JP7536048B2 - Computer program, information processing device, and information processing method - Google Patents

Description

The present invention relates to a computer program, an information processing device, and an information processing method.

Conventionally, in response to inquiries from users, a computer would automatically respond by searching knowledge consisting of manuals and past cases, or the respondent would use the searched knowledge to provide an answer. When a computer searches for knowledge, it obtains keywords by performing a semantic analysis of the sentences and words contained in the inquiry, and then performs a full-text search of the knowledge.

JP 2002-189837 A

However, when a computer performs a full-text search of knowledge, it is possible that a significant amount of unnecessary knowledge will be included, making it difficult to narrow down the knowledge to be provided to the user.

The present invention has been made in consideration of the above circumstances, and aims to provide a computer program, an information processing device, and an information processing method that can identify knowledge to be presented to a user based on priority.

The computer program disclosed herein is a computer program that causes a computer to execute a process of acquiring information posted by a user, calculating a priority of knowledge data to be presented to the user based on the acquired posted information by referring to the storage contents of a knowledge database that associates and stores multiple knowledge data including problems in a specific field and solutions to those problems with an evaluation of each piece of knowledge data, and outputting the knowledge data read from the knowledge database based on the calculated priority.

According to the present disclosure, it is possible to identify knowledge to be presented to a user based on priority.

1 is an explanatory diagram illustrating an overview of an information processing system according to a first embodiment; 2 is a block diagram showing an internal configuration of a management server according to the first embodiment; FIG. 2 is a block diagram showing the internal configuration of a user terminal. FIG. 2 is a conceptual diagram showing an example of the configuration of a knowledge DB. 11 is a flowchart illustrating a procedure for creating a knowledge DB. FIG. 11 is an explanatory diagram illustrating a method for extracting a speech part. 10 is a flowchart illustrating a procedure for presenting knowledge data in the first embodiment. FIG. 13 is a schematic diagram showing a display example on a user terminal. FIG. 13 is a schematic diagram showing an example of an evaluation input screen. FIG. 11 is a block diagram showing an internal configuration of a management server according to a second embodiment. FIG. 2 is a conceptual diagram showing an example of the configuration of a first knowledge DB and a second knowledge DB. 13 is a flowchart illustrating a procedure for presenting knowledge data in the second embodiment. 13 is a flowchart illustrating a procedure for presenting knowledge data in the third embodiment. 11 is a flowchart illustrating a procedure for introducing cross-disciplinary knowledge. FIG. 13 is a conceptual diagram illustrating an example of an attribute table. 13 is a flowchart illustrating a procedure of a process executed by a management server according to a fifth embodiment. FIG. 13 is a schematic diagram showing an example of display of user information. FIG. 2 is a schematic diagram showing a first display example of a knowledge graph. FIG. 13 is a schematic diagram showing a second display example of a knowledge graph.

Hereinafter, the present invention will be described in detail with reference to the drawings showing embodiments thereof.
(Embodiment 1)
Fig. 1 is an explanatory diagram for explaining an overview of an information processing system according to the first embodiment. The information processing system according to the first embodiment includes a management server 10 and a user terminal 20. These servers and terminals are communicably connected to each other via a communication network NW. In the configuration example of Fig. 1, for the sake of simplicity, one user terminal 20 is shown, but multiple user terminals 20 may be connected to the communication network NW.

The management server 10 includes a knowledge DB 121 (see FIG. 2), which will be described later. The knowledge DB 121 is a database that stores a plurality of knowledge data including problems in a specific field and solutions to the problems, as well as evaluations of each knowledge data in association with each other.

When the management server 10 acquires user information transmitted from the user terminal 20 via the communication network NW, it calculates the priority of the knowledge data to be presented to the user by referring to the contents stored in the knowledge DB 121, and returns the knowledge data read from the knowledge DB 121 to the user terminal 20 based on the calculated priority.

An example of the configuration of an information processing system will be described below.
2 is a block diagram showing the internal configuration of the management server 10 according to embodiment 1. The management server 10 is, for example, a general-purpose or dedicated server computer, and includes a control unit 11, a storage unit 12, a communication unit 13, an operation unit 14, a display unit 15, etc.

The control unit 11 includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The CPU of the control unit 11 controls the operation of the various hardware components described above by expanding into the RAM various programs pre-stored in the ROM or the storage unit 12 and executing them, causing the entire device to function as the information processing device (computer) of the present application.

The control unit 11 is not limited to the above configuration, and may be an arithmetic circuit or control circuit having one or more GPUs (Graphics Processing Units), FPGAs (Field Programmable Gate Arrays), volatile or non-volatile memories, etc. The control unit 11 may also have functions such as a clock that outputs date and time information, a timer that measures the elapsed time from when an instruction to start measurement is given to when an instruction to end measurement is given, and a counter that counts numbers.

The storage unit 12 includes a storage device using a hard disk, a flash memory, or the like. The storage unit 12 stores computer programs executed by the control unit 11, various data acquired from the outside, various data generated within the device, and the like. In this embodiment, the storage unit 12 includes a knowledge DB 121 that stores knowledge data. The configuration of the knowledge DB 121 will be described in detail later.

The communication unit 13 has a communication interface for connecting to the communication network NW. The interface provided in the communication unit 13 is, for example, a communication interface conforming to a wireless communication standard such as WiFi (registered trademark), 3G, 4G, 5G, or LTE (Long Term Evolution). The communication unit 13 transmits various information to be notified to the outside, and receives various information transmitted from the outside to the device itself.

