Information processing system, information processing method, and program

Abstract

To provide an information processing system and the like with which a user can more easily obtain quantitative prediction results relating to various economic events or the like.SOLUTION: According to one embodiment of the present invention, an information processing system is provided. This information processing system is configured in such a way to perform the following each step. In an acquisition step, input information accepted from a user, and predetermined reference information are acquired. The reference information is information including a plurality of causes and effects associating a cause and a result. In an extraction step, a relation numerical index is extracted based on the acquired input information and reference information. The relation numerical index is related to at least one item appearing in a chained causal relationship, with a keyword included in the input information as a starting point.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing method, and a program. [Background technology]

[0002] Patent Document 1 discloses a causal relationship extraction system that extracts causal relationships between economic events and corporate performance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-24689 A Summary of the Invention [Problem to be solved by the invention]

[0004] In recent years, there has been a demand for more quantitative prediction of various economic phenomena, etc., using information processing technology. However, although the technology disclosed in Patent Document 1 can extract causal relationships, it cannot make quantitative predictions such as increases and decreases in parameters.

[0005] In view of the above circumstances, the present invention provides an information processing system etc. that allows a user to more easily obtain quantitative prediction results regarding various economic phenomena etc. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is provided an information processing system. The information processing system is configured to execute the following steps: In the acquisition step, input information received from a user and preset reference information are acquired. The reference information is information including causality that associates multiple causes and effects. In the extraction step, a related numerical index is extracted based on the acquired input information and reference information. The related numerical index is related to at least one item appearing in a chain of causal relationships starting from a keyword included in the input information.

[0007] According to this aspect, a user can more easily obtain quantitative prediction results regarding various economic phenomena and the like. [Brief description of the drawings]

[0008] [Figure 1] 1 is a configuration diagram illustrating an information processing system 1 according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram showing a hardware configuration of a user terminal 2. [Diagram 3] FIG. 2 is a block diagram showing a hardware configuration of the server 3. [Figure 4] 2 is a block diagram showing functions realized by a control unit 33 and the like in the server 3. FIG. [Diagram 5] FIG. 11 is an activity diagram showing the flow of information processing relating to generation of reference information IF. [Figure 6] FIG. 13 is an activity diagram showing the flow of information processing regarding the presentation of the causal chain 4 and related numerical indicators RI. [Figure 7] FIG. 13 is a conceptual diagram for explaining a causal chain 4. [Figure 8] 2 is a diagram showing an example of a screen 5 of a display unit 24 visually recognized by a user. FIG. [Figure 9] 2 is a diagram showing an example of a screen 5 of a display unit 24 visually recognized by a user. FIG. [Figure 10] 11A to 11C are diagrams showing examples of different aspects of the screen 7 of the display unit 24 visually recognized by the user. [Figure 11]11A to 11C are diagrams showing examples of different aspects of the screen 7 of the display unit 24 visually recognized by the user. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various characteristic features shown in the following embodiments can be combined with each other.

[0010] Incidentally, the program for realizing the software appearing in this embodiment may be provided as a non-transitory computer-readable recording medium, or may be provided so as to be downloadable from an external server, or may be provided so that the program is started on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0011] In addition, in this embodiment, the term "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In addition, in this embodiment, various information is handled, and this information is represented, for example, by physical values ​​of signal values ​​representing voltage and current, high and low signal values ​​as a binary bit collection consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculation can be performed on the circuit in the broad sense.

[0012] In addition, a circuit in the broad sense is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, a memory, etc. In other words, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0013] 1. Hardware Configuration In this section, the hardware configuration of this embodiment will be described.

[0014] 1.1 Information Processing System 1 1 is a configuration diagram showing an information processing system 1 according to this embodiment. The information processing system 1 includes a user terminal 2 and a server 3, which are connected via a network 11. These components will be further described. Here, a system exemplified as the information processing system 1 is composed of one or more devices or components. Therefore, even the server 3 alone is an example of a system.

[0015] 1.2 User terminal 2 2 is a block diagram showing a hardware configuration of the user terminal 2. The user terminal 2 has a communication unit 21, a storage unit 22, a control unit 23, a display unit 24, and an input unit 25, and these components are electrically connected via a communication bus 20 inside the user terminal 2. Descriptions of the communication unit 21, the storage unit 22, and the control unit 23 will be omitted because they are substantially similar to the communication unit 31, the storage unit 32, and the control unit 33 in the server 3, which will be described later.

