WO2024090367A1

WO2024090367A1 - Information processing method, computer program, and information processing device

Info

Publication number: WO2024090367A1
Application number: PCT/JP2023/038134
Authority: WO
Inventors: 貴仁松沢
Original assignee: 東京エレクトロン株式会社
Priority date: 2022-10-24
Filing date: 2023-10-23
Publication date: 2024-05-02

Abstract

Provided are an information processing method, a computer program, and an information processing device that can be expected to support decision making by users.　In an information processing method according to the present embodiment, an information processing device acquires objective information input by a user, extracts a plurality of constituent elements from the obtained objective information, extracts, on the basis of the extracted plurality of constituent elements, a plurality of keywords from a database that stores a plurality of keywords and relationships between the keywords, and outputs the extracted plurality of keywords and relationships between the extracted plurality of keywords.

Description

Information processing method, computer program, and information processing device

This disclosure relates to an information processing method, a computer program, and an information processing device.

Patent Document 1 proposes a knowledge information extraction system that acquires video of a task performed by a task performer, acquires audio of speech about the task from at least one of the task performer and an observer, records the video of the task and the audio as work video data and work audio data, respectively, in a work behavior record database, and at least one of the task performer and observer is an expert who is familiar with the task, extracts and edits information about the task that the expert is familiar with based on the work video data and work audio data as knowledge information, and records the extracted knowledge information in a knowledge information database.

International Publication No. 2021/079414

The present disclosure provides an information processing method, a computer program, and an information processing device that are expected to assist users in making decisions.

In one embodiment of the information processing method, an information processing device acquires target information input by a user, extracts multiple components from the acquired target information, extracts multiple keywords from a database that stores multiple keywords and relationships between the keywords based on the extracted multiple components, and outputs the extracted multiple keywords and the relationships between the multiple keywords.

This disclosure is expected to assist users in making decisions.

1 is a schematic diagram for explaining an overview of an information processing system according to an embodiment of the present invention; FIG. 13 is a schematic diagram showing an example of a purpose input. FIG. 2 is a schematic diagram showing an example of a knowledge network displayed on a terminal device. 2 is a block diagram showing a configuration example of a server device according to the present embodiment. FIG. FIG. 2 is a schematic diagram showing an example of a configuration of a keyword DB; 2 is a block diagram showing a configuration example of a terminal device according to the present embodiment; FIG. 5 is a flowchart showing an example of a procedure of decision-making support processing performed by the server device 1 according to the present embodiment. FIG. 13 is a schematic diagram showing an example of a detailed display of a knowledge network. FIG. 13 is a schematic diagram showing an example of a detailed display of a knowledge network. FIG. 13 is a schematic diagram showing an example of a detailed display of a knowledge network. 10 is a flowchart showing an example of a procedure of a keyword DB update process performed by the server device according to the embodiment;

Specific examples of information processing systems according to embodiments of the present disclosure are described below with reference to the drawings. Note that the present disclosure is not limited to these examples, but is set forth in the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

<System Overview>
1 is a schematic diagram for explaining an overview of an information processing system according to the present embodiment. The information processing system according to the present embodiment is a system for supporting a user's thinking regarding, for example, recipe development for a semiconductor manufacturing process (etching process). The information processing system according to the present embodiment is configured to include a server device 1 having a database that accumulates various information related to the semiconductor manufacturing process, and a terminal device 3 that a user uses to input and output information. The server device 1 and the terminal device 3 can transmit and receive information via various networks, such as the Internet, a LAN (Local Area Network), or a mobile phone communication network.

The information processing system according to this embodiment accepts input of a purpose in natural language by the user, for example, by performing interactive information input and output with the user at the terminal device 3. FIG. 2 is a schematic diagram showing an example of purpose input. The server device 1 of the information processing system according to this embodiment causes the terminal device 3 to display the purpose input screen shown in FIG. 2. The purpose input screen displayed by the terminal device 3 can have a screen configuration similar to that of a message exchange application in which multiple users exchange messages, for example. The purpose input screen according to this embodiment has a title bar at the top with the characters "purpose input screen" attached, a message display area in the center, and a text box at the bottom for text input.

For example, an icon with the word "System" attached to it is displayed on the far left side of the message display area of the purpose input screen, and a message from the system is displayed in a speech bubble associated with this icon. Also, an icon with the word "User" attached to it is displayed on the far right side of the message display area, and text information entered by the user is displayed in a speech bubble associated with this icon. Multiple speech bubbles are displayed in chronological order, arranged from top to bottom of the message display area.

In this example, in response to the system's question "Please enter your objective," the user inputs the objective "Develop a process recipe." In response to the next question from the system, "Do you have any more detailed information?", the user inputs the information "XX process of NAND-Flash." In response to the next question from the system, "Do you have any more detailed information?", the user inputs the information "Actively reuse similar development." In response to the next question from the system, "Do you have any more detailed information?", the user inputs the keyword "End." This keyword is used to instruct the system to end objective input. Note that the system and user messages shown in Figure 2 are merely examples and are not limited to these. Any keyword other than "end" may be used as the keyword to end objective input, and the configuration may be such that objective input is ended by a method other than keyword input.

In the example shown in Figure 2, the system asks two questions: "Please enter your purpose" and "Do you have any more detailed information?" However, questions are not limited to these two. Questions from the system can include messages that prompt the user to enter information such as the "category," "constraints," or "prerequisites" related to the purpose.

For example, the system may display a message prompting the user to input information such as the user's name or ID, and accept the input of this information from the user. The system may, for example, restrict access to information based on the information about the user. The system may also store background information, such as the user's knowledge level or area of expertise, in a database, and refer to the background information based on the input information about the user to carry out various subsequent processes in a manner appropriate to the user.