The operation unit 14 is equipped with input devices such as a keyboard and a mouse, and accepts input of various information. The control unit 11 performs appropriate control based on the information input from the operation unit 14, and stores the input information in the memory unit 12 as necessary.

The display unit 15 is equipped with a display device such as a liquid crystal display panel or an organic EL display panel, and displays information to be notified to the administrator based on a control signal output from the control unit 11.

The management server 10 does not have to be a single computer, but may be a computer system consisting of multiple computers and peripheral devices. Furthermore, the management server 10 may be a virtual machine that is virtually constructed by software.

In addition, in this embodiment, the management server 10 is configured to include the knowledge DB 121, but the knowledge DB 121 may be provided outside the management server 10. For example, the knowledge DB 121 may be provided so as to be accessible from the management server 10 via the communication network NW.

Figure 3 is a block diagram showing the internal configuration of the user terminal 20. The user terminal 20 is a terminal device such as a smartphone, a tablet terminal, or a personal computer, and includes a control unit 21, a storage unit 22, a communication unit 23, a voice input unit 24, an operation unit 25, and a display unit 26.

The control unit 21 includes, for example, a CPU, a ROM, a RAM, etc. The CPU of the control unit 21 controls the operation of the various hardware components described above by expanding into the RAM various programs pre-stored in the ROM or the storage unit 22 and executing them, causing the entire device to function as the user terminal of the present application.

The control unit 21 is not limited to the above configuration, and may be configured as a single piece of hardware (SoC: System On a Chip) that integrates a processor, memory, storage, communication interface, etc. The control unit 21 may also have functions such as a clock that outputs date and time information, a timer that measures the elapsed time from when an instruction to start measurement is given to when an instruction to end measurement is given, and a counter that counts numbers.

The memory unit 22 includes a storage device using a flash memory, a hard disk, etc. The memory unit 22 stores various computer programs executed by the control unit 21, data necessary for executing the computer programs, etc.

The communication unit 23 has a communication interface for connecting to the communication network NW. The communication interface of the communication unit 23 is, for example, a communication interface conforming to a wireless communication standard such as WiFi (registered trademark), 3G, 4G, 5G, or LTE. The communication unit 23 transmits various information to be notified to the outside, and receives various information transmitted from the outside to the device itself.

The audio input unit 24 includes an audio input device such as a microphone. The audio input unit 24 amplifies an analog audio signal input through the microphone, converts it into digital audio data, and then outputs it to the storage unit 22. The audio data output from the audio input unit 24 is stored in the storage unit 22.

The operation unit 25 is equipped with input devices such as a touch panel and operation buttons, and accepts various types of operation information and setting information. The control unit 21 performs appropriate control based on the operation information input from the operation unit 25, and stores the setting information in the storage unit 22 as necessary.

The display unit 26 is equipped with a display device such as a liquid crystal display panel or an organic EL display panel, and displays information to be notified to the user based on a control signal output from the control unit 21.

The configuration of the knowledge DB 121 will be described below.
4 is a conceptual diagram showing an example of the configuration of the knowledge DB 121. The knowledge DB 121 stores knowledge data, the number of views for each piece of knowledge data, and user evaluations of each piece of knowledge data in association with each other. Each piece of knowledge data includes a situation in a specific field, a problem that may occur in that situation, and a solution to the problem.

Knowledge DB121, shown as an example in FIG. 4, is a database that accumulates knowledge data in the field of digital solutions. The knowledge data stored in this knowledge DB121 includes, for example, a situation in which "when proposing a solution that utilizes digital technology for the first time," in which there is an issue that "the customer representative is not good at digital technology," and a solution to that issue that "provides an invitation to a DX study group." Another example of knowledge data includes a situation in which "when considering the introduction of a specific solution," in which there is an issue that "there are concerns about security, and the internal customer proposal is not progressing," and a solution to that issue that "introduces security support measures" and "introduces examples from other companies." Knowledge DB121 stores a large amount of knowledge data that includes such situations, issues, and solutions.

The number of views indicates the number of times the knowledge data has been viewed by a user. The management server 10 counts the number of times knowledge data has been provided to the user terminal 20 for each piece of knowledge data, and registers the count value in the knowledge DB 121 as the number of views of the knowledge data. The evaluation of each piece of knowledge data indicates the average value of the evaluations by users who used the knowledge data. When the management server 10 provides knowledge data to a user, it accepts the user's evaluation (e.g., a 5-point evaluation), calculates the average value based on the evaluations previously accepted, and registers it in the knowledge DB 121.

In the example of FIG. 4, the knowledge data is data including a situation, a problem, and a solution, but the data may include only an explanation and a solution without including a situation. Furthermore, the knowledge data may be data including only a problem without including a situation and a solution.

The knowledge data registered in the knowledge DB 121 may be automatically extracted by the management server 10 based on dialogue data between users.

The procedure for creating the knowledge DB 121 will be described below.
5 is a flow chart for explaining the procedure for creating the knowledge DB 121. The management server 10 acquires user dialogue data through the communication unit 13 (step S101). In this embodiment, the dialogue data is, for example, voice data, which is input from the voice input unit 24 of the user terminal 20 when the user is talking to a customer and is stored in the storage unit 22. The management server 10 acquires the user dialogue data in real time or at an appropriate timing after the end of the dialogue by communicating with the user terminal 20 through the communication network NW. The management server 10 stores the acquired dialogue data in the storage unit 12.