[0016] The display unit 24 may be, for example, included in the housing of the user terminal 2, or may be externally attached. The display unit 24 displays a screen of a graphical user interface (GUI) that can be operated by the user. This is preferably implemented by using display devices such as a CRT display, a liquid crystal display, an organic EL display, and a plasma display depending on the type of the user terminal 2. Here, the display unit 24 will be described as being included in the housing of the user terminal 2.

[0017] The input unit 25 may be included in the housing of the user terminal 2, or may be externally attached. For example, the input unit 25 may be implemented as a touch panel integrated with the display unit 24. If it is a touch panel, the user can input a tap operation, a swipe operation, or the like. Of course, a switch button, a mouse, a QWERTY keyboard, or the like may be adopted instead of the touch panel. That is, the input unit 25 accepts an operation input made by the user. The input is transferred as a command signal to the control unit 23 via the communication bus 20, and the control unit 23 can execute a predetermined control or calculation as necessary.

[0018] 1.3 Server 3 3 is a block diagram showing the hardware configuration of the server 3. The server 3 has a communication unit 31, a storage unit 32, and a control unit 33, and these components are electrically connected via a communication bus 30 inside the server 3. Each component will be further described.

[0019] The communication unit 31 is preferably a wired communication means such as USB, IEEE1394, Thunderbolt, wired LAN network communication, etc., but may also include wireless LAN network communication, mobile communication such as 3G / LTE / 5G, Bluetooth (registered trademark) communication, etc. as necessary. In other words, it is more preferable to implement it as a collection of multiple communication means. In other words, the server 3 communicates various information with the user terminal 2 via the communication unit 31 and the network 11.

[0020] The storage unit 32 stores various information defined by the above description. This can be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the server 3 executed by the control unit 33, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to the calculation of the program. Also, a combination of these may be used. In particular, the storage unit 32 stores various programs and the like related to the server 3 executed by the control unit 33.

[0021] The control unit 33 performs processing and control of the overall operation related to the server 3. The control unit 33 is, for example, a central processing unit (CPU) not shown. The control unit 33 realizes various functions related to the server 3 by reading out a predetermined program stored in the storage unit 32. That is, information processing by the software stored in the storage unit 32 can be specifically realized by the control unit 33, which is an example of hardware, and executed as each functional unit included in the control unit 33. These will be described in more detail in the next section. Note that the control unit 33 is not limited to being single, and may be implemented with multiple control units 33 for each function. Also, a combination of these may be used.

[0022] 2. Functional configuration In this section, the functional configuration of the present embodiment will be described. As described above, information processing by the software stored in the storage unit 32 is specifically realized by the control unit 33, which is an example of hardware, and can be executed as each functional unit included in the control unit 33.

[0023] 4 is a block diagram showing functions realized by the control unit 33 and the like in the server 3. Specifically, the server 3, which is an example of the information processing system 1, includes an acquisition unit 331, a reference information generation unit 332, a chain processing unit 333, and a display control unit 334.

[0024] The acquisition unit 331 is configured to acquire various information necessary for information processing. For example, the acquisition unit 331 may read out various information previously stored in the storage unit 32, or may acquire various information from an external device via the network 11 and the communication unit 31. In particular, the acquisition unit 331 acquires an input text KW (input information) received from a user and a preset causal information database IF1 (reference information IF). This will be described in more detail later.

[0025] The reference information generating unit 332 is configured to generate the reference information IF. More specifically, the reference information generating unit 332 generates the causal information database IF1 (reference information IF) by extracting a plurality of causalities based on the text data TX. This will be described in more detail later.

[0026] The chain processing unit 333 is configured to generate a causal chain 4 that visually indicates a chain of causal relationships, as will be described in more detail later.

[0027] The display control unit 334 is configured to generate various display information and control display contents viewable by the user. The display information may be information itself generated in a form viewable by the user, such as a screen, an image, an icon, text, etc., or may be rendering information for displaying a screen, an image, an icon, text, etc. on the display unit 24 of the user terminal 2. More specifically, the display control unit 334 controls to display a screen including the related numerical index RI based on the acquired input text KW (input information) and causal information database IF1 (reference information IF). This will be described in more detail later.

