JP7835533B2 - Information processing system, information processing device, and control program - Google Patents

Description

This invention relates to an information processing system, an information processing device, and a control program.

In recent years, there has been a surge in the development of technologies that help users better understand the content of documents. Such technologies include, for example, the creation of document summaries, translation to make documents easier for readers to understand, and the addition of explanations (annotations) for specialized terminology (keywords) contained in documents.

For example, Patent Document 1 discloses a technology that provides explanations of technical terms included in a speaker's presentation, tailored to the specific area of expertise of each listener (user).

Japanese Patent Publication No. 2017-97488

However, while the technology described in Patent Document 1 provides explanations of technical terms based on each listener's area of expertise, it does not take into account each listener's knowledge level or individual characteristics. Therefore, there is a possibility that necessary explanations may not be provided to listeners, or conversely, that unnecessary explanations may be provided. For example, even if a technical term belongs to a listener's area of expertise, the listener may not necessarily possess knowledge of that term. If a technical term belongs to a listener's area of expertise, even if the listener does not possess knowledge of that term, it may be determined that the listener does, and therefore no explanation of the technical term is provided. However, this makes it difficult for listeners to understand the content of the lecture because they lack knowledge of the technical term.

On the other hand, even if a technical term does not belong to the listener's field of expertise, the listener may still possess knowledge of that term. In this case, because the technical term does not belong to the listener's field of expertise, it is determined that the listener does not possess knowledge of the term, and an explanation of the term is provided. However, this means that the listener is provided with an explanation even though they already possess knowledge of the term. As a result, the listener may feel burdened by the increased amount of information they need to confirm, and may have difficulty finding explanations of other technical terms they need.

This invention has been made in view of the above circumstances, and aims to provide an information processing system, an information processing device, and a control program that can add annotations to keywords in input data according to the user's knowledge.

The above objective of the present invention is achieved by the following means.

(1) An information processing system comprising: a data acquisition unit; an information storage unit that stores knowledge information for each user, which includes multiple elements corresponding to multiple pieces of information that constitute knowledge; an information comparison unit that compares keyword information related to keywords extracted from data acquired as input data by the data acquisition unit with knowledge information associated with the user; and an annotation addition unit that adds annotations to the keywords in the input data based on the comparison results by the information comparison unit , wherein the information comparison unit determines whether the keywords in the keyword information are included in the knowledge information , or whether the keyword information and the knowledge information are a match or similar , and the annotation addition unit adds annotations to the keywords among the multiple keywords that are not included in the user's knowledge information if the number of keywords among the multiple keywords in the keyword information that are included in the user's knowledge information, or the number of keywords that are a match or similar to the elements of the knowledge information, is less than a threshold.

(2) The information processing system according to (1) above , wherein the keyword information is information that includes multiple keywords , and the information comparison unit determines whether the keyword information and the knowledge information match or are similar by calculating a degree of similarity based on the vector of keywords included in the keyword information and the vector of elements included in the knowledge information .

(3) The information processing system described in ( 1 ) above , wherein the keyword information is information represented by a network including a plurality of nodes each representing a plurality of the keyword and links defining the relationships between the nodes .

(4) The information processing system according to (3) above, wherein the knowledge information is networked information representing the relationships between elements corresponding to multiple pieces of information constituting knowledge, and the information comparison unit determines whether the keyword information and the knowledge information match or are similar by calculating a similarity between the network of keyword information and the network of knowledge information based on the node matching rate and the distance between nodes .

(5) The information processing system according to (4) above, further comprising a subset extraction unit that extracts a subset of the user's knowledge information based on the keyword, wherein the subset extraction unit searches the user's knowledge information for nodes tagged with the keyword, extracts a range including nodes tagged with the keyword as a subset, and the information comparison unit compares the keyword information with the subset .

( 6 ) The information processing system according to any one of (1) to ( 5 ) above, further comprising an output generation unit that generates output data based on annotations added to the keyword by the annotation addition unit.

( 7 ) The information processing system according to any one of (1) to ( 6 ) above, wherein the data acquisition unit acquires the data uploaded by the user as the input data.

( 8 ) The information processing system according to any one of (1) to ( 6) above, further comprising a mail monitoring unit that monitors the user's received mail, and the data acquisition unit acquires the received mail as input data when the received mail satisfies predetermined conditions.

( 9 ) The information processing system according to any one of (1) to (6) above, further comprising: a file monitoring unit that monitors files being moved to a folder where the user's files are saved; and a file reading unit that reads files saved in the folder, wherein the data acquisition unit acquires the files in the folder, or the result of reading the files , as input data when predetermined conditions are met.

(10) An information processing system having a data server , a knowledge server and an information processing device, wherein the data server has a data acquisition unit and a keyword information generation unit that generates keyword information from data acquired as input data by the data acquisition unit, the knowledge server has an information storage unit that stores knowledge information for each user, which includes a plurality of elements corresponding to a plurality of pieces of information constituting knowledge, stores the keyword information generated by the keyword information generation unit and transmits the keyword information and the knowledge information to the information processing device, and the information processing device receives the acquired keyword information and the knowledge information associated with the user. An information processing system comprising: an information comparison unit that compares with; and an annotation unit that adds annotations to the keywords of the input data based on the comparison results by the information comparison unit, wherein the information comparison unit determines whether the keywords of the keyword information are included in the knowledge information, or whether the keyword information and the knowledge information match or are similar; and the annotation unit adds annotations to the keywords among the multiple keywords of the keyword information that are not included in the user's knowledge information if the number of keywords included in the user's knowledge information, or the number of keywords that match or are similar to the elements of the knowledge information, is less than a threshold;

( 11 ) The information processing system according to ( 10) above, wherein the annotation addition unit adjusts the threshold according to the attributes of the input data.

(12) The information storage unit is provided by the knowledge server and stores the knowledge information for each of the multiple users, according to any one of (1) to (9) above, the information processing system.

