Presentation device and presentation method

Abstract

A presentation device (10) has a generation unit (153) and a determination unit (154). The generation unit (153) generates a third text (divergence information o) related to a first text (manifest thought ex) representing a manifest thought of a user and a second text (latent thought lx) representing a latent thought of the user. The determination unit (154) determines whether the degree of divergence among the first text, the second text and the third text is at least a threshold value.

Description

Presentation device and presentation method

[0001] The present invention relates to a presentation device and a presentation method.

[0002] When humans try to generate new ideas or concepts, their thinking progresses through two phases: divergent thinking and convergent thinking. The divergent thinking phase is the phase in which as many thoughts as possible about the problem are collected. The convergent thinking phase is the phase in which the collected thoughts are organized and integrated. In the convergent thinking phase, the essence of the problem is pursued and a hypothesis to solve the problem is generated.

[0003] During the divergent thinking phase, it is known to be effective to externalize the thoughts of the person thinking.

[0004] 2. Description of the Related Art Conventionally, there is known a technique for supporting communication by visualizing a user's thought space by placing words corresponding to concepts the user is thinking about on an interface (see, for example, Non-Patent Document 1).

[0005] Yasuyuki Kado, Ryuta Ogawa, Koichi Hori, Setsuo Osuga, Kenji Mase, "A Communication Support Method by Visualizing Thinking Space," IEICE Transactions on Computer Science and Engineering, Vol. J79-A, No. 2, pp. 251-260, February 1996

[0006] However, conventional techniques may not be able to fully diverge the user's thoughts.

[0007] For example, with the technology described in Non-Patent Document 1, it is difficult to place words that the user has not uttered on the interface. Therefore, the thought space is not necessarily visualized in a state where the user's thoughts are sufficiently diverged.

[0008] The present invention has been made in view of the above, and has as its object to fully diverge the user's ideas.

[0009] In order to solve the above-mentioned problems and achieve the objectives, the presentation device of the present invention is characterized by having a generation unit that generates a third text related to a first text that represents a user's explicit thoughts and a second text that represents the user's implicit thoughts, and a determination unit that determines whether the degree of divergence between the first text, the second text, and the third text is equal to or greater than a threshold.

[0010] According to the present invention, it is possible to fully diverge the user's thoughts.

[0011] FIG. 1 is a schematic diagram illustrating a general configuration of a presentation device according to this embodiment. FIG. 2 is a diagram illustrating a data configuration of an ambiguous expression dictionary. FIG. 3 is a diagram for explaining the ambiguous expression dictionary. FIG. 4 is a flowchart illustrating a presentation processing procedure. FIG. 5 is a diagram illustrating an example of a method for a user to input information. FIG. 6 is a diagram illustrating an example of an interface. FIG. 7 is a diagram illustrating an example of a method for a user to input information. FIG. 8 is a diagram illustrating an example of a method for a user to input information. FIG. 9 is a diagram illustrating an example of a method for a user to input information. FIG. 10 is a diagram illustrating an example of a method for a user to input information. FIG. 11 is a diagram illustrating an example of extracted text. FIG. 12 is a diagram illustrating an example of a computer that executes a presentation program.

[0012] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to this embodiment. In addition, in the description of the drawings, the same parts are designated by the same reference numerals.

[0013] [Configuration of Presentation Device] Fig. 1 is a schematic diagram illustrating the overall configuration of a presentation device according to the present embodiment. As illustrated in Fig. 1, the presentation device 10 according to the present embodiment is realized by a general-purpose computer such as a personal computer, and includes an input unit 11, an output unit 12, a communication control unit 13, a storage unit 14, and a control unit 15.

[0014] The input unit 11 is realized using input devices such as a keyboard, mouse, camera, and microphone, and inputs various instruction information such as a command to start processing to the control unit 15 in response to an input operation by an operator. The output unit 12 is realized by a display device such as a liquid crystal display, a printing device such as a printer, etc. For example, the output unit 12 displays the results of a presentation process described below.