The control unit 11 of the management server 10 analyzes the acquired dialogue data and extracts speech portions including the situation, the problem, and the solution from the dialogue data (step S102). The control unit 11 can extract speech portions including the situation, the problem, and the solution from the dialogue data, for example, by estimating the emotions of the user (including the customer with whom the dialogue is being conducted) based on the dialogue data.

Figure 6 is an explanatory diagram explaining a method for extracting a speech part. The horizontal axis of the graph shown in Figure 6 indicates time, and the vertical axis indicates tension of the speech. The control unit 11 of the management server 10 can estimate the tension of the speech, for example, by using a learning model for emotion estimation. The learning model for emotion estimation is a known model such as SVM (Support Vector Machine), and is configured to output information on the level of tension when dialogue data (voice data) is input. Such a learning model is generated by learning according to a predetermined learning algorithm using a data set including the user's speech data and tension (correct answer data) identified by vital data such as heart rate and pulse rate as training data. The generated learning model may be stored in the storage unit 12 of the management server 10 or in an external server device.

When the control unit 11 of the management server 10 acquires dialogue data of a user through the communication unit 13, the control unit 11 inputs the acquired dialogue data into the trained learning model to estimate the tension of the speech. The control unit 11 distinguishes between sections where the tension of the conversation is relatively high and sections where the tension of the conversation is relatively low by referring to the calculation results of the learning model. In the example of FIG. 6, section C indicates a section that is identified as a section where the tension of the speech is relatively high as a result of the calculation by the learning model. The control unit 11 extracts the speech data of this section C as a speech portion including a solution. The control unit 11 also extracts the speech data of section B, which is connected to section C and where the tension of the speech is increasing, as a speech portion including a problem. Furthermore, the control unit 11 extracts the speech data of section A, which is connected to section B and where the tension of the speech is relatively low, as a speech portion including a situation.

The control unit 11 converts the dialogue data of the dialogue portion including the situation, the dialogue data of the dialogue portion including the problem, and the dialogue data of the dialogue portion including the solution into text data (step S103). A publicly known method such as a hidden Markov model or a statistical method can be used to convert the dialogue data (voice data) into text data (character string data).

After conversion to text data, the control unit 11 associates the text data of the dialogue portion including the situation, the text data of the dialogue portion including the problem, and the text data of the dialogue portion including the solution with each other, and registers them as knowledge data in the knowledge DB 121 (step S104). When registering the knowledge data in the knowledge DB 121, the control unit 11 may perform morphological analysis or context analysis, and may perform processing such as formatting the text to be registered or deleting unnecessary parts. Immediately after registration, the number of views of the knowledge data is registered as "0", and the evaluation is registered as "none".

The management server 10 acquires user speech data from the user terminal 20 at any time, not only before the system starts operating, but also after the system starts operating, and collects knowledge data and registers it in the knowledge DB 121 by extracting speech parts that include situations, issues, and solutions from the speech data.

In this embodiment, the tension of the speech is estimated from the voice data, and the dialogue parts corresponding to the situation, problem, and solution are extracted based on the estimated tension. However, the voice data may be converted into text data, and then the situation, problem, and solution may be extracted from the text data by understanding the user's intention using morphological analysis, contextual analysis, etc.

Furthermore, the speech data between users is not limited to voice data, but may be text data contained in chats or emails. In this case, the control unit 11 may use morphological analysis, contextual analysis, etc. to understand the user's intention, extract the situation, problem, and solution from the text data, and register the text data indicating the extracted situation, problem, and solution in the knowledge DB 121 as knowledge data.

The management server 10 may be configured to extract situations, issues, and solutions from text data equivalent to speech between users using natural language processing models such as BERT (Bidirectional Encoder Representations from Transformers) and GPT-3 (Generative Pre-Training-3). In this case, when text data is input, the situations, issues, and solutions may be extracted from the text data using a first natural language processing model tuned to output sentences indicating the situation, a second natural language processing model tuned to output sentences indicating the issue, and a third natural language processing model tuned to output sentences indicating the solution.

Furthermore, the users from whom dialogue data is obtained may be limited according to the attributes of the users. For example, if knowledge DB121 stores knowledge in the field of digital solutions, dialogue data may be obtained only from users who perform work related to digital solutions, and knowledge extracted from the dialogue data may be registered in knowledge DB121. For this reason, user authentication may be performed in advance, and dialogue data may be obtained only if the work of the authenticated user can be identified as being related to digital solutions.

The following describes the processing that the management server 10 executes after the knowledge DB 121 is created and the system begins operation.

Figure 7 is a flowchart explaining the procedure for presenting knowledge data in embodiment 1. The control unit 11 of the management server 10 acquires the transmitted information transmitted from the user through the communication unit 13 (step S121). The transmitted information transmitted from the user is any information transmitted from a user (e.g., a user registered in advance) who wishes to receive knowledge data. The transmitted information is input from the voice input unit 24 or the operation unit 25 of the user terminal 20, and is transmitted to the management server 10 via the communication network NW.

The control unit 11 extracts the task from the acquired user's transmitted information (step S122). The same method as described above is used to extract the task. That is, if the user's transmitted information is voice data, emotion estimation can be used to identify a section of the voice data where tension is gradually increasing, and the speech portion containing the task can be extracted. Also, if the user's transmitted information is text data, morphological analysis, context analysis, etc. can be used to understand the user's intention, and the speech portion corresponding to the task can be extracted.