( 13 ) The information processing system according to (11) above, wherein the annotation unit performs simple annotation, which adds an annotation containing only text to the keyword, or detailed annotation, which adds an annotation containing at least a figure to the keyword.

( 14 ) The information processing system according to (13) above, wherein the annotation unit calculates the difficulty level of the input data based on the threshold and the comparison result, and performs the detailed annotation if the difficulty level is high, and performs the simplified annotation if the difficulty level is low .

( 15 ) The information processing system according to (6) above, wherein the output data includes the input data and annotations attached to the keywords.

( 16 ) The information processing system according to any one of (1) to (15) above, wherein the knowledge information is generated based on at least one of the following: data relating to social networking services ( SNS ) used by the user, data relating to the web viewed by the user, and document data created by the user.

(17) An information processing device comprising: a keyword information acquisition unit that acquires keyword information from data acquired as input data; a knowledge information acquisition unit that acquires knowledge information for a specific user that includes multiple elements corresponding to multiple pieces of information constituting knowledge; an information comparison unit that compares the keyword information with the knowledge information associated with the specific user; and an annotation addition unit that adds annotations to the keywords of the input data based on the comparison results by the information comparison unit, wherein the information comparison unit determines whether the keywords of the keyword information are included in the knowledge information, or whether the keyword information and the knowledge information match or are similar; and the annotation addition unit adds annotations to the keywords among the multiple keywords that are not included in the knowledge information of the specific user if the number of keywords among the multiple keywords of the keyword information that are included in the knowledge information of the specific user, or the number of keywords that match or are similar to the elements of the knowledge information, is less than a threshold.

(18) A control program for causing a computer to execute a process comprising: a keyword information acquisition step (a) of acquiring keyword information from data acquired as input data; a knowledge information acquisition step (b) of acquiring knowledge information for a specific user that includes a plurality of elements corresponding to a plurality of pieces of information constituting knowledge; a comparison step (c) of comparing the keyword information with the knowledge information associated with the specific user; and an annotation addition step (d) of adding annotations to the keywords of the input data based on the comparison result in the comparison step (c), wherein in the comparison step (c), it is determined whether the keywords of the keyword information are included in the knowledge information, or whether the keyword information and the knowledge information match or are similar; and in the annotation addition step (d), if the number of keywords among the plurality of keywords of the keyword information that are included in the knowledge information of the specific user, or the number of keywords that match or are similar to the elements of the knowledge information, is less than a threshold, annotations are added to the keywords among the plurality of keywords that are not included in the knowledge information of the specific user.

According to this invention, annotations are added to the keywords in the input data based on a comparison between the keyword information of the input data and the user's knowledge information. Therefore, annotations are added to the keywords in the input data according to the user's knowledge.

This is a schematic block diagram illustrating the configuration of an information processing system according to the first embodiment. Figure 1 is a schematic block diagram illustrating the hardware configuration of the data server shown. This is a schematic diagram illustrating the generation of keyword information in a data server. Figure 1 is a network diagram illustrating a portion of the knowledge information stored in the knowledge database shown. Figure 1 is a schematic block diagram illustrating the configuration of the information processing device shown. Figure 5 is a functional block diagram illustrating the main functions of the information processing device shown. This is a conceptual diagram illustrating the similarity between keyword information and knowledge information. Figure 1 is a schematic block diagram illustrating the hardware configuration of the client terminal shown. This is a sequence chart illustrating a general processing procedure for the control method of the information processing system according to the first embodiment. Figure 1 is a schematic diagram showing an example of how output data from the information processing device is displayed. Figure 1 is a schematic diagram showing another example of how to display the output data of the information processing device shown in Figure 1. Figure 9 is a flowchart illustrating the general processing procedure for step S103 of the sequence chart. This is a schematic diagram illustrating the comparison between keyword information and knowledge information. This is a conceptual diagram illustrating a method for determining whether or not to add annotations to keywords in the examples and comparative examples. This is a conceptual diagram illustrating a method for determining whether or not to add annotations to keywords in the examples and comparative examples. This is a sequence chart illustrating the acquisition of input data in the second embodiment.

The embodiments of the present invention will be described below with reference to the attached drawings. In the description of the drawings, identical elements are denoted by the same reference numerals, and redundant explanations are omitted. Furthermore, the dimensional ratios in the drawings are exaggerated for illustrative purposes and may differ from actual ratios.

(First embodiment)
<Information Processing System 100>
Figure 1 is a schematic block diagram illustrating the configuration of an information processing system 100 according to the first embodiment. The information processing system 100 includes a data server 200, a knowledge server 300, and an information processing device 400, which are interconnected via a communication network 101, including, for example, a LAN (Local Area Network), a WAN (Wide Area Network), or the Internet. As will be described later, the data server 200 is responsible for generating keyword information based on keywords contained in the input data and providing it to the information processing device 400, and the knowledge server 300 is responsible for providing knowledge information stored for each user to the information processing device 400. The information processing device 400 adds annotations to the keywords based on the provided keyword information and knowledge information.

Furthermore, the information processing system 100 is connected to multiple client terminals A501 to C503 via a communication network 101. While Figure 1 illustrates a case where three client terminals A501 to C503 are connected to the information processing system 100, the number of client terminals may be more or less than three. The information processing system 100 may also include client terminals.

<Data Server 200>
Figure 2 is a schematic block diagram illustrating the hardware configuration of the data server 200 shown in Figure 1, and Figure 3 is a schematic diagram illustrating the generation of keyword information in the data server 200.

The data server 200 functions as a server (computer). As shown in Figure 2, the data server 200 includes a CPU (Central Processing Unit) 210, RAM (Random Access Memory) 220, ROM (Read Only Memory) 230, auxiliary storage unit 240, and communication unit 250, etc.