[0015] The communication control unit 13 is realized by a NIC (Network Interface Card) or the like, and controls communication between the control unit 15 and an external device via a telecommunication line such as a LAN (Local Area Network) or the Internet. For example, the communication control unit 13 controls communication between the control unit 15 and an external management device or the like that manages various types of information.

[0016] The storage unit 14 is realized by a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 14 stores in advance the processing program that operates the presentation device 10, data used during execution of the processing program, and the like, or temporarily stores the data each time processing is performed. The storage unit 14 may be configured to communicate with the control unit 15 via the communication control unit 13.

[0017] In this embodiment, the storage unit 14 stores an ambiguous expression dictionary 141. The ambiguous expression dictionary 141 includes information for extracting text that expresses implicit thoughts from information input by the user.

[0018] Specifically, the ambiguous expression dictionary 141 includes one or more pieces of information on the part of speech of each word, the number of meanings indicated by each word, and the ambiguity of the meaning indicated by each word. This information is collected via the input unit 11 or from a management device that manages various information, prior to or during the presentation process described below, and stored in the storage unit 14.

[0019] Here, Fig. 2 is a diagram illustrating an example of the data structure of the ambiguous expression dictionary. Fig. 3 is a diagram for explaining the ambiguous expression dictionary. First, as illustrated in Fig. 2, the ambiguous expression dictionary 141 includes information items such as part of speech, meaning, number of meanings, basic ambiguity, type, etc. for each word.

[0020] The number of meanings is the number of different meanings for the same word. In addition, the number of homonyms may be added. The basic ambiguity is a value set by the user for each word, taking into account, for example, the part of speech, type, field in which the user is involved, etc. Alternatively, the basic ambiguity may be a value based on a predefined classification table of ambiguous words, etc.

[0021] The type is, for example, a value ranging from 1 to 7 that indicates each class of ambiguity classification. Here, Fig. 3 shows an example of seven ambiguity classes. For example, type = 1 is set for a word that has multiple meanings.

[0022] The ambiguity expression dictionary 141 may include information items that change dynamically during use, such as the importance of each word in a sentence. For example, it may be possible to individually set a low ambiguity score for a word that has a high ambiguity score calculated in the presentation process described below if the word does not have a high importance.

[0023] Furthermore, the ambiguous expression dictionary 141 is not limited to information items for each word, but may include information items for each group of expressions made up of multiple words. In this case, expressions may include verbal expressions of facial expressions.

[0024] Returning to the explanation of FIG. 1 , the control unit 15 is realized using a CPU (Central Processing Unit) or the like, and executes a processing program stored in memory. As a result, the control unit 15 functions as a explicit thought extraction unit 151, an implicit thought extraction unit 152, a generation unit 153, a determination unit 154, and a presentation unit 155, as exemplified in FIG. 1 , and executes the presentation process. Note that each or some of these functional units may be implemented in different hardware. The control unit 15 may also include other functional units.

[0025] The processing of each functional unit and the overall processing flow of the presentation device 10 will be described with reference to Fig. 4. Fig. 4 is a flowchart showing the presentation processing procedure.

[0026] First, the input unit 11 supplements and outputs information transmitted by a user (user-transmitted information x) (step S101). For example, the user-transmitted information x is text input by a user or text generated based on the user's speech.

[0027] The explicit thought extraction unit 151 extracts and outputs explicit thoughts ex from the user-sent information x (step S102). For example, the explicit thoughts ex are information that the user intentionally intends to input.

[0028] The implicit thought extraction unit 152 extracts and outputs implicit thoughts lx from the user-sent information x (step S103). For example, implicit thoughts lx are information that the user has sent but has not intentionally intended to input into the system.

[0029] There are various methods for extracting latent thoughts lx. Here, we will explain one extraction method. In this case, we assume that the user is involved in AI research and development.