[0028] 3. Information processing method In this section, an information processing method of the above-mentioned information processing system 1 will be described.

[0029] 3.1 Generating Reference Information Interfaces First, the information processing related to the generation of the reference information IF will be described. Fig. 5 is an activity diagram showing the flow of the information processing related to the generation of the reference information IF. The reference information IF is information that the server 3 refers to when performing information processing related to the generation of the causal chain 4 (see Fig. 7) described later and the presentation of the related numerical indicators RI (see Figs. 8 and 9).

[0030] For example, the reference information IF may be a causal information database IF1 including multiple causal relationships that associate causes and effects. Alternatively, the reference information IF may further include information that associates multiple keywords with the associated numerical index RI. According to this embodiment, the associated numerical index RI associated with the extracted node 4N in the causal chain 4 can be easily presented. Preferably, the associated information is a trained model IF2 that has previously been machine-learned to learn the correlation between multiple keywords and the associated numerical index RI.

[0031] Also, an example will be shown in which the generation of the reference information IF is performed by the server 3. In this case, the user may directly operate the server 3, or may use the user terminal 2 or the like to remotely operate the server 3 by various methods typified by a remote desktop or the like. In the following, a case will be described in which the user successively generates a causal information database IF1, which is an example of the reference information IF, and a trained model IF2 for presenting a related numerical indicator RI.

[0032] First, a user who wants to construct a causal information database IF1, which is an example of reference information IF, selects any text data TX. In other words, the acquisition unit 331 in the server 3 acquires at least one piece of text data TX. The text data TX may be, for example, news, topic articles, academic papers, technical specifications, patent documents, research reports, etc. In addition, the text data TX is not limited to a text file, and may be a word processing software file, a PDF file, or an HTML file on the Internet (activity A001).

[0033] Next, the control unit 33 in the server 3 analyzes the text data TX selected in activity A001 by reading out the dedicated program stored in the storage unit 32. As a result, causal pairs related to various events are extracted from the text data TX (activity A002). There are no particular limitations on the method for extracting causal pairs, and existing natural language processing algorithms may be appropriately adopted.

[0034] Next, the control unit 33 in the server 3 reads out the dedicated program stored in the storage unit 32, and adds the causal set extracted in activity A002 to the causal information database IF1, thereby generating the causal information database IF1 (activity A003). In other words, the reference information generating unit 332 generates the causal information database IF1 (reference information IF) by extracting multiple causalities based on the text data TX. According to this embodiment, the causal information database IF1 can be generated according to the user's preferences and the desired usage environment, and therefore prediction of events with more limited uses can be realized.

[0035] If necessary, the user can select multiple text data TX. For example, after the causal information database IF1 is generated using the text data TX selected the first time, another text data TX can be selected as the second or subsequent processing. In other words, by appropriately repeating the activities A001 to A003, the causal information database IF1 with a larger amount of information can be generated.

[0036] Next, a user who wants to build a trained model IF2, which is an example of reference information IF, selects various keywords and numerical indicators that will be used as training data (activity A004). This may be the selection of text data TX used to extract causal pairs, or may be the selection of different data. Next, the user inputs the selected various keywords and numerical indicators into an existing machine learning algorithm, and causes the server 3 to execute machine learning (activity A005). As a result, a trained model IF2 is generated (activity A006). By performing machine learning in this way, it is possible to present related numerical indicators RI that are more accurate and relevant than methods using lookup tables, etc.

[0037] If necessary, the user can update the trained model IF2 by repeating machine learning using additional training data. For example, after generating the trained model IF2 using the keywords and numerical indicators selected the first time, further machine learning can be performed using different training data as input for the second or subsequent processing. In other words, by appropriately repeating the activities A004 to A006, a trained model IF2 with higher accuracy can be generated.

[0038] 3.2 Presentation of causal chain4 and related numerical indicators RI Next, a description will be given of information processing relating to the presentation of the causal chain 4 and the related numerical indicators RI. Fig. 6 is an activity diagram showing the flow of information processing relating to the presentation of the causal chain 4 and the related numerical indicators RI.