The control unit 11 calculates the priority of the knowledge data to be presented to the user by referring to the contents stored in the knowledge DB 121 using the extracted task as a search key (step S123). The control unit 11 calculates the similarity between the task extracted in step S122 and the tasks registered in the knowledge DB 121, and determines the priority according to the level of similarity.

For example, the control unit 11 counts the frequency of occurrence of words included in the task, generates a document vector having the frequency of occurrence as an element, and calculates the similarity between the generated document vectors to obtain the similarity between the tasks. For example, if the task registered in the knowledge DB 121 is "customer representative is not good at digital," the control unit 11 generates a document vector having elements of 1, 1, 1 because the number of occurrences of the words "customer representative," "digital," and "not good" is one each. On the other hand, if the task extracted in step S122 is "not good at digital," the number of occurrences of "customer representative" is 0, and the number of occurrences of "digital" and "not good" are each one, so the control unit 11 generates a document vector having elements of 0, 1, 1. The control unit 11 can obtain the similarity between the tasks by calculating the similarity between the two generated document vectors. Note that a known method such as cosine similarity is used to calculate the similarity between the document vectors.

In addition, since the knowledge DB 121 includes user evaluations of each piece of knowledge data, the control unit 11 may calculate the priority by taking into account the evaluation of each piece of knowledge data. If the similarity between tasks is X and the evaluation of the knowledge data is Y, the control unit 11 can calculate the priority by, for example, w1 x X + w2 x Y. Here, w1 and w2 are weights that are set appropriately.

The control unit 11 reads out knowledge data from the knowledge DB 121 based on the priority calculated in step S123 (step S124). For example, the control unit 11 reads out only the knowledge data with the highest priority from the knowledge DB 121. Alternatively, the control unit 11 may set a threshold value for the priority and read out one or more pieces of knowledge data having a priority exceeding the threshold value from the knowledge DB 121. The control unit 11 may also read out a predetermined number of pieces of knowledge data from the knowledge DB 121 in descending order of priority.

The control unit 11 outputs the knowledge data read in step S124 (step S125). Specifically, the control unit 11 transmits the knowledge data read from the knowledge DB 121 to the user terminal 20 via the communication unit 13. The knowledge data to be transmitted may be only a solution to the problem, or may include the situation, the problem, and the solution. The control unit 11 also accesses the knowledge DB 121 and increases the number of views of the knowledge data provided to the user by +1.

When the user terminal 20 receives knowledge data transmitted from the management server 10 via the communication unit 23, it displays the received knowledge data on the display unit 26. FIG. 8 is a schematic diagram showing an example of display on the user terminal 20. The display screen 200 shown as an example in FIG. 8 includes a problem display section 201 that displays the user's problem, a solution display section 202 that displays solutions provided by the management server 10, a problem selection section 203, a send button 204, and an end button 205.

In the example of FIG. 8, in response to a user's issue (tweet) that "I'm not good at digital," multiple solutions proposed by the management server 10, such as "We'll send you an invitation to a DX study group," "We'll introduce you to related books," ..., and "We'll introduce you to a support center," are displayed in the solution display field 202. Note that these solutions may be displayed in order of priority calculated by the management server 10.

The problem selection field 203 accepts the selection of a solution by the user. When the send button 204 is pressed, the number of the solution selected by the user (n numbers from 1 to n in the example of FIG. 8) is notified to the management server 10. When the end button 205 is pressed, the display screen 200 is closed without notifying the solution number. The management server 10 may execute a process to solve the user's problem according to the solution number notified from the user terminal 20. For example, when solution 1 is selected on the display screen 200 of FIG. 8, an invitation to a DX study session may be notified to the user terminal 20 by email or SNS (Short Messaging Service).

The control unit 11 of the management server 10 requests the user to evaluate the knowledge data at an appropriate timing after providing the knowledge data to the user (step S126). Specifically, the control unit 11 transmits screen data of the evaluation input screen to the user terminal 20 via the communication unit 13, and displays the evaluation input screen on the display unit 26 of the user terminal 20, thereby requesting the user to evaluate the knowledge data. FIG. 9 is a schematic diagram showing an example of the evaluation input screen. The evaluation input screen 210 shown as an example in FIG. 9 includes evaluation reception fields 211-1, 211-2, ..., 211-n for receiving evaluations for each piece of knowledge data, a send button 212 for sending the received evaluation to the management server 10, and an end button 213 for closing the evaluation input screen 210 without sending the evaluation. The evaluation reception fields 211-1, 211-2, ..., 211-n are used to input the user's evaluation of each piece of knowledge data presented in step S125, for example, on a five-point scale. When the send button 212 is pressed, the user's evaluation is sent from the user terminal 20 to the management server 10.

In this embodiment, the management server 10 is configured to accept evaluations for all knowledge data presented to the user terminal 20, but the configuration may be such that evaluations are accepted only for the solution selected by the user.

The control unit 11 of the management server 10 receives the evaluation sent from the user terminal 20 (step S127) and updates the evaluation of the knowledge data in the knowledge DB 121 (step S128). At this time, the control unit 11 calculates the average value of the evaluation based on the evaluations received in the past, and registers the calculated average value of the evaluation in the knowledge DB 121.

In this embodiment, when information transmitted from the user terminal 20 is received, the user's problem is extracted by emotion estimation, context analysis, etc., but the configuration may also be such that the problem from the user is directly accepted and the accepted problem is transmitted to the management server 10. In this case, the management server 10 refers to the knowledge DB 121 based on the problem received from the user terminal 20 and searches for a corresponding solution.