[0030] 5 is a diagram showing an example of how a user inputs information. As shown in FIG. 5, the user inputs the text "harmony with humans." The user also utters, "I want to create an amazing AI." The input unit 11 passes text based on the user's speech (e.g., transcription by speech recognition) and the text input by the user to other functional units as user-transmitted information x.

[0031] The explicit thought extraction unit 151 extracts the text input by the user as the explicit thought ex. For example, the explicit thought extraction unit 151 extracts the text "harmony with humans" as the explicit thought ex.

[0032] The implicit thought extraction unit 152 extracts, from text based on the user's speech, a portion (e.g., a sentence or a phrase) whose fuzzy score is equal to or greater than a threshold as the implicit thought lx. For example, from the text "I want to make an amazing AI," the implicit thought extraction unit 152 extracts the phrase "amazing AI" as the implicit thought lx.

[0033] The ambiguity score is an index that indicates the degree of ambiguity of a word, etc. A method for calculating the ambiguity score will be described. First, the implicit thought extraction unit 152 performs a morphological analysis of the user-sent information x to extract words.

[0034] The implicit thought extraction unit 152 refers to the ambiguity expression dictionary 141 in the storage unit 14 and calculates an ambiguity score that indicates the ambiguity of each word that makes up the input dialogue. The implicit thought extraction unit 152 calculates the ambiguity score using variables derived from information included in the ambiguity expression dictionary 141.

[0035] Here, the variables are derived based on, for example, the part of speech or basic ambiguity of the word, or the number of meanings of the word, or the type, etc. Note that context-related variables are not used.

[0036] Furthermore, the implicit thought extraction unit 152 may calculate a fuzzy score for each group of expressions made up of multiple words, which is a unit of information in the fuzzy expression dictionary 141. In this case, the expressions may include verbal expressions of facial expressions, etc.

[0037] The implicit thought extraction unit 152 calculates the ambiguity score using the product of multiple pieces of information that have been weighted in a predetermined manner. Specifically, the implicit thought extraction unit 152 calculates the ambiguity score r of the i-th word in the sentence using the following formula (1): ｉ Calculate.

[0038]

[0039] Here, the variable p based on the part of speech and basic ambiguity of the word ｉ１For example, the value of ambiguity can take three levels, from 1 to 3, with the larger value indicating higher ambiguity. For example, if the part of speech is a modifier such as an adjective or adverb, the value is set to 3 since it is highly ambiguous, while a demonstrative is set to 2, and others are set to 1. Alternatively, the basic ambiguity may be applied as is, or a value based on the part of speech may be used in combination with the basic ambiguity value.

[0040] Also, the variable p based on the number of meanings of a word ｉ２ For example, the value registered in the fuzzy expression dictionary 141 may be used as is, or the variable p ｉ１ Similarly, the values ​​may be classified into three levels.

[0041] Also, the variable p based on the type of word ｉ３ may be set to a value by the user, for example.

[0042] Also, the ambiguity score r ｉ The weight variable w is set to a value greater than 0.0. １ ~w ３ and variable p ｉ１ ~p ｉ３ is adjusted.

[0043] Additionally, each term in equation (1), or the final ambiguity score, may be normalized to fall within the range of 0 to 1.

[0044] The method for extracting the implicit thoughts lx is not limited to the method described here. For example, the implicit thought extraction unit 152 may extract, as the implicit thoughts lx, a rephrased version of the user-sent information x in a way that clarifies its meaning. For example, the implicit thought extraction unit 152 may extract, as the implicit thoughts lx, "I want to create an AI with high market value" or "I want to create an AI with functions different from conventional AI," which are obtained by rephrasing the "amazing" part of the text "I want to create an amazing AI." Furthermore, the implicit thought extraction unit 152 may identify the specific meaning of "amazing" from the context before and after the information sent by the user. The implicit thought extraction unit 152 may extract the implicit thoughts lx using a language model (e.g., ChatGPT (registered trademark)).