The purpose and associated information input by the user on the purpose input screen are transmitted as purpose information from the terminal device 3 to the server device 1. The server device 1 acquires the purpose information input by the user in natural language from the terminal device 3, and generates a knowledge network related to the user's purpose based on the acquired user's purpose information and information previously stored in the database, and transmits the knowledge network to the terminal device 3. The terminal device 3 displays the knowledge network received from the server device 1.

FIG. 3 is a schematic diagram showing an example of a knowledge network displayed by the terminal device 3. Based on the information received from the server device 1, the terminal device 3 displays a knowledge network according to the purpose entered by the user. In the illustrated screen, the terminal device 3 displays a triple oval as a background image, and arranges multiple keywords surrounded by rectangular frames overlaid on this background image, displaying the multiple keywords connected by straight lines as a knowledge network. The multiple keywords included in the knowledge network are classified in a hierarchical structure. In the knowledge network, keywords located closer to the center of the oval in the background image are higher-level keywords (larger category, abstract), and keywords located further outside the oval are lower-level keywords (smaller category, concrete). In the knowledge network, related keywords are associated with each other by connecting them with straight lines from higher-level keywords to lower-level keywords.

In the illustrated example, keywords such as "knowledge," "data," and "experience" are placed in the center of the triple oval as keywords for the "large category" of the top hierarchy. Second from the center of the triple oval, keywords such as "high frequency," "chemical reaction," "plasma," "similar process," "basic experiment," "point of interest," "trade-off," and "episode" are placed as keywords for the "medium category" of the second hierarchy. Third from the center of the triple oval, keywords such as "control orbit," "gas species," "electron orbit," "light emission," "knob," "recipe," "chamber atmosphere," "depth," "necking," "Boeing," "related reports," "in-house experts," and "related papers" are placed as keywords for the "principle/data" of the third hierarchy. Outside the triple oval, keywords such as "modern control theory," "first-principles calculation," and "simulation" are placed as more detailed information.

The knowledge network displayed by the terminal device 3 may also include three-dimensional rectangular frames surrounding keywords. In the illustrated example, these include "gas type," "plasma," "similar process," "trade-off," and "boeing." A keyword displayed three-dimensionally indicates that many lower-level keywords are associated with it, and that these lower-level keywords have not yet been displayed. The user can select a keyword displayed three-dimensionally to display a more detailed knowledge network.

In the information processing system according to this embodiment, the server device 1 generates a knowledge network according to the user's purpose based on information previously stored in a database, based on purpose information input by the user in natural language, and displays the knowledge network on the terminal device 3. This allows the user to understand, for example, what knowledge is needed for his or her purpose, or the correspondence between multiple pieces of knowledge, based on the knowledge network displayed on the terminal device 3. As a result, the information processing system according to this embodiment is expected to support decision-making regarding the user's purpose.

In this embodiment, a sentence in a natural language input by the user is acquired as the target information, but the information form and input method of the target information are not limited to this. For example, instead of a sentence, an input of a list of one or more words or a short sentence in a natural language may be accepted from the user as the target information. Also, for example, multiple options may be presented to the user in various formats such as a menu format, list box format, tab format, or table format, and one or more items selected from these multiple options may be accepted from the user as the target information. Any method of inputting the target information may be adopted.

Furthermore, as will be described in detail later, in the information processing system according to this embodiment, the user can edit the knowledge network displayed on the terminal device 3 by adding, changing, or deleting keywords, etc., and can create a knowledge network that is more suitable for the user. Information about the knowledge network edited and created by the user is fed back to the server device 1 and reflected in the database held by the server device 1.

<Device Configuration>
4 is a block diagram showing an example of the configuration of the server device 1 according to the present embodiment. The server device 1 according to the present embodiment is configured to include a processing unit 11, a storage unit 12, and a communication unit 13. Note that in the present embodiment, the processing is described as being performed by one server device 1, but the processing may be performed in a distributed manner by a plurality of server devices.

The processing unit 11 is configured using an arithmetic processing device such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit) or a quantum processor, a ROM (Read Only Memory) and a RAM (Random Access Memory), etc. The processing unit 11 reads out and executes a program 12a stored in the memory unit 12, thereby performing various processes such as a process of acquiring target information from a user via the terminal device 3 and a process of generating a knowledge network according to the target information.

The storage unit 12 is configured using a large-capacity storage device such as a hard disk or SSD (Solid State Drive). The storage unit 12 stores various programs executed by the processing unit 11 and various data necessary for the processing of the processing unit 11. In this embodiment, the storage unit 12 stores the program 12a executed by the processing unit 11. The storage unit 12 also includes a keyword DB (database) 12b that accumulates information about keywords necessary for generating a knowledge network, and a history storage unit 12c that stores the history of editing of the knowledge network by the user.

In this embodiment, the program (computer program, program product) 12a is provided in a form recorded on a recording medium 99 such as a memory card or optical disk, and the server device 1 reads the program 12a from the recording medium 99 and stores it in the memory unit 12. However, the program 12a may be written to the memory unit 12, for example, during the manufacturing stage of the server device 1. Also, for example, the program 12a may be distributed by another remote server device or the like and acquired by the server device 1 through communication. For example, the program 12a may be read from the recording medium 99 by a writing device and written to the memory unit 12 of the server device 1. The program 12a may be provided in a form distributed via a network, or may be provided in a form recorded on the recording medium 99.