The CPU 210 executes the OS (Operating System) and control programs for the data server 200, which are loaded into the RAM 220, and controls the operation of the data server 200. The control programs are pre-stored in the ROM 230 or auxiliary storage unit 240. The RAM 220 stores data temporarily generated by the processing of the CPU 210. The ROM 230 stores programs executed by the CPU 210, as well as data and parameters used for program execution.

The auxiliary storage unit 240 includes, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or flash memory.

The communication unit 250 has a communication device such as a network interface card (NIC) and transmits data to the knowledge server 300 and the information processing device 400 via the communication network 101.

As shown in Figure 3, the CPU 210 controls the communication unit 250 by executing a control program, thereby functioning as a data acquisition unit 261. In this embodiment, the data acquisition unit 261 acquires input data from, for example, a client terminal A501 used by user A. The input data may be a document containing at least one keyword (e.g., W1 to W9). This input data is also transmitted to the knowledge server 300.

Furthermore, the CPU 210 functions as a keyword information generation unit 262. The keyword information generation unit 262 extracts keywords from the input data using natural language processing and generates keyword information based on the extracted keywords. The generated keyword information is stored in the RAM 220 or the auxiliary storage unit 240.

More specifically, the keyword information generation unit 262 performs morphological analysis on the input data, dividing the sentence contained in the input data into parts of speech, and extracting, for example, words that appear repeatedly as keywords. The keyword information generation unit 262 also extracts the relationships between the extracted keywords and generates structured keyword information. The keyword information includes, for example, multiple keywords and the relationships (links) between those multiple keywords. For example, the keyword information can be represented as a network. In this case, each keyword can be represented as a node (shown as a circle), and the relationships between nodes can be represented by lines (links) connecting each node.

<Knowledge Server 300>
Figure 4 is a network diagram illustrating a portion of the knowledge information stored in the knowledge database 310 shown in Figure 1. Knowledge information is, for example, information that networks the relationships between elements corresponding to multiple pieces of information that constitute the knowledge possessed by a single user. The elements can be concepts represented by words, sentences, clauses, etc. In this embodiment, the knowledge information of a single user is modeled as a knowledge network (or semantic network) in which multiple concepts are linked according to their relationships. Knowledge information of multiple users is stored in the knowledge database 310. Knowledge information is generated or updated based on the usual work of a user on client terminal 501A (for example, the input data received from client terminal A501) and the usual work of another user on another client terminal, and is stored in the knowledge database 310.

The knowledge server 300 is a server (computer) that has a knowledge database 310 (see Figure 1). Since the knowledge server 300 has a hardware configuration similar to that of the data server 200, a detailed explanation is omitted. The knowledge database 310 is stored in an auxiliary storage unit (information storage unit).

Knowledge information includes multiple elements and the relationships (connections) between them. For example, as shown in the network diagram in Figure 4, in knowledge information, each element can be represented by a node (shown as a circle) 312, and the connections between nodes 312 can be represented by lines (links) 313 connecting each node 312. Figure 4 illustrates, for example, network 311, which is part of user A's knowledge information (knowledge network). For example, nodes N00, N01, N04, N08, and N09 correspond to the elements "hypertension," "guidelines," "treatment," "side effects," and "definition," and nodes N01, N04, N08, and N09 are connected to node N00. Relationships between nodes include superordinate/subordinate conceptual relationships (for example, the source node is a superordinate concept and the destination node is a subordinate concept), relationships where the destination node is an attribute of the source node, where the destination node is a required or optional attribute of the source node, and where the destination node is any possible value of the source node.

Furthermore, knowledge information may include information representing the relative importance of each element within a group of interconnected elements (hereinafter also referred to as an "element group"). Elements with higher importance compared to others may, for example, serve as the starting point or key point of the knowledge represented by the knowledge information. For instance, elements with many connections to other elements are considered relatively important. Therefore, the importance of an element can be judged by the degree of its connections to other elements. Alternatively, the knowledge database 310 may be configured to store the importance of elements numerically.

On a network diagram, nodes with many connections to other nodes are drawn larger than nodes with fewer connections. In the example shown in Figure 4, node N00, representing "hypertension," is drawn larger than other nodes because it has many connections to nodes N01, N04, N08, and N09. Furthermore, node N00 is the starting point of the knowledge represented by network 311.

Furthermore, knowledge information may also include information representing the strength of the connections between elements. The strength of these connections can also be stored numerically in the knowledge database 310. On the network diagram, for example, stronger connections are depicted with thicker lines.

Furthermore, knowledge information may include information about the field of knowledge. For example, each node in the knowledge information is associated with a tag indicating the field of knowledge, etc. For instance, the knowledge database 310 or a table stores information about the node number and the tag corresponding to that node number.

The knowledge database 310 stores knowledge information for multiple users, with each user having their own set of knowledge information. This knowledge information is generated or updated based on each user's usual activities on the client terminal and stored in the knowledge database 310. User activities on the client terminal include, for example, creating various documents using word processing software and email software, and viewing various documents and web pages. Knowledge information can be generated based on at least one of the following: document data created or reviewed by the user, data related to social networking services (SNS) used by the user, and data related to documents, websites, and emails viewed by the user.

Furthermore, considering that document creation generally requires deeper knowledge than simply viewing documents or web pages, it may be possible to rank the elements of knowledge information (knowledge) according to the type of work performed by the user. For example, knowledge gained through document creation could be given the highest rank, knowledge gained through document review could be given a medium rank, and knowledge gained through viewing web pages or emails could be given the lowest rank.

The knowledge server 300, in response to a request from the information processing device 400, reads the user's knowledge information from the auxiliary storage unit and transmits it to the information processing device 400.

In the example above, a knowledge network model of element sets was illustrated. However, it is also possible to model the network (semantic network) considering not only the presence or absence of connections between elements and the strength of those connections, but also the relationships between them (such as inclusion relationships).