[0039] (Overview of the process) The flow of information processing will be outlined below along each activity in Fig. 6. The user here is assumed to be a person who wants to predict an economic event using the service provided by the information processing system 1. The user, while referring to the screen 5 displayed on the display unit 24 of the user terminal 2, inputs an input text KW (input information) related to the event that the user wants to predict using the input unit 25 (activity A101). This input text KW is transmitted to the server 3 via the network 11.

[0040] Next, as an acquisition step, the acquisition unit 331 acquires the input text KW (input information) received from the user and the (pre-set) reference information IF generated by the information processing shown in Fig. 5 (activity A102). That is, the control unit 33 searches for a causal pair suitable for the input text KW by matching the input text KW received through the communication unit 31 with the causal information database IF1 stored in the storage unit 32 (activity A103) and determines it (activity A104). An example of the causal pair will be described later.

[0041] In addition, when the causal pair is determined, the control unit 33 inputs the keywords included in the causal pair to the trained model IF2 stored in the storage unit 32, whereby a related numerical index RI highly related to the keywords is proposed. Examples of the related numerical index RI will be described later.

[0042] Next, a screen such as the screen 5 is displayed in such a manner that the user can grasp the determined causal set and the proposed related numerical index RI (activity A105). In other words, the control unit 33 reads out a predetermined program stored in the storage unit 32, and extracts the related numerical index RI based on the acquired input text KW (input information) and causal information database IF1 (reference information IF) as an extraction step. Then, the display control unit 334 controls to display the screen 5 including the extracted related numerical index RI as a display control step. According to this embodiment, the user can input the input text KW related to an event in which the user is interested, and can grasp the related numerical index RI, which is a numerical index associated with the causal effect of various events related to the input text KW. That is, the user can obtain a more quantitative prediction result than ever before.

[0043] Note that the user may appropriately next input a different input text KW that the user is interested in. In other words, by appropriately repeating the activities A101 to A105, the user can obtain prediction results for a plurality of events.

[0044] Also, based on the keywords contained in the causal pair determined in activity A103, it is possible to subsequently search for another causal pair similar to this. Specifically, first, the user selects one of the causal pairs (node ​​4N) already displayed on the display unit 24 using the input unit 25 (activity A106).

[0045] Next, the control unit 33 reads out the reference information IF (activity A107) and searches for and determines causal pairs related to the selected causal pair by comparing the selected causal pair with the causal information database IF1 stored in the storage unit 32 (activity A108). Examples of related causal pairs will be described later.

[0046] In addition, once a related causal pair is determined, the control unit 33 inputs keywords contained in the causal pair into the learned model IF2 stored in the memory unit 32, thereby proposing a new related numerical indicator RI that is highly related to the keywords (activity A109).

[0047] Next, a screen such as screen 5 is updated in a manner that allows the user to grasp the determined causal set and the proposed related numerical indicators RI (activity A110).

[0048] That is, by the user appropriately repeating activities A101 to A105 or A106 to A110, a causal chain 4 is generated and displayed so as to be visible to the user. In other words, the chain processing unit 333, which is a functional unit of the control unit 33, can generate a causal chain 4 that visually shows a chain of causal relationships as a chain processing step. Then, together with the causal chain 4, an associated numerical index RI associated with each causal pair is proposed.

[0049] To summarize the above, this information processing method includes the steps of the information processing system 1, and more specifically, the following steps. In the acquisition step, an input text KW (input information) received from a user and a preset causal information database IF1 (reference information) are acquired. The causal information database IF1 (reference information) is information including causality that associates multiple causes with effects. In the extraction step, a related numerical index RI is extracted based on the acquired input text KW (input information) and causal information database IF1 (reference information). The related numerical index RI is related to at least one item appearing in a chain of causal relationships starting from a keyword included in the input text KW (input information).

[0050] (Details of information processing) Next, a detailed description will be given of the information processing outlined in Fig. 6. Fig. 7 is a conceptual diagram for explaining the causal chain 4. Figs. 8 and 9 are diagrams showing an example of the screen 5 of the display unit 24 visually recognized by the user.