The keyword DB 12b of the storage unit 12 is a database that stores keywords necessary for generating a knowledge network and relationships between multiple keywords. FIG. 5 is a schematic diagram showing an example of the configuration of the keyword DB 12b. The keyword DB 12b stores multiple keywords, for example, classified into multiple hierarchies from the first hierarchy to the Nth hierarchy. In the illustrated example, the keyword DB 12b stores "knowledge," "experience," "data," and "system" as keywords in the first hierarchy. The keyword DB 12b stores "high frequency," "temperature," "gas," "plasma," "process," "basic experiment," "focus point," "trade-off," "laboratory," "field," and "case study" as keywords in the second hierarchy. The keyword DB 12b stores "control," "chemical reaction," "furnace phenomenon," "use history," "parts," "light emission," and "knob" as keywords in the third hierarchy. The keyword DB 12b also stores "first principles," "plasma simulation," "electron energy transition," and "past report" as keywords in the Nth hierarchy.

In addition, the keyword DB 12b stores the relationship between keywords between each hierarchy. In the illustrated keyword DB 12b, for example, "knowledge" in the first hierarchy is associated with "high frequency" and "plasma" in the second hierarchy. Also, for example, "temperature" in the second hierarchy is associated with "chemical reaction" and "furnace phenomenon" in the third hierarchy. Note that in this example, keywords are associated between adjacent hierarchies, such as the first hierarchy and the second hierarchy, or the second hierarchy and the third hierarchy, but this is not limited to this. For example, keywords may be associated between the first hierarchy and the third hierarchy.

Although not shown in FIG. 5, each keyword is associated with numerical information indicating its importance (weight) and stored in keyword DB 12b. Importance can be expressed, for example, as a decimal value between 0 and 1, but is not limited to this. For example, when there are many candidates for keywords to be included in the knowledge network, server device 1 can prioritize and select keywords with high importance. Also, for example, when displaying the knowledge network, server device 1 can differentiate the display mode of keywords depending on their importance.

The keyword DB 12b is created in advance by, for example, a designer or manager of the information processing system according to this embodiment. The designer or manager can create the keyword DB 12b by collecting various information, such as books, papers, interviews with experts, past experimental information, manufacturing information, or device management information related to the semiconductor manufacturing process, extracting keywords from the collected information, and linking the keywords together. The processes of collecting information, extracting keywords, and linking the keywords together may be performed manually by, for example, a designer or manager, or may be performed using or in combination with software such as so-called AI (Artificial Intelligence). The keyword DB 12b created in advance is used to generate the above-mentioned knowledge network in this information processing system, and information is added or updated in response to editing of the knowledge network by a user using this information processing system. This allows the keyword DB 12b to accumulate the knowledge or will of more people, and it is expected that the server device 1 will generate a more accurate knowledge network.

In creating the keyword DB12b, it is important to collect detailed knowledge of related engineers or experts, including their backgrounds. For example, it is preferable to collect specialized knowledge that is the business background, information such as academic fields and papers on which the knowledge is based, and information such as important events in business experience, their results, customers involved in business, and manufacturing processes. Keywords are extracted from the collected information, and the extracted keywords are associated based on, for example, causal relationships or classification relationships. For example, known causal relationship analysis methods can be used to associate keywords. The server device 1 according to this embodiment presents a knowledge network to the user and accepts editing of the knowledge network by the user, and learns keywords or keyword associations that are important for the user's decision-making, and reflects them in the keyword DB12b. The server device 1 can reflect the subsequent generation of the knowledge network by learning keywords and their associations using post-event feedback and frequency measurement.

The history storage unit 12c of the storage unit 12 stores the history of editing operations performed by the user on the knowledge network. The user can perform editing operations such as adding, deleting, or changing keywords on the knowledge network displayed on the terminal device 3. The terminal device 3 transmits the contents of the editing operations performed by the user to the server device 1, and the server device 1, which receives this, stores the history of the editing operations performed by the user in the history storage unit 12c. The information stored in the history storage unit 12c is used when updating the information stored in the keyword DB 12b.

The communication unit 13 communicates with various devices via a wired or wireless network N, which may include, for example, the Internet, a LAN (Local Area Network), or a mobile phone network. In this embodiment, the communication unit 13 communicates with one or more terminal devices 3 via the network N. The communication unit 13 transmits data provided by the processing unit 11 to other devices, and provides data received from other devices to the processing unit 11.

The storage unit 12 may be an external storage device connected to the server device 1. The server device 1 may be a multi-computer including multiple computers, or may be a virtual machine virtually constructed by software. The server device 1 is not limited to the above configuration, and may include, for example, a reading unit that reads information stored in a portable storage medium, an input unit that accepts operational input, or a display unit that displays images.

In addition, in the server device 1 according to this embodiment, the processing unit 11 reads out and executes the program 12a stored in the memory unit 12, whereby the target information acquisition unit 11a, the component extraction unit 11b, the keyword extraction unit 11c, the knowledge network generation unit 11d, the editing processing unit 11e, the database update unit 11f, and the like are realized in the processing unit 11 as software-like functional units. Note that in this diagram, functional units related to processes such as the generation and display of a knowledge network are shown as functional units of the processing unit 11, and functional units related to other processes are not shown.

The objective information acquisition unit 11a performs processing to acquire information regarding the user's objective via the terminal device 3. In this embodiment, the objective information acquisition unit 11a acquires objective information by displaying questions for the user on the terminal device 3 and accepting input of answers to these questions from the user. The objective information acquisition unit 11a, for example, appropriately selects one question from a number of questions prepared in advance, and transmits the selected question to the terminal device 3 for display. At this time, the objective information acquisition unit 11a also acquires more objective information by presenting multiple questions in stages and accepting input of answers to each question.

In the example shown in FIG. 2, the target information acquisition unit 11a first presents the first question, "Please enter your purpose," to obtain a response from the user, and then presents the second question, "Is there any more detailed information?" to obtain a further response from the user. Furthermore, the target information acquisition unit 11a repeats the second question to obtain more target information from the user. Note that in this example, two types of questions are presented to the user to obtain responses, but the questions presented to the user are not limited to these two types. Questions may be prepared in a configuration of three or more stages, rather than two stages, and multiple questions may be prepared at each stage. Furthermore, the target information acquisition unit 11a may appropriately change the next question to be presented based on the response obtained from the user.