<Information Processing Device 400>
The information processing device 400 functions as a server (computer). As shown in Figure 5, the information processing device 400 includes a CPU 410, RAM 420, ROM 430, auxiliary storage unit 440, and communication unit 450, etc. Since the information processing device 400 has a hardware configuration similar to that of the data server 200 and the knowledge server 300, a detailed explanation is omitted. In this embodiment, the auxiliary storage unit 440 stores information processing programs executed by the CPU 410. The auxiliary storage unit 440 also stores data related to definitions and explanations of terms in various fields for adding annotations to keywords. This data includes text, diagrams, photographs, videos, audio, etc.

Figure 6 is a functional block diagram illustrating the main functions of the CPU 410 of the information processing device 400 shown in Figure 5. In this embodiment, the CPU 410 performs the functions of a subset extraction unit 411, an information comparison unit 412, an annotation addition unit 413, and an output generation unit 414 by executing a control program for the information processing device 400.

The subset extraction unit 411 extracts a subset (subnetwork) from the knowledge network of a specific user (e.g., user A). More specifically, the subset extraction unit 411 acquires keyword information from the data server 200 and the knowledge information (knowledge network) of the specific user from the knowledge server 300, and extracts a subset from the knowledge information based on the keywords contained in the input data. The subset extraction unit 411 functions as both a keyword information acquisition unit and a knowledge information acquisition unit. The subset extraction unit 411 determines the range of the subset to be extracted from the knowledge information based on the tag information attached to each node. For example, in the example shown in Figure 4, since the input data contains the keyword "hypertension," the subset extraction unit 411 searches for nodes tagged with "medical field" in the user's knowledge information. As a result of the search, for example, a range including nodes tagged with "medical field" is extracted as a subset.

The information comparison unit 412 compares the keyword information with the knowledge information (subset) associated with a specific user (the user who entered the input data). More specifically, the information comparison unit 412 determines whether the keywords in the keyword information are included in the knowledge information of the specific user. Alternatively, the information comparison unit 412 determines whether the keyword information and the knowledge information match (or are similar). To determine whether the keyword information and the knowledge information match (or are similar), for example, the similarity score between the keyword information and the knowledge information is calculated. If the similarity score is above a predetermined value, it is determined to be similar; if it is below the predetermined value, it is determined to be dissimilar.

Similarity is an index that represents the degree of similarity between keyword information and knowledge information. As shown in Figure 7, similarity can be represented, for example, by the distance between the centers of two conceptual regions 800 (represented by a user's knowledge information) and a conceptual region 801 (represented by keyword information), or by the overlapping region 802 (represented by the two regions). That is, the smaller the distance between the centers of the two regions, the higher the similarity; the larger the distance, the lower the similarity. Similarly, the larger the overlapping region 802, the higher the similarity; the smaller the region 802, the lower the similarity. Region 802 is defined by the keyword groups and element groups included within it and the relationships between them. Note that regions 800, 801, and 802 are usually multidimensional regions, but in Figure 7, they are represented as two-dimensional regions for simplicity of explanation.

Furthermore, similarity can be calculated based on the vector of keywords included in the keyword information and the vector of elements included in the knowledge information. In this specification, these two vectors are referred to as the keyword information vector and the knowledge information vector, respectively. The keyword information vector may be the average or representative value of the vectors of multiple keywords included in the keyword information, while the knowledge information vector may be the average or representative value of the vectors of multiple elements included in the knowledge information.

More specifically, the information processing device 400 converts multiple keywords contained in keyword information into vector values, for example, using a method such as Word2Vec, calculates the average value of these vectors, and obtains the keyword information vector. Keywords with similar meanings, such as synonyms, will have similar vector values. Similarly, for knowledge information, multiple elements contained in the knowledge information are converted into vector values, calculates the average value of these vectors, and obtains the knowledge information vector. Elements with similar meanings, such as synonyms, will have similar vector values. The information processing device 400 then calculates the cosine similarity based on the keyword information vector and the knowledge information vector. The cosine similarity between the keyword information vector and the knowledge information vector is also called the distance between the two vectors.

Furthermore, similarity can be calculated between a keyword information network and a knowledge information network based on node agreement rates and inter-node distances. For example, in the network exemplified in Figure 4, where "hypertension" is the root keyword, knowledge far from the root is unimportant, so subsequent levels are ignored. Using the nodes from the root to a predetermined level, the node agreement rate and inter-node distances can be calculated for nodes up to the nth level.

Furthermore, the information comparison unit 412 can assign weights to specific keywords or groups of keywords in the keyword information when calculating similarity. This allows the contribution of specific keywords or groups of keywords to be greater than that of other keywords or groups of keywords in the similarity calculation. For example, if the input data contains the keyword "treatment," assigning a weight to the keyword "treatment" makes its contribution to the similarity calculation greater than that of other keywords. This allows for an approximate comparison between keyword information and knowledge information. For instance, if the comparison point is "treatment" for "hypertension," and other keywords are not particularly important, similarity can be calculated without being constrained by minor details. Therefore, the comparison between keyword information and knowledge information can be performed more efficiently.

Thus, the information comparison unit 412, in comparing the keyword information of the input data with the knowledge information of a specific user, determines whether the keyword information is included in the knowledge information, or whether the keyword information and the knowledge information match (or are similar). When determining whether the keyword is included in the knowledge information, it can accurately determine whether the user has knowledge about the keyword in the input data. However, if the knowledge information does not include the keyword in the input data, but does include a synonym for that keyword, the user's knowledge of the synonym is not taken into account in the comparison result between the keyword information and the knowledge information. On the other hand, when calculating the similarity between the keyword information and the knowledge information and determining whether they match or are similar, a certain degree of variation is allowed in the comparison between the keyword information and the knowledge information. If the knowledge information includes a synonym for the keyword, the user's knowledge of the synonym may be taken into account. This allows for an increase in the similarity between the keyword information and the knowledge information even if there is no element in the user's knowledge information that perfectly matches the keyword, as long as a synonym or similar concept exists in the knowledge information.