[0045] The generation unit 153 acquires (generates) information (divergent information o) that diverges concepts, opinions, and ideas, taking into account the explicit thoughts ex and implicit thoughts lx (step S104). That is, the generation unit 153 generates the divergent information o, which is text related to the explicit thoughts ex, which are text representing the user's explicit thoughts, and the implicit thoughts lx, which are text representing the user's implicit thoughts.

[0046] The generation unit 153 causes the language model to generate divergent information o, which indicates information in which concepts, opinions, and thoughts diverge, for the explicit thoughts ex and the implicit thoughts lx.

[0047] For example, the generation unit 153 inputs a prompt to the language model, such as, "We are generating ideas for creating new technology. The user has input 'amazing AI'. To encourage divergence of thoughts, please output as many synonyms, antonyms, words that delve deeper, and words from a different angle as possible for the current input. The user has not yet decided on lx, so please leave some room for variation and suggest a variety of words." to generate divergence information o.

[0048] The determination unit 154 calculates the divergence degree d from the explicit thoughts ex, implicit thoughts lx, and divergence information o, and determines whether divergence has been exhausted (step S105). That is, the determination unit 154 determines whether the degree of divergence between the explicit thoughts ex, implicit thoughts lx, and divergence information o is equal to or greater than a threshold. For example, the determination unit 154 determines whether the degree of divergence (divergence degree d) based on the similarity between the explicit thoughts ex, implicit thoughts lx, and divergence information o is equal to or greater than a threshold (whether divergence has occurred).

[0049] For example, the determination unit 154 vectorizes the explicit thoughts ex, implicit thoughts lx, and divergence information o, all of which are text, using a technique such as word2vec, and then calculates the similarity (e.g., cosine similarity or Euclidean distance) or variance between the vectors as the divergence d.

[0050] The determination unit 154 may also determine whether divergence has occurred using a language model. In this case, the determination unit 154 inputs a prompt into the language model, such as, "Can you come up with synonyms, antonyms, words that delve deeper, or words from a different angle for 'amazing AI,' 'harmony with humans,' and 'human-understandable AI'? Please let me know if all of these have already been exhausted." and determines whether the divergence information o has been exhausted (whether divergence has occurred). Note that "amazing AI," "harmony with humans," and "human-understandable AI" are examples of implicit thought lx, explicit thought ex, and divergence information o, respectively.

[0051] The determination unit 154 may also use the result of determining whether or not divergence has occurred by a person as the determination result.

[0052] Furthermore, the determination unit 154 may calculate the similarity or dispersion of the topics, fields, or topics represented by the set of explicit thoughts ex, implicit thoughts lx, and divergent information o as the divergence d.

[0053] For example, the determination unit 154 can extract topics from a set of explicit thoughts ex, implicit thoughts lx, and divergent information o using LDA (Latent Dirichlet Allocation), which is a type of topic analysis method. Then, the determination unit 154 calculates the average vector of all text belonging to each topic and calculates the cosine similarity between the average vectors.

[0054] If the determination unit 154 determines that the divergence has not been completed (No at step S106), the presentation device 10 returns to step S101 and repeats the process.

[0055] If the judgment unit 154 determines that divergence has been exhausted (step S106, Yes), the presentation unit 155 converts the explicit thoughts ex, implicit thoughts lx, and divergence information o into a format that is easy for humans to understand and presents them (step S107).

[0056] The presentation unit 155 presents information via an interface (screen) shown in Fig. 6. Fig. 6 is a diagram showing an example of the interface.

[0057] Screen 200 is output to a display or the like via output unit 12. Screen 200 includes text input fields 211, 212, and 213, each corresponding to a plurality of users. Input unit 11 acquires text entered in text input fields 211, 212, and 213 as user-transmitted information x. When button 214 is pressed, input unit 11 collects voice via a microphone or the like, and acquires the collected voice as user-transmitted information x.