In addition, in this embodiment, the target information acquisition unit 11a acquires target information by repeating a question and answer session, but the method of acquiring target information is not limited to this. For example, the target information acquisition unit 11a may not ask the user questions, but may acquire a sentence in a natural language input by the user as target information. In addition, for example, the target information acquisition unit 11a may display a list of multiple keywords or sentences, etc. prepared in advance on the terminal device 3. The target information acquisition unit 11a may acquire target information from the user by accepting a selection of keywords or sentences, etc. related to the user's target from the displayed list. The target information acquisition unit 11a may acquire target information by combining these multiple methods.

The component extraction unit 11b performs processing to extract components contained in the target information based on the target information acquired by the target information acquisition unit 11a. The component extraction unit 11b performs processing such as morphological analysis on the natural language sentence acquired as target information to extract, for example, nouns and verbs contained in the natural language sentence, and treats the extracted nouns and verbs as components. The component extraction unit 11b also calculates, for example, the frequency of appearance in the target information sentence for each extracted component, and calculates the importance (weight) of each component based on the calculated frequency of appearance. The component extraction unit 11b also ranks the multiple components extracted from the target information based on the calculated importance.

The keyword extraction unit 11c performs a process of extracting keywords from the keyword DB 12b based on the multiple components extracted and ranked by the component extraction unit 11b. For example, the keyword extraction unit 11c first obtains the highest-ranked component, and searches for and extracts keywords that match or are similar to this component from the keyword DB 12b. The keyword extraction unit 11c then extracts one or more other keywords associated with the extracted keyword from the keyword DB 12b.

For example, if the component extraction unit 11b extracts "high frequency" as a component, the keyword extraction unit 11c can extract the "high frequency" keyword stored in the second layer in the keyword DB 12b shown in FIG. 5, as well as the "knowledge" keyword in the first layer and the "control" keyword in the third layer that are associated with this keyword. The keyword extraction unit 11c may further extract one or more keywords in the fourth layer that are associated with "control" in the third layer, and similarly extract keywords associated up to the Nth layer in order. Note that the keyword extraction unit 11c may extract keywords by setting a limit of, for example, two layers above and below each layer, rather than extracting related keywords for all layers from the first layer to the Nth layer.

The keyword extraction unit 11c extracts the above keywords from the ranked components, starting from the top component, and repeats keyword extraction until, for example, a component of a predetermined rank is reached or the number of extracted keywords reaches a predetermined number.

The knowledge network generation unit 11d performs processing to generate a knowledge network to be displayed on the terminal device 3 based on the multiple keywords extracted by the keyword extraction unit 11c. That is, the knowledge network generation unit 11d generates information for the terminal device 3 to display the knowledge network shown in FIG. 3. The knowledge network generation unit 11d places the keyword at the highest level (e.g., the first level) from among the multiple keywords extracted from the keyword DB 12b at the most central position, and places one or more keywords related to this keyword outward in hierarchical order. At this time, the knowledge network generation unit 11d may limit the keywords to be placed, for example, to the top four levels. In this case, keywords belonging to the fourth level or lower levels are displayed by the terminal device 3 when a detailed display request is given by the user.

The knowledge network generation unit 11d indicates relationships between multiple keywords that are appropriately arranged on the background image by connecting related keywords with straight lines. The knowledge network generation unit 11d transmits information such as the coordinates at which each keyword is arranged, the coordinates of the lines connecting the keywords, and the size at which each keyword is displayed to the terminal device 3 as information for displaying the knowledge network. The terminal device 3 that receives this information can reproduce the knowledge network generated by the knowledge network generation unit 11d of the server device 1, and display it on the display unit to present it to the user.

The editing processing unit 11e performs processing to accept editing operations by the user on the knowledge network displayed on the terminal device 3. In the information processing system according to this embodiment, the user can edit the knowledge network displayed on the terminal device 3, for example, by adding keywords, deleting keywords, changing keywords, moving the display position of keywords, changing the relevance between keywords, or changing the ranking of keywords. The terminal device 3 accepts editing operations by the user and transmits them to the server device 1, and in response to these editing operations, the editing processing unit 11e of the server device 1 modifies the knowledge network and transmits information for displaying the modified knowledge network to the terminal device 3. The editing processing unit 11e also stores the history of editing by the user in the history storage unit 12c.

The database update unit 11f performs a process of updating the information stored in the keyword DB 12b based on the history information stored in the history storage unit 12c by the editing processing unit 11e. For example, when a user performs editing to add a keyword not included in the keyword DB 12b to the knowledge network, the database update unit 11f can add this keyword to the keyword DB 12b. Also, for example, when an editing operation such as moving the display position of a keyword or changing the association between keywords is performed, the database update unit 11f can change the hierarchy in which these keywords are stored or change the link between these keywords. Also, for example, when an editing operation to change the ranking of a keyword is performed, the database update unit 11f can increase or decrease the importance associated with this keyword. Note that the database update unit 11f may update the keyword DB 12b not in response to an editing operation performed by one user, but when, for example, more than a predetermined number of users perform similar editing operations.

FIG. 6 is a block diagram showing an example of the configuration of the terminal device 3 according to this embodiment. The terminal device 3 according to this embodiment is configured to include a processing unit 31, a storage unit 32, a communication unit 33, a display unit 34, and an operation unit 35. The terminal device 3 can be configured using a general-purpose information processing device such as a personal computer, a smartphone, or a tablet terminal device.

The processing unit 31 is configured using an arithmetic processing device such as a CPU or MPU, a ROM, and the like. The processing unit 31 reads out and executes a program 32a stored in the storage unit 32, thereby performing various processes such as accepting input of target information by a user, displaying a knowledge network, and accepting editing operations on the knowledge network.