[0051] As shown in FIG. 7, the causal chain 4 is a network structure having nodes 4N hierarchically, each of which represents a causal set. The causal chain 4 indicates a chain of causal relationships starting from an input text KW input by a user. More specifically, the causal chain 4 includes a plurality of nodes 4N indicating causality in the causal information database IF1. In this example, when a user inputs an input text KW, nodes 41, 42, and 43, which are nodes 4N representing three causal sets, are presented based on the causal information database IF1. As described above, the causal information database IF1 is an example of reference information IF that is preset based on text data TX.

[0052] Also, in relation to the causal set shown in node 41, nodes 411, 412, and 413 are presented as nodes 4N succeeding node 41. Similarly, in relation to the causal set shown in node 42, nodes 421, 422, and 423 are presented as nodes 4N succeeding node 42, and in relation to the causal set shown in node 43, nodes 431, 432, and 433 are presented as nodes 4N succeeding node 43. According to such a causal chain 4, it is possible to form a causal flow of related events.

[0053] Here, it is preferable that the causal effects indicated by the mutually connected nodes 4N are similar to each other. More specifically, the multiple nodes 4N are configured to be capable of calculating the similarity between them. The causal chain 4 has a predetermined causal chain 4' including multiple nodes 4N connected in series. For example, in FIG. 7, the input text KW, the node 41, and the node 411 correspond to the predetermined causal chain 4'. Then, the chain processing unit 333 adds a new node 4N following the terminal node 4N of the predetermined causal chain 4', and the added new node 4N may have a similarity with the terminal node 4N of a predetermined value or more. The concept of similarity is not particularly limited, and the similarity may be determined based on a distance defined in existing natural language processing, for example.

[0054] More specifically, the causal chain 4 may be presented on a screen 5 as shown in Fig. 8. In the example of Fig. 8, the user inputs "infectious disease" as the input text KW in an input field 50 while referring to the screen 5. When the user presses a search icon 501 in this state, a node 4N indicating a causal pair consisting of a first cause 51 and a first result 52 is displayed in an area R1. The first cause 51 includes the keywords "world situation", "infectious disease", and "impact", and the first result 52 caused by these includes the keywords "Japanese", "exodus company", "year-on-year comparison", and "ten thousand people".

[0055] Furthermore, when the user selects the keyword "Japanese" 521 in node 4N corresponding to region R1, a node 4N is further displayed in region R2 indicating a causal pair consisting of a second cause 61 and a second result 62. The second cause 61 includes the keywords "Japanese", "guests", and "decrease", and the second result 62 resulting from these includes the keywords "supply and demand", "mitigation", "direction", "guest rooms", and "unit price".

[0056] Although not shown, each time a node 4N is added, a new region and the corresponding node 4N may be displayed under the region R2. When a subsequent node 4N is added, the method of selecting the terminal node 4N is not particularly limited. For example, when the user selects the keyword "Japanese" 521 in the region R1 and presses the button 55, a new node 4N may be displayed in the region R2. In the example of FIG. 8, the input text KW is used as the starting point, and a first cause 51 and a first result 52, which is a result of the first cause 51, are presented, and a keyword related to the first result 52 is set as a second cause 61, and a second result 62, which is a further result of the first cause 51, are presented. In other words, the causal chain 4 is presented in a direction from the cause to the result.

[0057] Furthermore, when a node 4N corresponding to the region R2 is selected, a related numerical indicator RI related to the second result 62 is presented together with the causal chain 4. That is, the related numerical indicator RI is related to at least one item appearing in the chain of causal relationships starting from the input text KW. Here, the at least one item corresponds to the "guest room". More specifically, the related numerical indicator RI is related to at least one node 4N appearing in the causal chain 4. In addition, when generating such a screen 5, the display control unit 334 controls to display a screen including the causal chain 4 and the related numerical indicator RI. According to this embodiment, the related numerical indicator RI is displayed on the screen together with the causal chain 4 in a list-like manner, so that the user can grasp the causal flow and the related numerical indicator RI from a bird's-eye view.