As described above, in the first embodiment, knowledge data that matches the user's problem can be extracted from knowledge DB121 and presented to the user, and knowledge DB121 can be updated based on the user's evaluation of the presented knowledge data. The optimization of knowledge DB121 progresses based on the user's evaluation, and the optimization of the knowledge DB can increase the number of users and the frequency of use by users.

(Embodiment 2)
In the second embodiment, a configuration including knowledge DBs for multiple fields will be described.

Figure 10 is a block diagram showing the internal configuration of a management server 10 according to the second embodiment. The management server 10 is, for example, a general-purpose or dedicated server computer, and includes a control unit 11, a storage unit 12, a communication unit 13, an operation unit 14, a display unit 15, etc. The configuration of each of these hardware units is the same as in the first embodiment.

In the second embodiment, a configuration will be described in which the management server 10 includes a knowledge DB 121 that accumulates knowledge data in the digital solutions field, and a knowledge DB 122 in the e-commerce field. In the following description, the knowledge DB 121 will be referred to as the first knowledge DB 121, and the knowledge DB 121 will be referred to as the second knowledge DB 122.

Figure 11 is a conceptual diagram showing an example of the configuration of the first knowledge DB 121 and the second knowledge DB 122. The first knowledge DB 121 stores knowledge data in the digital solution field, the number of views for each piece of knowledge data, and user evaluations for each piece of knowledge data, in association with each other. The second knowledge DB 122 stores knowledge data in the e-commerce field, the number of views for each piece of knowledge data, and user evaluations for each piece of knowledge data, in association with each other. The management server 10 can create the first knowledge DB 121 and the second knowledge DB 122 using a method similar to that of embodiment 1.

When creating the first knowledge DB121, dialogue data is obtained only from users who perform digital solution-related tasks, and the knowledge extracted from the dialogue data is registered in the first knowledge DB121. When creating the second knowledge DB122, dialogue data is obtained only from users who perform e-commerce-related tasks, and the knowledge extracted from the dialogue data is registered in the second knowledge DB122.

The following describes the processing that the management server 10 executes after the first knowledge DB 121 and the second knowledge DB 122 are created and operation of the system is started.

Figure 12 is a flowchart explaining the procedure for presenting knowledge data in embodiment 2. The control unit 11 of the management server 10 acquires information sent by a user through the communication unit 13 (step S201), and extracts issues from the sent information (step S202).

Next, the control unit 11 determines the knowledge DB to be searched (step S203). For example, the control unit 11 may determine the knowledge DB to be searched by extracting a specific keyword from the acquired transmitted information or issue and identifying a field that includes the extracted keyword. The control unit 11 may also perform user authentication in advance and determine the knowledge DB to be searched based on the user attributes identified at the time of authentication. In this embodiment, either the first knowledge DB 121 or the second knowledge DB 122 is determined.

The control unit 11 refers to the knowledge DB determined in step S203 (for example, the first knowledge DB 121) and calculates the priority of the knowledge data to be presented in the same procedure as in embodiment 1 (step S204).

Next, the control unit 11 judges whether or not to provide knowledge from a different field (step S205). For example, if the control unit 11 is unable to extract a predetermined number or more (e.g., 1 or more) of knowledge whose priority is higher than a set value from the first knowledge DB 121, the control unit 11 judges to provide knowledge data from the second knowledge DB 122 in which knowledge data from a different field is registered. Alternatively, the control unit 11 may judge to provide knowledge from a different field if the number of knowledge data registered in the knowledge DB determined in step S203 is not sufficiently large (i.e., if it is less than the set number). If it is judged not to provide knowledge from a different field (S205: NO), the control unit 11 executes the processing from step S207 onwards.

When it is determined that knowledge from a different field is to be provided (S205: YES), the control unit 11 refers to a knowledge DB from the different field (e.g., the second knowledge DB 122) and calculates the priority of the knowledge data to be presented in the same procedure as in embodiment 1 (step S206). That is, the control unit 11 determines the similarity between the task extracted in step S202 and the task registered in the knowledge DB from the different field, and determines the priority according to the level of similarity.

The control unit 11 reads out the knowledge data from the knowledge DB based on the calculated priority of the knowledge data (step S207). The control unit 11 may read out the knowledge data from both the first knowledge DB 121 and the second knowledge DB 122, or may read out the knowledge data from either one of them. The control unit 11 outputs the read out knowledge data to the outside (step S208).

The control unit 11 requests the user to evaluate the knowledge data at an appropriate timing after providing the knowledge data to the user (step S209). The control unit 11 receives the evaluation of the knowledge data sent from the user terminal 20 (step S210), and updates the knowledge DB based on the received evaluation (step S211).

As described above, in the second embodiment, in cases where a solution to a problem in a particular field (e.g., the digital solutions field) is not registered in the knowledge DB for that field, knowledge data can be searched for from a knowledge DB for a different field and provided to the user.

(Embodiment 3)
In the third embodiment, a configuration will be described in which knowledge DBs in a plurality of fields are searched across to provide knowledge data to a user.

The management server 10 according to the third embodiment is assumed to have a first knowledge DB 121 and a second knowledge DB 122, similar to the second embodiment.

Figure 13 is a flowchart explaining the procedure for presenting knowledge data in embodiment 3. The control unit 11 of the management server 10 acquires information sent by a user through the communication unit 13 (step S301), and extracts issues from the information sent (step S302).

