information input receiving program" 0410 from the storage device into the main memory and executes it. It receives input of question-related information, which is information linked to the knowledge attributes to be collected, and stores it at a predetermined address in the main memory.

[0050] (Specific structure of the question extraction section) The question extraction unit consists of a computer program and computer hardware. Specifically, the CPU reads the "question extraction program" 0420 from the storage device into the main memory and executes it, reading the received question-related information and extracting one or more questions related to that information.

[0051] (Specific configuration of the question output section) The question output unit consists of a computer program and computer hardware. Specifically, the CPU reads the "question output program" 0430 from the storage device into the main memory and executes it, and outputs the questions extracted by the execution of the question extraction program to a designated provider terminal.

[0052] (Specific structure of the response recording section) The response recording unit consists of a computer program and computer hardware. Specifically, the CPU reads the "response recording program" 0440 from the storage device into the main memory and executes it. It then obtains the response to the question output through the execution of the question output program from the provider terminal that is the output destination, and stores it in a predetermined address in the main memory, linked to the question.

[0053] <Processing flow> Figure 5 shows an example of the processing flow in the knowledge accumulation support system of this embodiment. The processing flow in the figure consists of the following steps. First, in step S0501, the system receives input of question-related information, which is information linked to the knowledge attributes to be accumulated, from the provider terminal (question-related information input reception step). In step S0502, one or more questions related to the received question-related information are extracted (question extraction step), and in step S0503, the extracted questions are output to the provider terminal (question output step).

[0054] Subsequently, in step S0504, it is determined whether or not there is a response to the outputted question. If there is a response, in step S0505, the response is recorded in association with the outputted question (response recording step).

[0055] <Effects> By using a knowledge accumulation support system that employs the above configuration, it becomes possible to collect and accumulate highly practical and accurate knowledge information provided by experts.

[0056] <<Embodiment 2>> <Overview> The knowledge accumulation support system of this embodiment is basically the same as the technical features of the knowledge accumulation support system described in Embodiment 1, but it is further characterized by a configuration that controls the order in which questions are output according to the knowledge attributes associated with the questions.

[0057] <Functional configuration> Figure 6 shows an example of a functional block of the knowledge accumulation support system of this embodiment. As shown in the figure, the "knowledge accumulation support system" 0600 of this embodiment includes a "question-related information input receiving unit" 0601, a "question extraction unit" 0602, a "question output unit" 0603, and an "answer recording unit" 0604, and the question output unit further includes a "sequence control means" 0613. The basic configuration is the same as the knowledge accumulation support system described using Figure 3 of Embodiment 1, so the difference, the function of the "sequence control means" 0613, will be described below.

[0058] The "order control means" 0613 is configured to control the order in which questions are output in the question output unit according to the knowledge attributes associated with the questions.Embodiment 1 described a configuration that accepts the selection of questions to be output and the specification or modification of the order in which questions to be output.However, the order control means of this embodiment has the technical feature of setting the order in which multiple questions should be output according to their knowledge attributes, without accepting such selections, and performing output processing according to that setting.For example, in the case where knowledge attributes are divided into multiple levels, such as major categories, medium categories, and minor categories, as described in Embodiment 1, it is conceivable to control the output of questions in order from those associated with the major knowledge attributes of the category.In this case, by controlling the output of questions that are more abstract to those that are more specific and individual, it is possible to reduce the likelihood of discrepancies in the accuracy of responses by the provider.

[0059] Another possible configuration is to set a minimum number of required answers for questions possessing a specific knowledge attribute, output questions with the corresponding knowledge attribute, and then output questions with a different knowledge attribute. By adopting this configuration, it becomes possible to output questions that are related to each other and possess certain attributes in a concentrated manner, rather than outputting questions sporadically in a somewhat unrelated context, making it possible to elicit answers without confusing the provider.

[0060] Incidentally, if a required number of answers is set here, and the number of pre-extracted questions does not exceed a predetermined threshold, it is conceivable to configure the system to provide supplementary information to providers, such as comments that supplement the answers from the provider's own perspective. Even if the aim is to obtain a highly accurate knowledge base, the accuracy of the questions asked by users is not always high. Therefore, the system is configured to record information that providers feel should be provided, even if it was not asked, as knowledge, positioned as supplementary information. By adopting this configuration, truly valuable knowledge that may not have been intended by users can be appropriately accumulated from the provider's perspective.

[0061] <Specific Structure> The hardware configuration of each device constituting the knowledge accumulation support system of this embodiment is basically the same as the hardware configuration of the knowledge accumulation support system of Embodiment 1 described with reference to Figure 4. Therefore, the following will describe the specific processing of the "sequence control means" which has not been described so far.