[0058] Here, when presenting the related association numerical index RI, the learned model IF2 may be referred to. The method of determining the degree of association is not particularly limited, but more specifically, for example, it may be determined based on the corrected odds ratio as shown in Equation 1. Also, instead of or in addition to the degree of association, the degree of non-association may be determined to present the association numerical index RI.

number

[0059] Preferably, the keywords included in the causal set shown in the node 4N include those indicating large and small increases and decreases, such as "decrease," "increase," "large," and "small," and more preferably, the multiple nodes 4N have polarities that indicate numerical increases or decreases, and the related numerical index RI is determined based on the polarity. In predicting economic events, etc., increasing and decreasing changes are extremely important, so according to this embodiment, the preferred related numerical index RI desired by the user can be presented in consideration of the direction of polarity changes along the causal chain 4 from the input text KW (input information) entered by the user to the related numerical index RI, resulting in a user-friendly specification. Furthermore, in addition to presenting the related numerical index RI, it is also possible to obtain a predicted value of the numerical value.

[0060] Furthermore, when the user selects the related numerical indicator "room occupancy rate" 63, which is an example of the related numerical indicator RI, while referring to the screen 5, it is preferable to transition to a screen 5' including a graph of the related numerical indicator RI as shown in Fig. 9. A graph relating to the related numerical indicator "room occupancy rate" 63 is displayed on the screen 5'. By allowing the user to refer to such a graph, the user can more intuitively understand the prediction results.

[0061] In this manner, in this embodiment, an information processing system 1 is proposed that searches the causal relationships of economic events and the like in a chain reaction manner from the causal information database IF1 extracted from the text data TX, and presents the related numerical index RI related to the ripple effect. According to the information processing method by such an information processing system 1, the user can more easily obtain quantitative prediction results regarding various economic events and the like.

[0062] 4.Other The information processing system 1 according to this embodiment may adopt the following aspects.

[0063] In the above embodiment, the configuration of the information processing system 1 has been described, but a program may be provided that causes a computer to execute each step of the information processing system 1. Such a program may be configured to be executable offline in a stand-alone computer separated from the network 11.

[0064] In presenting the related numerical index RI, rather than merely presenting the index itself, it may be implemented to provide a forecast of time-dependent changes, such as an upward change in value in X days or within X months.

[0065] At least one text data TX selected in generating reference information such as the causal information database IF1 may be a plurality of text data TX written in different languages. For example, the first extraction of causal pairs may be performed using text data TX written in Japanese, and the second extraction of causal pairs may be performed using text data TX written in a foreign language such as English or Chinese. In such a case, the chain processing unit 333 generates a causal chain 4 that visually shows a chain of causal relationships, and the causal chain 4 includes a plurality of nodes 4N that show causal relations described in different languages. That is, even in different languages, by extracting common concepts or keywords with the same meaning, it is possible to compare the similarity of the nodes 4N extracted in each different language. According to such an embodiment, a more universal causal chain 4 that transcends language barriers can be generated.

[0066] In the present embodiment, a new node 4N related to the terminal node 4N is searched and determined based on the causal information database IF1, but it may be implemented to refer to not only the terminal node 4N but also other nodes 4N. For example, other nodes 4N located in parallel in the same hierarchy, such as node 411 or node 413 as seen from node 412, may be referenced. However, it is particularly preferable that the chain processing unit 333 weights a plurality of nodes 4N in a predetermined causal chain 4' and determines a subsequent new node 4N based on the weighted evaluation value. According to such an embodiment, not only the immediately preceding node 4N but also a new node 4N following the previous causal flow is proposed, so that the user can make a more accurate prediction.

[0067] In this embodiment, particularly in Fig. 8, the causal chain 4 is presented in a direction from a cause to a result, but the user may input an input text KW related to a result, and the causal chain 4 may be presented in a direction from the result to a cause. Figs. 10 and 11 are diagrams showing examples of different aspects of the screen 7 of the display unit 24 visually recognized by the user.

[0068] In FIG. 10, the user inputs "infectious disease" as the input text KW in the input field 70 while referring to the screen 7. When the user presses the search icon 701 in this state, a node 81, a node 82, a node 83, and a plurality of nodes 4N are presented on the right side of the input field 70. When the user presses the button 751, as shown in FIG. 11, a node 821, a node 822, and a node 823 are further added on the right side. The added nodes 821, 822, and 823 are connected by a curved line so that it is visually recognizable that they are nodes 4N following the node 82. When the user presses the button 752, another node 4N having a parallel relationship with the nodes 81, 82, and 83 may be displayed.