The control unit 11 refers to a knowledge DB for a specific field (for example, the first knowledge DB 121) and calculates the priority of the knowledge data to be presented in the same procedure as in the first embodiment (step S303). The control unit 11 reads out the knowledge data from the first knowledge DB 121 based on the calculated priority of the knowledge data (step S304).

Next, the control unit 11 reads out knowledge data related to the read knowledge data in the specific field from a knowledge DB in a different field (e.g., the second knowledge DB 122) (step S305). The relatedness of the knowledge data is derived based on the similarity of the problems or solutions. For example, the control unit 11 can find the similarity between problems or solutions by counting the frequency of occurrence of words included in the problem or solution, generating a document vector having the frequency of occurrence as an element, and calculating the similarity between the generated document vectors. The control unit 11 can identify problems or solved problems with a high degree of similarity as related knowledge data.

The control unit 11 outputs the knowledge data read in steps S304 and S305 to the outside (step S306).

At an appropriate timing after providing the knowledge data to the user, the control unit 11 requests the user to evaluate the knowledge data (step S307). The control unit 11 receives the evaluation of the knowledge data sent from the user terminal 20 (step S308), and updates the knowledge DB based on the received evaluation (step S309).

As described above, in the third embodiment, in addition to knowledge data extracted from a knowledge DB in a specific field (e.g., the digital solutions field), knowledge data from a different field (e.g., the e-commerce field) that is related to the knowledge data can also be provided to the user.

In this embodiment, the configuration is such that knowledge data from a specific field and knowledge data from a different field that is related to the knowledge data are provided to the user together, but the configuration may also be such that knowledge data from a different field that is not related to the knowledge data from the specific field is output together. In this case, in step S305 of the above-mentioned flowchart, knowledge data that has a low similarity to the problem or problem to be solved can be identified as related knowledge data.

(Embodiment 4)
In the fourth embodiment, a configuration for introducing knowledge data from a different field into a knowledge DB in a specific field will be described.

Figure 14 is a flowchart explaining the procedure for introducing knowledge from different fields. At an appropriate timing after creating a knowledge DB in a specific field, the control unit 11 counts the amount of knowledge data stored in the knowledge DB (step S401).

The control unit 11 compares the counted accumulated number with a set value and determines whether the accumulated number is less than the set value (step S402). The set value is appropriately set by an administrator, etc.

When it is determined that the accumulated number is less than the set value (S402: YES), the control unit 11 introduces the knowledge data registered in the knowledge DB of the different field to the knowledge DB of the specific field (step S403). At this time, the control unit 11 reads out a part or all of the knowledge data registered in the knowledge DB of the different field, and registers the read out knowledge data in the knowledge DB of the specific field as knowledge data of the specific field. When reading out a part of the knowledge data registered in the knowledge DB of the different field, the control unit 11 may, for example, extract knowledge data of the different field that is related to the knowledge data of the specific field, and register them in the knowledge DB as knowledge data of the specific field. The control unit 11 may also extract knowledge data in the knowledge DB of the different field that has an evaluation higher than the set value, and register them in the knowledge DB as knowledge data of the specific field.

In step S402, if it is determined that the accumulated number is equal to or greater than the set value and a sufficient amount of knowledge data is registered in the knowledge DB (S402: NO), the control unit 11 ends the processing according to this flowchart without introducing knowledge data from a different field.

As described above, in the fourth embodiment, knowledge data from a different field is introduced to a knowledge DB in a specific field where the number of registered knowledge data is small and the knowledge data may not be fully utilized. As a result, for example, knowledge data from a completely different field, such as the medical field, may be introduced to a knowledge DB in the financial field, making it possible to provide stimulating solutions to users in the financial field and encouraging them to use the knowledge DB.

In this embodiment, the number of accumulated knowledge data is used as a judgment index to determine whether or not to introduce knowledge data from a different field, but the judgment index is not limited to the accumulated number, and may be the accumulation speed or accumulation acceleration. For example, the control unit 11 may calculate the number of accumulated knowledge data per specified period as the accumulation speed, and if this is less than a set value, determine to introduce knowledge data from a different field. The control unit 11 may also calculate the accumulation speed of knowledge data per specified period as the accumulation acceleration, and if this is less than a set value, determine to introduce knowledge data from a different field.

Furthermore, the frequency of use of knowledge data in a specific field may be used as an index to determine whether or not to introduce knowledge data from a different field. The control unit 11 may refer to a knowledge DB in a specific field and count the number of times knowledge data has been provided to a user as the frequency of use, and if this is less than a set value, may determine to introduce knowledge data from a different field.

(Embodiment 5)
In the fifth embodiment, a configuration for prompting a user to register knowledge data will be described.

The management server 10 according to the fifth embodiment includes an attribute table 120 in the storage unit 12, which stores user attribute information. FIG. 15 is a conceptual diagram showing an example of the attribute table 120. The attribute table 120 stores user attribute information such as a user ID, password, name, email address, department, field, and frequency of use in association with each other. Here, the ID and password are authentication information used when performing user authentication in the management server 10. The management server 10 can perform user authentication using the ID and password when using this system. The field indicates the field for which the user wishes to be provided with knowledge data. The frequency of use indicates the number of times the user has received knowledge data from the management server 10. The attribute information is not limited to the above, and may also include information indicating whether or not the user is expected to provide a good solution.