The storage unit 32 is configured using a non-volatile memory element such as a flash memory or a storage device such as a hard disk. The storage unit 32 stores various programs executed by the processing unit 31 and various data necessary for the processing of the processing unit 31. In this embodiment, the storage unit 32 stores the program 32a executed by the processing unit 31. In this embodiment, the program 32a is distributed by a remote server device or the like, and the terminal device 3 acquires it through communication and stores it in the storage unit 32. However, the program 32a may be written to the storage unit 32, for example, during the manufacturing stage of the terminal device 3. For example, the program 32a may be read by the terminal device 3 from a recording medium 99 such as a memory card or an optical disk and stored in the storage unit 32. For example, the program 32a may be read from the recording medium 99 by a writing device and written to the storage unit 32 of the terminal device 3. The program 32a may be provided in the form of distribution via a network, or in the form of being recorded on the recording medium 99.

The communication unit 33 communicates with various devices via a wired or wireless network N, which may include a mobile phone communication network, a wireless LAN, the Internet, etc. In this embodiment, the communication unit 33 communicates with the server device 1 via the network N. The communication unit 33 transmits data provided by the processing unit 31 to other devices, and provides data received from other devices to the processing unit 31.

The display unit 34 is configured using a liquid crystal display or the like, and displays various images, characters, etc. based on the processing of the processing unit 31. The operation unit 35 accepts user operations and notifies the processing unit 31 of the accepted operations. For example, the operation unit 35 accepts user operations through an input device such as a mechanical button or a touch panel provided on the surface of the display unit 34. Furthermore, for example, the operation unit 35 may be an input device such as a mouse and a keyboard, and these input devices may be configured to be removable from the terminal device 3.

In addition, in the terminal device 3 according to this embodiment, the processing unit 31 reads and executes the program 32a stored in the storage unit 32, whereby the target information acquisition unit 31a, the display processing unit 31b, the editing processing unit 31c, and the like are realized in the processing unit 31 as software-like functional units. Note that the program 32a may be a program dedicated to the information processing system according to this embodiment, or may be a general-purpose program such as an internet browser or web browser.

The objective information acquisition unit 31a performs processing to acquire information related to the user's objective. In this embodiment, the objective information acquisition unit 31a acquires objective information by displaying questions for the user on the display unit 34 and accepting input of answers to these questions from the user via the operation unit 35. The objective information acquisition unit 31a receives questions sent from the server device 1 and displays the received questions on the display unit 34. The objective information acquisition unit 31a accepts answers to the displayed questions from the user and transmits information such as natural language sentences accepted as answers to the server device 1 as objective information.

The display processing unit 31b performs processing to display various information on the display unit 34. In this embodiment, the display processing unit 31b displays the objective input screen shown in FIG. 2 and the knowledge network shown in FIG. 3 based on the information received from the server device 1.

The editing processing unit 31c accepts operations performed on the operation unit 35, thereby performing a process of accepting editing operations by the user for the knowledge network displayed on the display unit 34. The editing processing unit 31c accepts editing operations such as adding a keyword to the knowledge network, deleting a keyword, changing a keyword, moving the display position of a keyword, changing the relevance between keywords, or changing the ranking of keywords. The editing processing unit 31c transmits information indicating the content of the editing operation accepted from the user to the server device 1.

Decision Support Processing
7 is a flowchart showing an example of the procedure of the decision support process performed by the server device 1 according to the present embodiment. The objective information acquisition unit 11a of the processing unit 11 of the server device 1 according to the present embodiment appropriately selects one question from a plurality of questions prepared in advance, and transmits information on the selected question to the terminal device 3 (step S1). At this time, the objective information acquisition unit 11a may select a question from the plurality of questions in a predetermined order, for example, or may select the next question according to information input by the user, or may select a question in any other appropriate manner.

The target information acquisition unit 11a acquires target information including a response to the question sent in step S1 in the form of a sentence in natural language by the user from the terminal device 3 (step S2). The target information acquisition unit 11a performs an interpretation process on the sentence in natural language by the target information acquired in step S2 (step S3). Based on the result of the interpretation process in step S3, the target information acquisition unit 11a determines whether acquisition of the target information is complete depending on, for example, whether the user has entered a keyword to end input of the target information (step S4). If input of the target information is not complete (S4: NO), the target information acquisition unit 11a returns the process to step S1, selects and sends the next question based on, for example, the result of the interpretation process in step S3, and repeats the above-mentioned process.

When acquisition of the target information is complete (S4: YES), the component extraction unit 11b of the processing unit 11 extracts components contained in the target information from the user (step S5). At this time, the component extraction unit 11b can extract nouns, verbs, etc. by, for example, performing morphological analysis on natural language sentences, etc. contained in the target information, and treat the extracted nouns, verbs, etc. as components. At this time, the component extraction unit 11b also calculates the occurrence frequency of each extracted component, and calculates the importance (weight) of each component based on the calculated occurrence frequency. The component extraction unit 11b also ranks the multiple components extracted from the target information based on the calculated importance.

The keyword extraction unit 11c of the processing unit 11 performs a process of extracting keywords from the keyword DB 12b based on the components extracted in step S5 (step S6). At this time, the keyword extraction unit 11c extracts a predetermined number of components that are ranked, for example, from the multiple components. The keyword extraction unit 11c extracts keywords that match or are similar to these components from the keyword DB 12b. Furthermore, the keyword extraction unit 11c extracts one or more other keywords associated with the extracted keyword from the keyword DB 12b.