The annotation unit 413 adds annotations to keywords in the input data based on the comparison results from the information comparison unit 412. For example, the annotation unit 413 can determine whether the user can understand the content of the input data, and if it determines that the user would have difficulty understanding the content of the input data, it can add annotations to the keywords.

Whether a user can understand the content of the input data can be determined, for example, by setting a predetermined threshold for the parameters of the input data to determine whether the user can understand the input data, and checking whether the parameters exceed the threshold. For example, the number of keywords could be set as a parameter, and the threshold could be set to the minimum number of keywords that a user is considered to be able to understand the input data. For example, if the number of keywords included in the user's knowledge information, or the number of keywords that match or are similar to elements of the knowledge information, is less than the threshold, the annotation unit 413 determines that it is difficult for the user to understand the content of the input data and adds annotations to some of the keywords in the input data. In this case, for example, the annotation unit 413 may add annotations to keywords not included in the knowledge information, i.e., keywords unknown to the user, based on the comparison results by the information comparison unit 412.

The threshold can be adjusted according to attributes such as the field and type of input data. For example, if the input data is a document in a highly specialized field, it will likely contain many technical terms, making it difficult to understand without knowing these terms. Therefore, a higher threshold will be set for such highly specialized documents. Conversely, general documents used in everyday life tend to use simpler language and fewer technical terms. Therefore, a lower threshold will be set for general documents. This improves the accuracy of determining whether a user will find the content of the input data difficult to understand.

Furthermore, the difficulty level of the input data can also be defined. The difficulty level of the input data can be calculated, for example, based on the threshold mentioned above. For instance, if the content of the input data belongs to a field generally considered difficult to understand, the number of keywords required for the user to understand the content of the input data tends to be large, and a high threshold may be set. Therefore, the annotation unit 413 is configured to calculate a higher difficulty level the higher the threshold.

The annotation unit 413 can also add annotations according to the difficulty level of the input data. For example, the annotation unit 413 can adjust the number of keywords to be annotated and the quality (detail) of the annotations to be added, according to the difficulty level of the input data.

For example, for input data with high difficulty, the system can consider the user's knowledge level and add annotations not only for keywords not included in their knowledge, but also for keywords corresponding to lower-ranked elements (knowledge). This allows users to deepen their understanding of even high-difficulty input data.

Furthermore, the annotation unit 413 can perform simple annotations, which add only text to keywords, or detailed annotations, which add diagrams to keywords. For example, simple annotations may be performed when the difficulty level is low. This is because, in such cases, it is assumed that users can understand the content of the input data through text-based explanations alone, without the need for diagrams or other illustrations. On the other hand, detailed annotations may be performed when the difficulty level is high. This is because, in such cases, it is assumed that users would have difficulty understanding the content through text-based explanations alone.

Furthermore, the annotation unit 413 can add annotations to keywords based on the comparison results, regardless of whether the user can understand the content of the input data. For example, the annotation unit 413 can add annotations to keywords in the keyword information that are not included in the knowledge information.

The output generation unit 414 generates output data based on the annotations added to the keywords by the annotation addition unit 413. The output data includes the input data and the data of the annotations added to the keywords. The output data is stored in the auxiliary storage unit 440.

<Client terminals A501 to C503>
Figure 8 is a schematic block diagram illustrating the hardware configuration of client terminals A501 to C503 shown in Figure 1. Each of the client terminals A501 to C503 is a computer equipped with a CPU 510, RAM 520, ROM 530, auxiliary storage unit 540, communication unit 550, and operation display unit 560, etc. Client terminals A501 to C503 can be, for example, a personal computer, a PDA (Personal Digital Assistant), a smartphone, etc.

Furthermore, the configurations of the CPU 510, RAM 520, ROM 530, auxiliary storage unit 540, and communication unit 550 are the same as those of the CPU 210, RAM 220, ROM 230, auxiliary storage unit 240, and communication unit 250 of the data server 200, so a detailed explanation is omitted. In this embodiment, the auxiliary storage unit 540 stores user information, such as the user's name, department, area of expertise, extension number, and email address.

The operation display unit 560 has an input unit and an output unit. The input unit includes, for example, a keyboard, mouse, etc., and is used by the user to input various instructions (inputs) such as character input and various settings using the keyboard, mouse, etc. The output unit includes a display 561 (see Figures 10 and 11) and is used to present documents created by application software to the user. In this embodiment, the output unit displays output data on the display 561 to the user in accordance with instructions from the CPU 510. The output unit also has a speaker and can provide the user with audio information about the content of the input data and annotations attached to keywords in the input data.

<Control methods for information processing systems>
Figure 9 is a sequence chart illustrating a schematic processing procedure for the control method of the information processing system 100 according to the first embodiment. The processing shown in the sequence chart is realized by the CPU 210 executing a control program for the data server 200 and the CPU 410 executing a control program for the information processing device 400. Figure 10 is a schematic diagram showing an example of displaying the output data of the information processing device shown in Figure 1, and Figure 11 is a schematic diagram showing another example of displaying the output data.

In this embodiment, for example, we assume a scenario where user A manually uploads input data to the information processing system 100, and the information processing device 400 adds annotations to the keywords in the input data and displays them on the display of client terminal A501. In the example shown in Figure 9, it is assumed that the input data is not a document created by user A, and that user A is unaware of the content of the input data before uploading it. Before uploading the input data, the input data contains keywords unknown to user A. User A understands the content of the input data by checking the annotations displayed on the display.

As shown in Figure 9, the input data stored in client terminal A501 is transmitted (uploaded) to the data server 200 along with user information. The user information includes information about user A. The data server 200 retrieves the input data and user information and temporarily stores them in the RAM 220 or auxiliary storage unit 240 (step S101).