Figure 16 is a flowchart explaining the procedure of the process executed by the management server 10 according to the fifth embodiment. The control unit 11 of the management server 10 uses an internal timer to measure the time that has elapsed since the knowledge data was last registered in the knowledge DB 121 (step S501), and determines whether the set time has elapsed (step S502). If the set time has not elapsed (S502: NO), the control unit 11 returns the process to step S501.

When it is determined that a set time has elapsed since the last time knowledge data was registered (S502: YES), the control unit 11 refers to the attribute table 120 and identifies users who should be prompted to register knowledge data (step S503). When the registration target is the knowledge DB 121 in the digital solutions field, the control unit 11 searches the attribute table 120 for users who belong to the digital solutions field, and among the searched users, for example, identifies the user with the highest frequency of use as a user who should be prompted to register knowledge data. The control unit 11 may also identify users who are expected to provide good solutions as users who should be prompted to register knowledge data.

The control unit 11 requests the identified user to register the knowledge data (step S504). Specifically, the control unit 11 refers to the attribute table 120, reads the email address of the requesting user, and sends an email addressed to the read email address, thereby requesting the registration of the knowledge data.

After requesting the user to register knowledge data, the control unit 11 determines whether or not knowledge data transmitted from the user terminal 20 of the user has been received (step S505). If knowledge data has not been received (S505: NO), the control unit 11 waits until knowledge data is received.

When knowledge data is received, the control unit 11 registers the situation, problem, and solution contained in the received knowledge data in a knowledge DB for the relevant field (e.g., knowledge DB 121) (step S506).

As described above, in the fifth embodiment, users who should be encouraged to register are identified based on the user's attribute information, and the identified users are encouraged to register knowledge data, so that the knowledge DB can be updated on an ongoing basis. Furthermore, by updating the knowledge data on an ongoing basis, the frequency of use by users increases.

In this embodiment, if knowledge data is not received from a user who has been prompted to register, the system waits until knowledge data is received from that user. However, instead of waiting, it is also possible to search for another user and prompt the searched user to register knowledge data.

In this embodiment, users who should be encouraged to register are identified based on the user's attribute information, but it is also possible to ask each user whether or not they can register knowledge data, and obtain knowledge data from users who respond that they can register, and register the knowledge data in the knowledge DB.

(Embodiment 6)
In the sixth embodiment, a configuration for displaying information about a user who is using a knowledge DB will be described.

When the management server 10 according to the sixth embodiment receives a request from the user terminal 20, it reads user information from the attribute table 120 and provides the read information to the user terminal 20. At this time, the management server 10 may aggregate the information read from the attribute table 120 and provide it to the user terminal 20.

Figure 17 is a schematic diagram showing an example of user information display. The display screen 220 shown in Figure 17 shows an example of a screen displayed on the display unit 26 of the user terminal 20. The display screen 220 shows a list of which departments have a high frequency of use of knowledge data for each field. The management server 10 can generate the display screen 220 shown in Figure 17 by referring to the attribute table 120 and tallying up the frequency of use of knowledge data for each department for each field.

In FIG. 17, an example of a display screen 220 that displays the frequency of use for each department, aggregated by field, is shown, but the frequency of use for each user, aggregated by field, may also be displayed on the display screen 220.

As described above, in the sixth embodiment, information on users who use the knowledge DB is displayed, so that the browsing behavior of frequently-used users can induce browsing behavior in other users.

(Seventh embodiment)
In the seventh embodiment, a configuration for displaying a knowledge graph will be described.

FIG. 18 is a schematic diagram showing a first display example of a knowledge graph. The display screen 230 shown in FIG. 18 as a first display example includes a knowledge graph in the digital solutions field and a knowledge graph in the e-commerce field. The nodes in each knowledge graph represent words extracted from knowledge data in the knowledge DB. In the example of FIG. 18, the white circles represent nodes in the digital solutions field, and the black circles represent nodes in the e-commerce field. The size of the node represents the number of views. The larger the node, the more views it has, and the smaller the node, the fewer views it has. In addition, nodes that are related to each other are connected by edges (also called links). The length of the edge represents the degree of relatedness. The shorter the edge, the higher the relatedness, and the longer the edge, the lower the relatedness. Furthermore, the arrow pointing from the node in the e-commerce field to the node in the digital solutions field indicates that knowledge data has been introduced from the e-commerce field to the digital solutions field by the cross-field knowledge introduction function.

In addition, the nodes may be highlighted by blinking or changing their display color depending on the rate of increase in the number of views over a certain period of time. Furthermore, the thickness of the edges may be changed depending on the rating of the knowledge data.

Furthermore, areas with low node density may be identified, and the user may be prompted to register knowledge data in order to increase the number of nodes in the identified areas. In the example of Figure 18, an area with low node density exists in the knowledge graph for the e-commerce field, and the user is prompted to register knowledge data for that area.

19 is a schematic diagram showing a second display example of a knowledge graph. The display screen 240 shown in FIG. 19 as a second display example includes an issue input field 241. When an issue is input in the issue input field 241, the control unit 11 of the management server 10 searches across multiple knowledge DBs and creates a knowledge graph based on the search results. That is, the control unit 11 creates a knowledge graph based on the words and the relevance between words contained in the issues searched by the cross-search, regardless of the field. As described above, the size of the node represents the number of views, and the length of the edge represents the degree of relevance. In addition, the node may be highlighted by blinking or changing the display color depending on the rate of increase in the number of views over a certain period of time. Furthermore, the thickness of the edge may be changed depending on the evaluation of the knowledge data.