[0058] The presentation unit 155 displays the explicit thoughts ex, implicit thoughts lx, and divergent information o in the area 220. The explicit thoughts ex, implicit thoughts lx, and divergent information o are converted into two-dimensional vectors and placed at positions indicated by the coordinates corresponding to the vectors. In this way, the results of processing by the presentation device 10 are fed back to the user.

[0059] Text 221 and text 222 are explicit thoughts ex and implicit thoughts lx, respectively. Text 223 and text 224 are divergent information o.

[0060] The user-generated information x does not have to be both text and speech. As shown in Figure 7, the user-generated information x may be input only by speech. Figure 7 shows an example of how a user can input information. In this case, both explicit thoughts ex and implicit thoughts lx are extracted from the speech-based text.

[0061] As shown in Figure 8, user-generated information x may be input only as text. Figure 8 shows an example of how a user can input information. In this case, both explicit thoughts ex and implicit thoughts lx are extracted from the input text.

[0062] As shown in Fig. 9, user-contributed information x may be input by multiple users. For example, utterances in a discussion between multiple users are input as user-contributed information x. Fig. 9 is a diagram showing an example of a method for inputting information by a user.

[0063] The implicit thought extraction unit 152 may extract implicit thoughts lx from information previously transmitted by the user. FIG. 10 is a diagram showing an example of how a user inputs information. In the example of FIG. 10, the user-transmitted information x is "I guess what I want to visualize is a vague thought." The explicit thought ex is "cooperate with people."

[0064] Here, it is assumed that the user's past communications are stored in a database. The user's past communications are not limited to inputs to the presentation device 10, but also include speeches in meetings, texts written in diaries, posts on social media, etc. In addition, the user's past communications may also include voice messages, memos, etc. from a life log.

[0065] 11, the implicit thought extraction unit 152 searches the database using the user-sent information x and the explicit thoughts ex as search indexes, and extracts highly relevant text as implicit thoughts lx. FIG. 11 is a diagram showing an example of the extracted text.

[0066] The user may specify which part of the user-transmitted information x to use as the search index. For example, the implicit thought extraction unit 152 accepts an operation to specify "vague" as the search index in the sentence "I kind of want to visualize a vague thought."

[0067] For example, assume that the database contains vectorized versions of the user's past messages. The implicit thought extraction unit 152 uses the vectorized user-transmitted information x and explicit thoughts ex as a search index. The implicit thought extraction unit 152 then extracts from the database the user's past messages whose similarity (e.g., cosine similarity) with the search index is equal to or greater than a threshold.

[0068] The implicit thought extraction unit 152 may extract implicit thoughts lx from the database within a specified period (for example, within one year) via a function on the screen 200 or the like.

[0069] Furthermore, the implicit thought extraction unit 152 may acquire not only the implicit thought lx but also the entire document containing the implicit thought lx and present it to the user as reference information. For example, the implicit thought extraction unit 152 can acquire a transcript of the entire meeting in which the implicit thought lx is included in the speech, or the entire diary entry for a certain day in which the implicit thought lx is included. The user can use the presented reference information to help diverge their thoughts.

[0070] [Effect] As described above, the presentation device 10 includes the generation unit 153 and the determination unit 154. The generation unit 153 generates a third text (divergence information o) related to a first text (explicit thoughts ex) representing a user's explicit thoughts and a second text (implicit thoughts lx) representing the user's implicit thoughts. The determination unit 154 determines whether the degree of divergence among the first text, the second text, and the third text is equal to or greater than a threshold.

[0071] This allows the user to check their implicit thoughts in addition to their explicit thoughts, and further allows them to check divergence information that diverges their explicit and implicit thoughts. Furthermore, the presentation device 10 can generate divergence information while determining whether or not divergence has been sufficient, allowing the user's thoughts to diverge sufficiently.