The knowledge network generation unit 11d of the processing unit 11 performs processing to generate a knowledge network based on the multiple keywords extracted in step S6. At this time, the knowledge network generation unit 11d places the highest hierarchical keyword from the multiple extracted keywords at the center, for example, on a background image prepared in advance, and places one or more keywords related to this keyword outward in hierarchical order. The knowledge network generation unit 11d generates a knowledge network by connecting related keywords with straight lines for the multiple placed keywords. The knowledge network generation unit 11d transmits information about the generated knowledge network to the terminal device 3 (step S8).

The processing unit 11 determines whether or not the knowledge network has been edited based on whether or not information regarding editing has been received from the terminal device 3 that displayed the knowledge network transmitted in step S8 (step S9). If an editing operation has been performed on the knowledge network (S9: YES), the editing processing unit 11e of the processing unit 11 stores the history of the editing operation performed by the user in the history storage unit 12c based on the information received from the terminal device 3 (step S10). The editing processing unit 11e updates the knowledge network in accordance with the editing operation performed by the user (step S11). The editing processing unit 11e transmits the knowledge network updated in step S11 to the terminal device 3 (step S12) and returns the process to step S9.

If no editing operation has been performed on the knowledge network (S9: NO), the processing unit 11 determines whether or not the process should be terminated based on, for example, whether or not the user has performed an operation to terminate on the terminal device 3 (step S13). If it is determined that the process should not be terminated (S13: NO), the processing unit 11 returns the process to step S9. If it is determined that the process should be terminated (S13: YES), the processing unit 11 terminates the decision support process.

In the information processing system according to this embodiment, the server device 1 acquires the objective information input by the user on the terminal device 3, and the terminal device 3 displays the knowledge network created by the server device 1 based on the acquired objective information. An example of the knowledge network displayed by the terminal device 3 is as shown in FIG. 3. In the knowledge network illustrated in FIG. 3, for example, keywords such as "gas type", "plasma", "similar process", "trade-off", and "boeing" are represented three-dimensionally. A keyword represented three-dimensionally indicates that many lower-level keywords are associated with it. A user can display a detailed knowledge network by performing an operation to select a keyword represented three-dimensionally in the knowledge network.

FIGS. 8 to 10 are schematic diagrams showing an example of a detailed display of a knowledge network. The example shown in FIG. 8 is a detailed display when "similar process" is selected by the user in the knowledge network shown in FIG. 2. The server device 1, which has received a request for detailed display of the "similar process" of the knowledge network from the terminal device 3, searches for the specified "similar process" from the keyword DB 12b. The server device 1 extracts keywords that are associated with this "similar process" and belong to a lower hierarchy than the "similar process" from the keyword DB 12b. Note that at this time, the server device 1 may limit the keywords extracted from the keyword DB 12b, for example, to a hierarchy lower than the "similar process". Based on one or more keywords extracted from the keyword DB 12b, the server device 1 generates a detailed network in which the "similar process" is the top level (root) and one or more keywords associated with it are linked in a so-called tree structure. The server device 1 transmits the generated detailed network to the terminal device 3, which receives it and displays the detailed network shown in FIG. 8.

In the example shown in FIG. 8, the terminal device 3 displays a detailed network in a tree structure with "similar processes" at the top, expanded left and right on the display unit 34. In the detailed network in this example, three keywords, "recipe," "knob," and "light emission," are displayed as keywords that belong to one level below "similar processes." Note that the detailed network may also display keywords at levels lower than these three keywords, but these are not shown in this example.

In addition, in the detailed network displayed in this example, multiple keywords at the same level are displayed vertically, with more important keywords at the top and less important keywords at the bottom. The more important the keyword, the larger the rectangular frame surrounding the keyword, and the less important it is, the smaller it is. In other words, the display position and size of keywords at the same level in the detailed network indicate the ranking according to the importance of the keyword. In the example shown in Figure 8, "recipe" is ranked first, "knob" is ranked second, and "light emission" is ranked third.

For example, when the terminal device 3 is displaying a detailed network, the user can perform an editing operation to change the ranking of keywords. When the detailed network shown in FIG. 8 is displayed, for example, the user uses the mouse or the like of the operation unit 35 to drag the keyword "recipe" and drop "recipe" into the area between "knob" and "light emission". This allows the user to swap the rankings of "recipe" and "knob". FIG. 9 shows an example of the display of a knowledge network after swapping the rankings. As a result of swapping the rankings, the size of the rectangular frame surrounding "knob" and "recipe" is also changed. Furthermore, in response to the change in the positions of "recipe" and "knob", the display positions of the lower hierarchical keywords associated with "recipe" and the lower hierarchical keywords associated with "knob" are also changed.

The user can also perform editing operations to add keywords to the knowledge network (detailed network) displayed by the terminal device 3. For example, the user can display a menu by right-clicking the mouse on the operation unit 35 and select an item for adding a keyword included in this menu, thereby adding a keyword to the knowledge network. The terminal device 3 displays a text box for inputting a keyword at the position in the knowledge network where the mouse was right-clicked, and accepts input of a character string into this text box, thereby accepting a new keyword from the user. The newly added keyword is added, together with a rectangular frame surrounding it, to the knowledge network at the position where the mouse was right-clicked.

Furthermore, the terminal device 3 accepts an operation from the user to associate one or more keywords already included in the knowledge network with the newly added keyword. The terminal device 3 accepts an operation from the user to connect two keywords with a line, for example, based on a mouse click operation, and associates the two keywords connected by the line. The terminal device 3 transmits information regarding the keyword added by the user and its relevance to other keywords to the server device 1. The server device 1 receives this information from the terminal device 3, updates the knowledge network based on the received information, and transmits the updated knowledge network to the terminal device 3, thereby displaying the updated knowledge network on the terminal device 3. The server device 1 also stores the information received from the terminal device 3 in the history storage unit 12c.