[0069] Here, as shown in FIG. 10 and FIG. 11, the screen 7 has a selection button 76 and a selection button 77 below the input field 70. When the user selects the selection button 76, the causal chain 4 can be displayed in a direction from the cause to the result. Although not shown, when the user selects the selection button 77, the causal chain 4 can be displayed in a direction from the result to the cause. In general, since one item causes multiple results, the presentation of the direction from the cause to the result is considered to be a forward problem. Conversely, the presentation of the direction from the result to the cause is an inverse problem, and in order to perform more accurate cause prediction, it is preferable to set a separate constraint condition in addition to the judgment of the similarity. Such constraint conditions are not particularly limited, and it is possible to appropriately assume that the weighting value is changed, the subsequent node 4N is restricted, etc.

[0070] In this embodiment, the causal chain 4 can theoretically be added almost without limit as long as the information in the causal information database IF1 is enriched, but it is assumed that as the number of nodes 4N increases, the relevance to the input text KW initially input by the user becomes weaker. In order to prevent this, the number of layers of the nodes 4N that can be included in one causal chain 4 may be limited. The number of layers to be limited is not particularly limited, but specifically, for example, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, and 20 layers, and may be within a range between any two of the numerical values ​​exemplified here. More preferably, 1 to 5 layers, and particularly preferably, 3 layers.

[0071] In the present embodiment, an example has been described in which a causal chain 4 and an associated numerical index RI are generated starting from an input text KW and are presented, but the causal chain 4 including a causal node 4N associated with the associated numerical index RI may be presented using the associated numerical index RI as an input. In other words, the acquisition unit 331 acquires a numerical index (associated numerical index RI) received from a user and preset reference information IF as an acquisition step. The reference information IF is information that associates a plurality of keywords with numerical indexes, for example, a trained model IF2. Then, the control unit 33 extracts a causal chain 4, which is a chain of causal relationships including keywords, based on the acquired numerical index and reference information as an extraction step.

[0072] Furthermore, it may be provided in the following aspects: The information processing system is further configured to execute a display control step, in which a screen including the extracted related numerical indicator is controlled to be displayed. In the information processing system, the reference information further includes information associating a plurality of keywords with the associated numerical indicator. In the information processing system, the associated information is a trained model that has been machine-learned in advance to determine the correlation between the multiple keywords and the related numerical indicators. The information processing system further includes a chain processing step, in which a causal chain is generated that visually shows a chain of causal relationships, the causal chain including a plurality of nodes that indicate causality in the reference information, and the display control step controls to display a screen including the causal chain and the associated numerical indicator, the associated numerical indicator being related to at least one node that appears in the causal chain. In the information processing system, the multiple nodes are configured to be able to calculate similarities between each other, the causal chain includes a predetermined causal chain including multiple nodes arranged in series, and in the chain processing step, a new node is added following the terminal node of the predetermined causal chain, and the new node added here has a similarity with the terminal node that is equal to or greater than a predetermined value. In the information processing system, the chain processing step weights a plurality of nodes in the predetermined causal chain, and determines a new subsequent node based on the weighted evaluation value. In the information processing system, the plurality of nodes have a polarity representing a numerical increase or decrease, and the associated numerical indicator is determined based on the polarity. The information processing system further includes a reference information generating step, in which the acquisition step acquires at least one text data, and the reference information generating step generates the reference information by extracting multiple causalities based on the text data. The information processing system further includes a chain processing step, wherein the at least one text data is a plurality of text data written in different languages, and the chain processing step generates a causal chain that visually indicates a linked causal relationship, wherein the causal chain includes a plurality of nodes that indicate causality each described in the different languages. An information processing system configured to execute the following steps: in an acquisition step, a numerical indicator received from a user and preset reference information are acquired, the reference information being information that associates a plurality of keywords with the numerical indicator; and in an extraction step, a chain of causal relationships including the keywords are extracted based on the acquired numerical indicator and the reference information. An information processing method comprising the steps of the information processing system. A program that causes a computer to execute each step of the information processing system. Of course, this is not the case.

[0073] Finally, although various embodiments of the present invention have been described, these are presented as examples and are not intended to limit the scope of the invention. The novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments and their modifications are included within the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims.