[0072] Furthermore, the generator 153 causes the language model to generate a third text representing information in which concepts, opinions, and thoughts diverge from the first text and the second text, thereby enabling the presentation device 10 to automatically generate divergent information.

[0073] The determination unit 154 also determines whether the degree of divergence based on the similarity between the first text, the second text, and the third text is equal to or greater than a threshold, thereby enabling the presentation device 10 to quantitatively evaluate whether the user's thoughts have been sufficiently diverged.

[0074] [Program] In one embodiment, the presentation device 10 can be implemented by installing a detection program that executes the above-described processing as package software or online software on a desired computer. For example, by executing the above-described detection program on an information processing device, the information processing device can function as the presentation device 10. The information processing device referred to here includes desktop and notebook personal computers. Other examples of information processing devices include smartphones, tablet terminals, and the like.

[0075] 12 is a diagram showing an example of a computer that executes a detection program. The computer 1000 includes, for example, a memory 1010 and a CPU 1020. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0076] The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012. The ROM 1011 stores, for example, a boot program such as a BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to a hard disk drive 1090. The disk drive interface 1040 is connected to a disk drive 1100. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to, for example, a mouse 1110 and a keyboard 1120. The video adapter 1060 is connected to, for example, a display 1130.

[0077] The hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, a program that defines each process of the presentation device 10 is implemented as a program module 1093 in which computer-executable code is written. The program module 1093 is stored, for example, in the hard disk drive 1090. For example, a program module 1093 for executing processes similar to those of the functional configuration of the presentation device 10 is stored in the hard disk drive 1090. Note that the hard disk drive 1090 may be replaced with an SSD.

[0078] Furthermore, setting data used in the processing of the above-described embodiment is stored as program data 1094, for example, in the memory 1010 or the hard disk drive 1090. The CPU 1020 then reads the program module 1093 or the program data 1094 stored in the memory 1010 or the hard disk drive 1090 into the RAM 1012 as necessary, and executes the processing of the above-described embodiment.

[0079] The program module 1093 and program data 1094 may not necessarily be stored in the hard disk drive 1090, but may also be stored in, for example, a removable storage medium and read by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and program data 1094 may be stored in another computer connected via a network (such as a local area network (LAN) or a wide area network (WAN)). The program module 1093 and program data 1094 may then be read by the CPU 1020 from the other computer via the network interface 1070.

[0080] Although the present invention has been described above as an embodiment, the present invention is not limited to the description and drawings that form part of the disclosure of the present invention. In other words, other embodiments, examples, and operational techniques that can be made by those skilled in the art based on the present invention are all included in the scope of the present invention.

[0081] REFERENCE SIGNS LIST 10 Presentation device 11 Input unit 12 Output unit 13 Communication control unit 14 Storage unit 141 Ambiguous expression dictionary 15 Control unit 151 Explicit thought extraction unit 152 Implicit thought extraction unit 153 Generation unit 154 Determination unit 155 Presentation unit

Claims

1. A presentation device comprising: a generation unit that generates a third text related to a first text representing a user's explicit thoughts and a second text representing the user's implicit thoughts; and a judgment unit that determines whether the degree of divergence between the first text, the second text, and the third text is greater than or equal to a threshold.

2. The presentation device described in claim 1, characterized in that the generation unit causes a language model to generate the third text, which indicates information that diverges from the first text and the second text in terms of concepts, opinions, and thoughts.

3. The presentation device according to claim 1, characterized in that the determination unit determines whether the degree of divergence based on the similarity between the first text, the second text, and the third text is equal to or greater than a threshold value.

4. A presentation method executed by a presentation device, comprising: a generation step of generating a third text related to a first text representing a user's explicit thoughts and a second text representing the user's implicit thoughts; and a determination step of determining whether the degree of divergence between the first text, the second text, and the third text is greater than or equal to a threshold.

Images