Figure 10 shows an example in which the user has added a new keyword, "related papers", to the detailed network shown in Figure 9. In this example, the user has added the keyword "related papers" to a position below "light emission". As a result, in the detailed network, the keyword "related papers" is added as a keyword at the same hierarchical level as "knobs", "recipe", and "light emission", and the size of the rectangular frame is made smaller than that of "light emission" as it is a keyword with a lower ranking than "light emission". Also, in this example, the user has connected "related papers" and "similar processes" with a straight line, and the two keywords "similar processes" and "related papers" are now associated.

In this example, the "related papers" keyword added by the user is a keyword that has already been registered in keyword DB 12b and is a keyword that has not been associated with "similar processes." When such a keyword is added, the server device 1 extracts lower-level keywords that are associated with the "related papers" keyword and adds them to the detailed network together with the "related papers" keyword. When the added keyword is not registered in keyword DB 12b, the server device 1 need only add this keyword to the detailed network.

In this example, an example of adding keywords to a detailed network has been described, but adding keywords is not limited to detailed networks. For example, keywords may be added to the knowledge network shown in Figure 3 in a similar manner.

When a user performs an editing operation on the knowledge network as described above, the server device 1 stores and accumulates information on the editing history in the history storage unit 12c. In the information processing system according to this embodiment, the server device 1 performs an update process of the keyword DB 12b based on the information stored in the history storage unit 12c at a predetermined cycle, for example, once a month or once a week. In the update process, the server device 1 reads out information stored in the history storage unit 12c during the period from the previous update process to the current update process, for example, and counts the number of times similar editing operations were performed. For example, the number of times keyword A was added, the number of times keyword B was lowered in rank, the number of times keyword C was deleted, etc. The server device 1 extracts editing operations whose counted number exceeds a predetermined number (for example, 10 times), and updates the keyword DB 12b according to the extracted editing operations.

For example, if keyword A has been added more than a predetermined number of times, the server device 1 adds this keyword A to the keyword DB 12b. At this time, the server device 1 further counts the hierarchies to which keyword A has been added, and adds this keyword A to the hierarchies to which keyword A has been added the most number of times. The server device 1 also counts other keywords in higher hierarchies associated with keyword A, and associates this keyword A with other keywords in higher hierarchies with which it has been associated the most number of times. The server device 1 also counts other keywords in lower hierarchies associated with keyword A, and associates this keyword A with one or more keywords in lower hierarchies with which it has been associated more than a predetermined number of times (for example, five times).

For example, if the number of times that the ranking of keyword B has been lowered exceeds a predetermined number of times, the server device 1 reduces the importance attached to this keyword B. At this time, the server device 1 may determine the amount of reduction in importance depending on the number of times that the ranking of keyword B has been lowered.

For example, if the number of times keyword C has been deleted exceeds a predetermined number, the server device 1 deletes this keyword C from the keyword DB 12b. However, in this case, the server device 1 may reduce the importance of keyword C instead of deleting it from the keyword DB 12b. It is preferable that the amount of reduction in importance is greater than the amount of reduction when an editing operation that lowers the ranking is performed as described above. Furthermore, when deleting keyword C from the keyword DB 12b, the server device 1 may also delete other keywords that are associated only with this keyword C from the keyword DB 12b.

FIG. 11 is a flow chart showing an example of the procedure for updating the keyword DB 12b performed by the server device 1 according to this embodiment. The database update unit 11f of the processing unit 11 of the server device 1 according to this embodiment determines whether it is time to perform an update, for example, once a month or once a week (step S21). If it is not time to perform an update (S21: NO), the database update unit 11f waits until it is time to perform an update.

If it is time to perform an update (S21: YES), the database update unit 11f reads out the history information stored in the history memory unit 12c that has been accumulated since the previous update (step S22). The database update unit 11f counts the number of times the same or similar editing operations have been performed for the history information read out in step S22 (step S23). Based on the counting result in step S23, the database update unit 11f extracts editing operations that have been performed more than a predetermined number of times (step S24). Based on the editing operation extracted in step S24, the database update unit 11f updates the information on the keyword related to this editing operation stored in the keyword DB 12b (step S25), and ends the process.

<Summary>
In the information processing system according to the present embodiment having the above configuration, the server device 1 acquires purpose information input by a user in natural language via the terminal device 3, and extracts a plurality of components from the acquired purpose information. The server device 1 extracts a plurality of keywords from the keyword DB 12b that stores a plurality of keywords and relationships between the keywords, based on the extracted plurality of components. The server device 1 outputs a knowledge network indicating the extracted plurality of keywords and the relationships between the plurality of keywords to the terminal device 3. This is expected to enable the user to understand, based on the knowledge network displayed on the terminal device 3, what kind of knowledge is necessary for his/her purpose, and what kind of correspondence there is between the plurality of pieces of knowledge. This is expected to enable the information processing system according to the present embodiment to support the user in making decisions regarding the purpose.

In addition, in the information processing system according to this embodiment, the server device 1 displays questions on the terminal device 3 and receives the user's responses to the questions at the terminal device 3, thereby acquiring objective information. The server device 1 may also display multiple questions in stages and receive the user's responses to each question. As a result, the server device 1 can be expected to obtain detailed information about the objective the user is trying to achieve.

Furthermore, in the information processing system according to this embodiment, the server device 1 weights the components extracted from the target information. The server device 1 extracts multiple keywords from the keyword DB 12b based on the weight of each component. The server device 1 ranks the multiple keywords based on the importance assigned to the extracted keywords. This allows the user to rank and understand the importance of multiple keywords included in the knowledge network.

In addition, in the information processing system according to this embodiment, multiple keywords are displayed on the terminal device 3 in a manner according to their ranking, editing operations to change the ranking are accepted from the user, and the displayed knowledge network is updated according to the accepted change operation. This allows the user to easily grasp the ranking of multiple keywords, and for example, by performing an operation to change the ranking of the keywords, the user can think about which keywords are important for his or her own purpose.