[0062] (Specific configuration of the sequence control means) The sequence control mechanism specifically consists of a computer program and computer hardware. When the question output program is executed, the CPU reads a "sequence control subprogram" from the storage device into main memory and executes it, which then performs a process to control the order in which questions are output according to the knowledge attributes associated with the questions.

[0063] <Processing flow> Figure 7 shows an example of the processing flow in the knowledge accumulation support system of this embodiment. The processing flow in the figure consists of the following steps. First, in step S0701, the system receives input of information linked to the knowledge attributes to be accumulated from the provider terminal (question-related information input reception step). In step S0702, one or more questions related to the received information are extracted (question extraction step), and in step S0703, the extracted questions are output to the provider terminal in an order corresponding to the knowledge attributes linked to the questions (order-controlled question output step).

[0064] Subsequently, in step S0704, it is determined whether or not there is a response to the outputted question. If there is a response, in step S0705, the response is recorded in association with the outputted question (response recording step).

[0065] <Effects> By using the knowledge accumulation support system of this embodiment, it is possible to accumulate knowledge with higher accuracy compared to using the knowledge accumulation support system of Embodiment 1.

[0066] <<Embodiment 3>> <Overview> The knowledge accumulation support system of this embodiment is basically the same as the technical features of the knowledge accumulation support system described in Embodiment 1, but is further characterized by outputting related questions in response to the question output, depending on the response result to the question output.

[0067] <Functional configuration> Figure 8 shows an example of a functional block of the knowledge accumulation support system of this embodiment. As shown in the figure, the "knowledge accumulation support system" 0800 of this embodiment has a "question-related information input receiving unit" 0801, a "question extraction unit" 0802, a "question output unit" 0803, and an "answer recording unit" 0804, and the question output unit further has a "related question means" 0813. The basic configuration is the same as the knowledge accumulation support system described using Figure 3 of Embodiment 1, so the difference, the function of the "related question means" 0813, will be described below.

[0068] The "related question means" 0813 is configured in the question output unit to output further questions related to the question in accordance with the response result to the question output. As described in Embodiment 1, when the output order of multiple questions is set to be selectable, one possible configuration is to apply the response result from the provider to the newly set output order and output the next question. In some cases the response content is simply divided into branches such as "yes" and "no," while in other cases the response content is composed of a certain sentence, a configuration is also possible in which the meaning of the sentence is analyzed and the next question is output according to the result of such analysis. If such a configuration is adopted, the provider may feel as if they are providing knowledge while interacting with this system, and it may be possible to provide highly accurate knowledge based on the recall of memories such as past experiences during the communication process.

[0069] <Specific Structure> The hardware configuration of each device constituting the knowledge accumulation support system of this embodiment is basically the same as the hardware configuration of the knowledge accumulation support system of Embodiment 1 described with reference to Figure 4. Therefore, the following will describe the specific processing of the "related questioning means" which has not been described so far.

[0070] (Specific structure of related questioning methods) The related questioning means specifically consists of a computer program and computer hardware. When the question output program is executed, the CPU reads the "related question receiving subprogram" from the storage device into the main memory and executes it. Based on the response results to the question output, which are recorded by the execution of the answer recording program, it outputs further questions related to that question.

[0071] <Processing flow> Figure 9 shows an example of the processing flow in the knowledge accumulation support system of this embodiment. The processing flow in the figure consists of the following steps. First, in step S0901, the system receives input of information linked to the knowledge attributes to be accumulated from the provider terminal (question-related information input reception step). In step S0902, one or more questions related to the received information are extracted (question extraction step), and in step S0903, the extracted questions are output to the provider terminal (question output step).

[0072] Then, in step S0904, it is determined whether or not there is a response to the outputted question. If there is a response, in step S0905, the response is recorded in association with the outputted question (response recording step). Then, in step S0906, it is determined whether or not the response result allows for related questions. If it is determined that it does, in step S0907, the related questions are output again (related question output step), and the process thereafter is the same as in steps S0904 and below.

[0073] <Effects> By using the knowledge accumulation support system of this embodiment, it becomes possible to efficiently accumulate more relevant knowledge compared to using the knowledge accumulation support system of Embodiment 1.

[0074] <<Embodiment 4>> <Overview> The knowledge accumulation support system of this embodiment is basically the same as the technical features of the knowledge accumulation support system described in Embodiment 1, but it is further characterized in that, when recording the content of the response to a question, it records the summary result of the output response to the question in association with the question.