Next, the data server 200 generates keyword information and stores it in the RAM 220 or auxiliary storage unit 240 (step S102). The user information, input data, and generated keyword information are transmitted to the information processing device 400. The information processing device 400 also requests knowledge information for user A from the knowledge server 300, and the knowledge server 300 transmits user A's knowledge information to the information processing device 400 in response to the request.

Next, the information processing device 400 adds annotations to the keywords based on the keyword information and knowledge information, and generates output data (step S103). The content of the annotations is read from the auxiliary storage unit 440. The generated output data is transmitted to the client terminal A501. Detailed processing of the information processing device 400 will be described later with reference to Figure 12.

Next, the client terminal A501 displays the output data on the display 561 (step S104). As shown in Figure 10, the screen 600 of the display 561 displays not only the content of the input data but also annotations 603 of the keywords of the input data. In the example shown in Figure 10, since user A does not have knowledge of the two keywords "heart disease" and "antihypertensive drugs," annotations are added for these two keywords. Also, as shown by reference numerals 605 and 606 in Figure 11, the annotations may be configured to be displayed, for example, by callouts.

In this way, user A can easily understand the meaning of keywords by checking the annotations displayed on the display 561, thus efficiently understanding the content of the input data.

<Processing by information processing device 400 (S103)>
Figure 12 is a flowchart illustrating the general processing procedure for step S103 of the sequence chart in Figure 9. The processing in the flowchart is realized by the CPU 410 executing a control program. Figure 13 is a schematic diagram illustrating the comparison between keyword information and knowledge information. Figures 14 and 15 are conceptual diagrams illustrating the method for determining whether or not to add annotations to keywords in the examples and comparative examples.

As shown in Figure 12, the subset extraction unit 411 acquires keyword information and user A's knowledge information, and extracts a subset of user A's knowledge information based on the keyword information (steps S201, S202).

The information comparison unit 412 compares the keyword information with the knowledge information (step S203). More specifically, as shown in Figure 13, the information comparison unit 412 compares the keyword information network with the user A's knowledge information network. Keywords not included in the subset correspond to keywords unknown to user A (for example, the gray area in Figure 13).

The annotation unit 413 adds annotations to the keywords in the input data based on the comparison results from the information comparison unit 412 (step S204). For example, as shown in Figure 14, consider a case where the input data includes keywords W1 to W8 (area enclosed by solid lines), and the subset of user A includes keywords W2, W3, W6, W7, W9, W10, and W11 (area enclosed by dashed lines). In this embodiment, of the keywords W1 to W8 in the input data, keywords W2, W3, W6, and W7 are included in user A's knowledge information and are keywords that user A knows, so there is no need to add annotations to them. Therefore, the keywords to which annotations are added are W1, W4, W5, and W8, and the keywords to which annotations are not added are W2, W3, W6, and W7.

On the other hand, in the comparative example, while User A's area of expertise is considered, User A's existing knowledge is not. Keywords belonging to User A's area of expertise are stored in the area of expertise database, for example, W2, W12, and W13 (indicated by the dashed lines). The area of expertise database comprehensively stores keywords used in each area of expertise. Of the input data keywords W1 to W8, keyword W2, which belongs to the user's area of expertise, is a keyword that User A is highly likely to know, so there is no need to add an annotation. Therefore, the keywords to which annotations are added are W1, W3, W4, W5, W6, W7, and W8, while the keyword to which no annotation is added is W2. As a result, keywords W2, W3, W6, and W7 are annotated even though the user already possesses the knowledge. However, this increases the amount of information User A needs to check, which may be cumbersome and make it difficult to find annotations for other keywords that are needed.

Furthermore, as shown in Figure 15, we assume a case where the input data includes keywords W1 to W8 and W13 (area enclosed by solid lines), and the subset of user A includes keywords W3, W5, W9, W10, and W11 (area enclosed by dashed lines). In this embodiment, of the keywords W1 to W8 and W13 in the input data, keywords W3 and W5 are included in user A's knowledge information and are keywords that user A knows, so there is no need to add annotations. Therefore, the keywords to which annotations are added are W1, W2, W4, W6, W7, W8, and W13, and the keywords to which annotations are not added are W3 and W5.

On the other hand, in the comparative example, while User A's area of expertise is considered, User A's existing knowledge is not. Keywords belonging to User A's area of expertise include, for example, W2, W6, W7, W12, and W13 (within the dashed line). Of the input data keywords W1-W8 and W13, keywords W2, W6, W7, and W13, belonging to the user's area of expertise, are likely to be known by User A, so annotations are not necessary. Therefore, the keywords to be annotated are W1, W3, W4, W5, and W8, while the keywords to be unannotated are W2, W6, W7, and W13. As a result, keywords W3 and W5 are not annotated, even though User A does not possess knowledge about them. This makes it difficult for User A to understand the input content because they lack knowledge about keywords W3 and W5.

The output generation unit 414 generates output data (step S205). The generated output data is transmitted to the client terminal A501, and the information processing device 400 terminates processing (end).

As described above, in the sequence chart in Figure 9 and the flowchart in Figure 12, input data is acquired, keyword information is compared with user A's knowledge information, and annotations are added to the keywords in the input data based on the comparison results. The annotated input data is sent to client terminal A501 as output data, and the output data is displayed on display 561.

According to the information processing system 100 of the first embodiment described above, annotations are added to the keywords in the input data based on the comparison result between the keyword information and knowledge information of the input data. Therefore, annotations are added to the keywords in the input data according to the knowledge possessed by the user.

Furthermore, by comparing keyword information with a subset, comparisons can be performed within a limited scope of knowledge information, thereby improving the efficiency of the comparison process.

(Second embodiment)
In the first embodiment, a case was described in which input data is uploaded from the client terminal A501 to the data server 200 and the information processing device 400 acquires the input data from the data server 200. In the second embodiment, a case is described in which the information processing device 400 acquires documents that meet the monitoring conditions as input data.