In addition, the information processing system according to this embodiment accepts input of additional keywords from the user to be added to the knowledge network displayed on the terminal device 3. The server device 1 updates the knowledge network to indicate the relationship between the multiple keywords included in the knowledge network and the added keywords, and outputs the updated knowledge network to the terminal device 3. This allows the user to think about keywords that are missing from the displayed knowledge network, and to add the missing keywords to the knowledge network to update it.

In addition, in the information processing system according to this embodiment, when a keyword is added by a user, the server device 1 stores the input history of the added keyword in the history storage unit 12c, and updates the keyword DB 12b by adding the keyword based on the stored history. This allows the server device 1 to reflect the keyword added by one user in the knowledge network for other users.

In addition, in the information processing system according to this embodiment, the server device 1 arranges multiple keywords on a two-dimensional background image, generates a diagram of a knowledge network in which related keywords are wired together, and outputs the diagram to the terminal device 3. In this embodiment, the server device 1 arranges multiple keywords on a two-dimensional plane, but this is not limited to this, and it is also possible to generate a diagram of a knowledge network in which multiple keywords are arranged in a (virtual) three-dimensional space.

Furthermore, in the information processing system according to this embodiment, the multiple keywords stored in the keyword DB 12b are hierarchically related to one another. The server device 1 causes the terminal device 3 to display a knowledge network showing the higher-level keywords and relatedness. The server device 1 accepts the selection of a keyword included in the displayed knowledge network, and causes the terminal device 3 to display a detailed network showing the keywords and relatedness at a lower level than the selected keyword. In this way, the server device 1 can present the user with many keywords related to the user's objectives, which is expected to assist the user in making decisions regarding the objectives.

In this embodiment, the server device 1 presents information to the user in the form of a network in which multiple keywords are connected by lines, but this is not limited to this. For example, the server device 1 may present multiple keywords related to the user's purpose in a list display, or may present multiple keywords to the user in various other ways. Also, the knowledge network and detailed network illustrated in this embodiment are merely examples and are not limited to this.

In addition, the information processing system according to this embodiment is configured such that a user uses a terminal device 3 to access the server device 1, and the server device 1 displays information such as a knowledge network on the display unit 34 of the terminal device 3, but this is not limited to the above. The information processing system may be configured such that, for example, a user directly operates the server device 1 (or an equivalent information processing device), and information such as a knowledge network may be displayed on the display unit of the server device 1 or on a display device connected to the server device 1.

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

　The matters described in each embodiment can be combined with each other. Furthermore, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations, regardless of the citation format. Furthermore, the claims use a format in which a claim cites two or more other claims (multi-claim format), but this is not limited to this. They may also be written in a format in which a multiple claim cites at least one other multiple claim (multi-multi-claim).

1. Server device (information processing device, computer)
3 Terminal device 11 Processing unit 11a Target information acquisition unit 11b Component extraction unit 11c Keyword extraction unit 11d Knowledge network generation unit 11e Editing processing unit 11f Database update unit 12 Storage unit 12a Program (computer program)
12b Keyword DB
12c History storage unit 13 Communication unit 31 Processing unit 31a Object information acquisition unit 31b Display processing unit 31c Edit processing unit 32 Storage unit 32a Program 33 Communication unit 34 Display unit 35 Operation unit 98, 99 Recording medium N Network

Claims

An information processing device,
Obtain the purpose information entered by the user,
Extracting a plurality of components from the acquired target information;
Extracting a plurality of keywords from a database storing a plurality of keywords and relationships between the keywords based on the extracted plurality of components;
outputting the extracted keywords and the relationships between the extracted keywords;
Information processing methods.
acquiring the target information according to a sentence input by a user in a natural language or according to a plurality of items selected by the user;
The information processing method according to claim 1 .
outputting questions and receiving input of responses to the questions, thereby acquiring the target information;
The information processing method according to claim 1 .
outputting a plurality of questions in a stepwise manner and receiving responses to each of the questions, thereby acquiring the target information;
The information processing method according to claim 3.
weighting the extracted components;
Extracting a plurality of keywords from the database based on the weights;
Ranking the extracted keywords;
The information processing method according to claim 1 .
outputting the plurality of keywords in a manner according to their ranking;
Accepting an operation to change the order,
Update the output according to the operation received,
The information processing method according to claim 5.
accepting input of an additional keyword to be added to the outputted plurality of keywords;
outputting a relationship between the plurality of keywords and the additional keyword;
The information processing method according to claim 1 .
storing a history of input of the additional keyword;
adding keywords to the database based on the stored history;
The information processing method according to claim 7.
arranging the plurality of keywords on a two-dimensional plane or in a three-dimensional space, and outputting a network diagram in which related keywords are wired;
The information processing method according to claim 1 .
The plurality of keywords are related to each other hierarchically,
outputting a first network diagram showing hierarchically higher keywords and relevance among the plurality of keywords;
Accepting a selection of a keyword from the first network diagram;
outputting a second network diagram showing keywords and associations hierarchically lower than the received keyword;
The information processing method according to claim 9.
On the computer,
Obtain the purpose information entered by the user,
Extracting a plurality of components from the acquired target information;
Extracting a plurality of keywords from a database storing a plurality of keywords and relationships between the keywords based on the extracted plurality of components;
and outputting the extracted plurality of keywords and relationships between the extracted plurality of keywords.
an acquisition unit for acquiring target information input by a user;
a first extraction unit that extracts a plurality of components from the acquired target information;
a second extraction unit that extracts a plurality of keywords from a database that stores a plurality of keywords and relationships between the keywords, based on the extracted plurality of components;
and an output unit that outputs the extracted plurality of keywords and relationships between the plurality of keywords.