[0075] <Functional configuration> Figure 10 shows an example of a functional block of the knowledge accumulation support system of this embodiment. As shown in the figure, the "knowledge accumulation support system" 1000 of this embodiment includes a "question-related information input receiving unit" 1001, a "question extraction unit" 1002, a "question output unit" 1003, and an "answer recording unit" 1004, and the answer recording unit further includes a "summary recording means" 1014. The basic configuration is the same as the knowledge accumulation support system described using Figure 3 of Embodiment 1, so the difference, the function of the "summary recording means" 1014, will be described below.

[0076] The "summary recording means" 1014 is configured in the response recording unit to record the summarized result of the response to the outputted question, linked to the question. As described above, the provision of knowledge by providers is not always concise, and in some cases it may be abstract or difficult to put into text. Therefore, by summarizing these responses, it is possible to eliminate as much as possible expressions that make it difficult for users to understand, and to provide information that is easy to understand intuitively.

[0077] The specific method of summarization can be carried out using appropriate publicly available techniques. This may involve summarizing the content of the information obtained in text format, or it may involve summarizing the response by providing link information to other information, such as links to records of past cases linked to the obtained response. In any case, by adopting this configuration, even highly verbose responses or responses that are difficult to verbalize can be improved in terms of their potential for subsequent utilization as knowledge through summarization as an alternative method of information recording.

[0078] <Specific Structure> The hardware configuration of each device constituting the knowledge accumulation support system of this embodiment is basically the same as the hardware configuration of the knowledge accumulation support system of Embodiment 1 described with reference to Figure 4. Therefore, the following will describe the specific processing of the "summary recording means" which has not been described so far.

[0079] (Specific configuration of the summary recording device) The summary recording means specifically consists of a computer program and computer hardware. When the response recording program is executed, the CPU reads a "summary recording subprogram" from the storage device into the main memory and executes it, and stores the summarized result of the response to the output question, linked to the question, at a predetermined address in the main memory.

[0080] <Processing flow> Figure 11 shows an example of the processing flow in the knowledge accumulation support system of this embodiment. The processing flow in the figure consists of the following steps. First, in step S1101, information linked to the knowledge attributes to be accumulated is received from the provider terminal (question-related information input reception step). In step S1102, one or more questions related to the received information are extracted (question extraction step), and in step S1103, the extracted questions are output to the provider terminal (question output step).

[0081] Subsequently, in step S1104, it is determined whether or not there is a response to the outputted question. If there is a response, in step S1105, the summary result of the response to the outputted question is recorded in association with the question (response summary recording step).

[0082] <Effects> By using the knowledge accumulation support system of this embodiment, it becomes possible to efficiently accumulate knowledge that is more focused and versatile compared to using the knowledge accumulation support system of Embodiment 1. [Explanation of symbols]

[0083] 0300...Knowledge accumulation support system, 0301...Question-related information input reception unit, 0302...Question extraction unit, 0303...Question output unit, 0304...Answer recording unit

Claims

1. A question-related information input acceptance step that accepts input of one or more knowledge attributes to be aggregated and question-related information which is a question for aggregating knowledge related to said knowledge attributes, A question extraction step of extracting one or more questions from pre-stored question data that are determined to be related to the results of language analysis of the question-related information, A question output step that outputs the extracted questions, A response recording step which involves obtaining the response to the outputted question and recording the obtained result in association with the question, A knowledge accumulation method comprising a related question output step, in which, if multiple questions are extracted in the question extraction step, the response result to one of the output questions is analyzed, and other extracted questions are output according to the results of the analysis.

2. The response recording step is: The knowledge accumulation method according to claim 1, further comprising a summary recording substep of summarizing the responses to the outputted questions and recording the summarization results in association with the questions.

3. A question-related information input acceptance step that accepts input of one or more knowledge attributes to be aggregated and question-related information which is a question for aggregating knowledge related to said knowledge attributes, A question extraction step of extracting one or more questions from pre-stored question data that are determined to be related to the results of language analysis of the question-related information, A question output step that outputs the extracted questions, A response recording step which involves obtaining the response to the outputted question and recording the obtained result in association with the question, In the question extraction step, if multiple questions are extracted, the related question output step involves analyzing the response result to one of the output questions and outputting other extracted questions according to the results of that analysis. A knowledge accumulation program for execution on a computer.

4. A question-related information input receiving unit that accepts input of one or more knowledge attributes to be aggregated and question-related information which is a question for aggregating knowledge related to said knowledge attributes, A question extraction unit extracts one or more questions from pre-stored question data that are determined to be related to the results of language analysis of the question-related information, A question output unit that outputs the extracted questions, A response recording unit that acquires the response to the outputted question and records the acquired result in association with the question, When the question extraction unit extracts multiple questions, the related question output unit analyzes the response result to one of the output questions and outputs other extracted questions according to the result of that analysis. A knowledge accumulation support system equipped with [features / equipment].

Images