Figure 16 is a sequence chart illustrating the acquisition of input data in the second embodiment. In the second embodiment, the information processing system 100 further includes an application server and a mail server. The application server stores user information for multiple users, including users A, B, and C, in an auxiliary storage unit. While the method of acquiring input data differs from the first embodiment in this embodiment, the configurations of the data server 200, knowledge server 300, and information processing device 400 are the same as in the first embodiment. To avoid repetition, the same configurations as in the first embodiment will be omitted from the following description.

As shown in Figure 16, login information is sent from client terminal A501 to the application server. The application server performs login authentication for user A (step S301), and if the login authentication is successful, it notifies client terminal A501. The authentication information is then sent from the application server to the mail server. The mail server performs authentication based on the authentication information (step S302), and if the authentication is successful, it notifies the application server.

Next, the application server sets monitoring conditions (predetermined conditions) for emails addressed to user A that the mail server has received (step S303). The monitoring conditions may, for example, be whether the sender of the received email belongs to a specific group. The monitoring conditions are sent to the mail server.

Next, monitoring of the mail server begins (step S304). The application server monitors emails addressed to user A on the mail server using email software such as MICROSOFT® OUTLOOK®. The mail server checks whether each email addressed to user A meets the monitoring conditions (step S305) and sends the confirmation result to the application server. The application server and mail server function as a mail monitoring unit.

The application server retrieves user information for user A based on the login information. Furthermore, based on the verification results, the application server retrieves emails addressed to user A that meet the monitoring conditions as input data (step S306) and sends them to the data server 200 along with the user information.

The data server 200 generates and stores keyword information. User information, input data, and keyword information are sent to the information processing device 400. The information processing device 400 requests knowledge information for a specific user (user A) from the knowledge server 300, and the knowledge server 300 sends user A's knowledge information to the information processing device 400. The information processing device 400 adds annotations to the keywords based on the keyword information and user A's knowledge information (subset) and generates output data (step S307). Details of the processing by the information processing device 400 are as described above and will be omitted here. The generated output data is sent to the mail server.

The mail server updates the email (step S308). More specifically, the mail server updates the original email data that formed the basis of the input data using the output data generated by the information processing device 400. The mail server also sends the updated email data to the application server upon request.

The application server generates email display data using the email software and sends it to client terminal A501. Client terminal A501 displays the email based on the display data (step S309). The annotated email is displayed on display 561.

Although not shown in the diagram, the system may also be configured to include a file monitoring unit that constantly monitors files moved to the user's folder on the application server or client terminal, in addition to received emails. In this case, the information processing device 400 acquires files moved to the user's folder as input data, or acquires files moved to the user's folder as input data if they have predetermined attributes (e.g., a predetermined file name or predetermined creation date and time). Furthermore, it may have a file reading unit that reads files moved to the user's folder and determines whether the file reading result satisfies the monitoring conditions. For example, the system may be configured to determine that the monitoring conditions are met if the content of a file moved to the user's folder is a document in a specific field.

According to the information processing system 100 of the second embodiment described above, the documents to be annotated (for example, received emails or files in a specific folder) are selected based on monitoring conditions, so the user does not need to upload input data to be annotated.

As described above, the information processing system 100, the information processing device 400, and the control program have been explained in the embodiments. However, the present invention can be appropriately added to, modified, and omitted by those skilled in the art within the scope of its technical concept.

For example, in the first and second embodiments described above, a case was explained in which keyword information is generated from input data in the data server 200 and user knowledge information is stored in the knowledge server 300. However, the present invention is not limited to this case, and the information processing device 400 can also be configured to generate keyword information from input data and store user knowledge information. In this case, the CPU 410 and the communication unit 460 perform the functions of the data acquisition unit, the CPU 410 performs the function of the keyword generation unit, and the auxiliary storage unit 440 performs the function of the information storage unit.

Furthermore, in the first and second embodiments described above, a case was explained in which user knowledge information is read from the knowledge server 300 to the information processing device 400, and the information comparison unit 412 of the information processing device 400 performs a comparison between keyword information and knowledge information. However, the system is not limited to this case; the knowledge server 300 can also be configured to perform the comparison between keyword information and knowledge information.

Furthermore, while the first embodiment described a case where user information is stored in the auxiliary storage unit 540 and input data and user information are transmitted to the data server 200, the system is not limited to this case. For example, similar to the second embodiment, the information processing system 100 may have an application server, perform user login authentication, and be configured to acquire user information based on login information.

Furthermore, the control program may be provided on a computer-readable recording medium such as a USB memory stick, flexible disk, or CD-ROM, or it may be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable recording medium is usually transferred and stored in memory or storage. This control program may also be provided as, for example, a standalone application software, or it may be incorporated into the software of each device as a function of the server.

Furthermore, in this embodiment, some or all of the processing performed by the control program may be replaced and executed by hardware such as circuits.

100 Information Processing Systems,
200 data servers,
210 CPU,
220 RAM,
230 ROM,
240 Auxiliary storage unit,
250 Communications Department,
261 Data acquisition unit,
262 Keyword Information Generation Unit,
300 Knowledge Servers,
310 databases,
400 information processing devices,
410 CPU,
411 Subset extraction unit,
412 Information comparison department,
413 Annotation section,
414 Output generation unit,
420 RAM,
430 ROM,
440 Auxiliary storage unit,
450 Communications Department,
501-503 Client terminals A-C.