As described above, in embodiment 7, the registration status of knowledge data can be visualized.

The embodiments disclosed herein are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is indicated by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims.

REFERENCE SIGNS LIST 10 Management server 11 Control unit 12 Storage unit 13 Communication unit 14 Operation unit 15 Display unit 20 User terminal 21 Control unit 22 Storage unit
23 Communication unit 24 Voice input unit 25 Operation unit 26 Display unit 121, 122 Knowledge DB

Claims (16)

Acquire information transmitted by the user including a speech portion corresponding to the user's problem;
a knowledge database that stores a plurality of knowledge data including problems in a specific field and solutions to the problems, and an evaluation of each piece of knowledge data in association with each other, based on the problem of the user extracted from the acquired information transmitted by the user, and calculates a priority of the knowledge data to be presented to the user;
outputting the knowledge data read from the knowledge database based on the calculated priority;
A computer program for causing a computer to execute a process of mapping a plurality of knowledge data in the knowledge database using nodes indicating each piece of knowledge data and edges indicating the relationships between the nodes, and displaying the nodes and edges in different display modes depending on attributes of each piece of knowledge data including the strength of the relationship between the knowledge data, the number of views for each piece of knowledge data, and the rate of increase in the number of views over a certain period of time. Accepting an evaluation of the knowledge data;
The computer program product according to claim 1 , further comprising: updating the contents of the knowledge database based on the received evaluation. Acquire dialogue data relating to a dialogue between users;
Extracting problems and solutions to the problems from the acquired dialogue data to generate knowledge data;
3. The computer program product according to claim 1, for causing the computer to execute a process of storing the generated knowledge data in the knowledge database. 4. The computer program product according to claim 1, for causing the computer to execute a process of: associating a plurality of knowledge data including problems and solutions to the problems in a field different from the specific field with an evaluation of each piece of knowledge data and storing them in the knowledge database. The computer program product according to claim 4 , for causing the computer to execute a process of searching for knowledge data in the different field in response to a search result for knowledge data in the specific field. The computer program according to claim 4 or 5, for causing the computer to execute a process of outputting, when outputting the knowledge data in the specific field, the knowledge data in the different field that is related to the knowledge data in the specific field. The computer program according to claim 4 or claim 5, for causing the computer to execute a process of outputting, when outputting the knowledge data in the specific field, the knowledge data in the different field that is not related to the knowledge data in the specific field. The computer program according to claim 6 or 7, for causing the computer to execute a process of deriving a relevance between the knowledge data in the specific field and the knowledge data in the different field based on a similarity between problems or solutions between the two. Counting the amount of accumulated knowledge data in the specific field;
9. The computer program product according to claim 4, for causing the computer to execute a process of: introducing the knowledge data of the different field as knowledge data of the specific field when the counted accumulated number is less than a set value. Calculating the accumulation speed or accumulation acceleration of knowledge data in the specific field;
9. The computer program according to claim 4, for causing the computer to execute a process of: introducing the knowledge data of the different field as knowledge data of the specific field when the calculated accumulation speed or accumulation acceleration is less than a set value. Calculating the frequency of use of knowledge data in the specific field;
9. The computer program product according to claim 4, for causing the computer to execute a process of: introducing the knowledge data of the different field as knowledge data of the specific field when the calculated frequency of use is less than a set value. Acquire attribute information of multiple users,
Based on the acquired attribute information, users who should be encouraged to register knowledge data are identified;
The computer program product according to claim 1 , for causing the computer to execute a process of notifying the identified user that the knowledge data should be registered. measuring an elapsed time from the registration of the knowledge data;
The computer program product according to claim 12, for causing the computer to execute a process of notifying the identified user that knowledge data should be registered when a set time has elapsed. 14. The computer program product according to claim 1, for causing the computer to execute a process of displaying information on users who are using the knowledge database. An acquisition unit that acquires information transmitted by the user including a speech portion corresponding to a problem of the user;
a calculation unit that calculates a priority level of knowledge data to be presented to a user by referring to a storage content of a knowledge database that stores a plurality of knowledge data including problems in a specific field and solutions to the problems, and an evaluation of each piece of knowledge data, based on the problem of the user extracted from the acquired information transmitted by the user;
an output unit that outputs the knowledge data read from the knowledge database based on the calculated priority ;
a display unit that, when mapping a plurality of pieces of knowledge data in the knowledge database with nodes indicating each piece of knowledge data and edges indicating a relationship between the nodes, maps and displays the nodes and edges in different display modes according to attributes of each piece of knowledge data including the strength of the relationship between the pieces of knowledge data, the number of views of each piece of knowledge data, and an increase rate of the number of views in a certain period of time;
An information processing device comprising: Acquire information transmitted by the user including a speech portion corresponding to the user's problem;
a knowledge database that stores a plurality of knowledge data including problems in a specific field and solutions to the problems, and an evaluation of each piece of knowledge data in association with each other, based on the problem of the user extracted from the acquired information transmitted by the user, and calculates a priority of the knowledge data to be presented to the user;
outputting the knowledge data read from the knowledge database based on the calculated priority;
An information processing method that executes a process by a computer to map multiple knowledge data in the knowledge database using nodes indicating each piece of knowledge data and edges indicating the relationships between the nodes, and to display the nodes and edges in different display modes depending on attributes of each piece of knowledge data including the strength of the relationship between the knowledge data, the number of views for each piece of knowledge data, and the rate of increase in the number of views over a certain period of time.

Images