[0074] 1: Information processing system 11: Network 2: User terminal 20: Communication bus 21: Communications Department 22: Storage section 23: Control section 24: Display section 25: Input section 3: Server 30: Communication bus 31: Communications Department 32: Storage section 33: Control section 331: Acquisition Department 332:Reference information generation unit 333: Chain processing section 334: Display control unit 4: Causal chain 4' : given causal chain 4N : Node 41 : Node 411 : Node 412 : Node 413 : Node 42 : Node 421 : Node 422: Node 423 : Node 43 : Node 431 : Node 432 : Node 433 : Node 5: Screen 5': Screen 50: Input field 501: Search icon 51: First Cause 52: First result 521: Keyword "Japanese" 55: Button 61: Second Cause 62: Second result 63: Related numerical indicator: "Room occupancy rate" 7: Screen 70: Input field 701: Search icon 751: Button 752: Button 76: Selection button 77: Selection button 81 : Node 82 : Node 821 : Node 822: Node 823 : Node 83 : Node IF : Reference information IF1: Causal information database IF2: Trained model KW : Input text R1: area R2: area RI: Related Numerical Indicators TX: Text data

Claims

1. An information processing system, A control unit is provided, The control unit is configured to execute the following steps: In the acquisition step, input information received from a user and preset reference information are acquired; Here, the reference information is information indicating a causal relationship that associates a plurality of items indicating causes with items indicating results, and further includes information that associates a plurality of keywords with related numerical indicators, Here, the related numerical index is a numerical index related to at least one of the items indicating the cause and the items indicating the result appearing in the chain of causal relationships starting from a keyword included in the input information, In the extraction step, the system extracts the relevant numerical indicators based on the acquired input information and the acquired reference information.

2. 2. The information processing system according to claim 1, The system, wherein the associated information is a trained model that has been machine-learned to determine the correlation between the multiple keywords and the related numerical indicators.

3. 3. The information processing system according to claim 1, Furthermore, in the display control step, the system displays a screen including the extracted related numerical indicators.

4. 4. The information processing system according to claim 3, Furthermore, in the chain processing step, a causal chain is generated that visually indicates the linked causal relationships, and the causal chain includes a plurality of nodes that respectively indicate the causal relationships in the reference information; In the display control step, a screen including the causal chain and the associated numerical indicators is displayed, A system in which the relevant numerical indicator is related to the causal relationship indicated by at least one node appearing in the causal chain.

5. 5. The information processing system according to claim 4, The plurality of nodes are configured to be able to calculate a similarity between each other, the causal chain comprises a predetermined causal chain including a plurality of nodes in series; In the chain processing step, a new node following the terminal node of the predetermined causal chain is A new node is added, and the new node is connected to the terminal node by a predetermined Systems having a similarity greater than or equal to the value.

6. 6. The information processing system according to claim 5, In the chain processing step, the system weights the similarity between each of the new node and multiple nodes serially connected in the specified causal chain, and determines the subsequent new node based on the weighted similarity.

7. In the information processing system according to any one of claims 4 to 6, the plurality of nodes have polarities representing numerical increase or decrease; A system in which the related numerical indicator is determined to be an indicator suitable for predicting events of the directional numerical change indicated by the polarity of the node indicating the causal relationship to which the related numerical indicator is related.

8. In the information processing system according to any one of claims 4 to 7, In the display control step, a list including a plurality of the related numerical indicators is displayed, and when any one of the related numerical indicators is selected from the list, a graph relating to the related numerical indicator is displayed.

9. In the information processing system according to any one of claims 1 to 8, In the acquiring step, at least one piece of text data is acquired, Furthermore, in the reference information generating step, a plurality of causal effects are extracted based on the text data. and generating the reference information.

10. 10. The information processing system according to claim 9, The at least one text data may include a plurality of text data written in different languages. It is In addition, the chain processing step creates a causal chain that visually shows the chain of causal relationships. wherein the causal chain is a set of causal chains described by different languages. A system including a plurality of nodes representing

11. 1. An information processing method, comprising: A method comprising the steps of the information processing system according to any one of claims 1 to 10, wherein each of the steps is executed by a control unit of the information processing system.

12. A program, A program for causing a computer to execute each step of the information processing system according to any one of claims 1 to 10.

Images