Claims (18)

Data acquisition unit,
An information storage unit that stores knowledge information for each user, which includes multiple elements corresponding to multiple pieces of information that constitute knowledge,
An information comparison unit compares keyword information related to keywords extracted from data acquired as input data by the data acquisition unit with knowledge information associated with the user.
The system includes an annotation unit that adds annotations to the keywords of the input data based on the comparison results from the information comparison unit,
The information comparison unit determines whether the knowledge information contains the keywords from the keyword information, or whether the keyword information and the knowledge information match or are similar.
The annotation unit is an information processing system that, when the number of keywords among the multiple keywords in the keyword information that are included in the user's knowledge information, or the number of keywords that match or are similar to elements of the knowledge information, is less than a threshold, adds annotations to the keywords among the multiple keywords that are not included in the user's knowledge information. The aforementioned keyword information is information that includes multiple keywords,
The information processing system according to claim 1, wherein the information comparison unit determines whether the keyword information and the knowledge information match or are similar by calculating a degree of similarity based on the vector of keywords included in the keyword information and the vector of elements included in the knowledge information. The information processing system according to claim 1, wherein the keyword information is represented by a network including multiple nodes, each representing one of the multiple keywords, and links defining the relationships between the nodes. The aforementioned knowledge information is information that networks the relationships between the elements corresponding to multiple pieces of information that constitute the knowledge,
The information processing system according to claim 3, wherein the information comparison unit determines whether the keyword information and the knowledge information match or are similar by calculating a similarity between the network of keyword information and the network of knowledge information based on the node match rate and the distance between nodes. The system further includes a subset extraction unit that extracts a subset of the user's knowledge information based on the aforementioned keywords.
The subset extraction unit searches the user's knowledge information for nodes tagged with the keyword, and extracts a subset containing the nodes tagged with the keyword.
The information processing system according to claim 4, wherein the information comparison unit compares the keyword information with the subset. The information processing system according to any one of claims 1 to 5, further comprising an output generation unit that generates output data based on annotations added to the keyword by the annotation addition unit. The data acquisition unit acquires the data uploaded by the user as the input data, according to any one of claims 1 to 6. The system further includes a mail monitoring unit that monitors emails received by the aforementioned user,
The information processing system according to any one of claims 1 to 6, wherein the data acquisition unit acquires the received email as input data when the received email satisfies predetermined conditions. The system further includes a file monitoring unit that monitors files being moved to a folder where the user's files are saved, and a file reading unit that reads files saved in the folder.
The information processing system according to any one of claims 1 to 6, wherein the data acquisition unit acquires the file in the folder, or the result of reading the file, as input data when predetermined conditions are met. An information processing system having a data server , a knowledge server , and an information processing device,
The aforementioned data server is
Data acquisition unit,
It includes a keyword information generation unit that generates keyword information from data acquired as input data by the data acquisition unit,
The aforementioned knowledge server has an information storage unit that stores knowledge information for each user, which includes multiple elements corresponding to multiple pieces of information that constitute knowledge.
The keyword information generated by the keyword information generation unit is stored, and the keyword information and the knowledge information are transmitted to the information processing device.
The aforementioned information processing device is
An information comparison unit that compares the acquired keyword information with the knowledge information associated with the user,
The system includes an annotation unit that adds annotations to the keywords of the input data based on the comparison results from the information comparison unit,
The information comparison unit determines whether the knowledge information contains the keywords from the keyword information, or whether the keyword information and the knowledge information match or are similar.
The annotation unit is an information processing system that, when the number of keywords among the multiple keywords in the keyword information that are included in the user's knowledge information, or the number of keywords that match or are similar to elements of the knowledge information, is less than a threshold, adds annotations to the keywords among the multiple keywords that are not included in the user's knowledge information. The information processing system according to claim 10, wherein the annotation unit adjusts the threshold according to the attributes of the input data. The information storage unit is provided by a knowledge server and stores the knowledge information for each of the multiple users, according to any one of claims 1 to 9 . The information processing system according to claim 11, wherein the annotation unit performs either a simplified annotation that adds an annotation containing only text to the keyword, or a detailed annotation that adds an annotation containing at least a figure to the keyword. The information processing system according to claim 13, wherein the annotation unit calculates the difficulty level of the input data based on the threshold and the comparison result, and performs the detailed annotation if the difficulty level is high, while performing the simplified annotation if the difficulty level is low. The information processing system according to claim 6, wherein the output data includes the input data and annotations attached to the keywords. The information processing system according to any one of claims 1 to 15, wherein the knowledge information is generated based on at least one of the following: data relating to social networking services (SNS) used by the user, data relating to the web viewed by the user, and document data created by the user. A keyword information acquisition unit that acquires keyword information from data acquired as input data,
A knowledge information acquisition unit acquires knowledge information for a specific user that includes multiple elements corresponding to multiple pieces of information that constitute knowledge,
An information comparison unit that compares the keyword information with the knowledge information associated with the specific user,
The system includes an annotation unit that adds annotations to the keywords of the input data based on the comparison results from the information comparison unit,
The information comparison unit determines whether the keyword information contains the keyword, or whether the keyword information and the knowledge information match or are similar.
The annotation unit is an information processing device that, when the number of keywords among the multiple keywords in the keyword information that are included in the knowledge information of the specific user, or the number of keywords that match or are similar to elements of the knowledge information, is less than a threshold, adds annotations to the keywords among the multiple keywords that are not included in the knowledge information of the specific user. Keyword information acquisition step (a) involves acquiring keyword information from data obtained as input data,
Knowledge information acquisition step (b) acquires knowledge information for a specific user that includes multiple elements corresponding to multiple pieces of information that constitute knowledge,
A comparison step (c) involves comparing the keyword information with the knowledge information associated with the specific user,
A process comprising : an annotation step (d) which adds annotations to the keywords of the input data based on the comparison results in the comparison step (c);
In the comparison step (c) above, it is determined whether the knowledge information contains the keywords of the keyword information, or whether the keyword information and the knowledge information are identical or similar.
In the annotation step (d) described above, if the number of keywords among the multiple keywords in the keyword information that are included in the knowledge information of the specific user, or the number of keywords that match or are similar to elements of the knowledge information, is less than a threshold, a control program to cause a computer to perform a process of adding annotations to the keywords among the multiple keywords that are not included in the knowledge information of the specific user.