JP6869849B2 - Information providing equipment, information providing method, and information providing program - Google Patents

Description

The present invention relates to an information providing device, an information providing method, and an information providing program.

Conventionally, there is known a technique of searching or generating information related to the input information based on the analysis result of the input information, and outputting the searched or generated information as a response. As an example of such a technique, words, sentences, and contexts contained in the input text are converted into a multidimensional vector and analyzed, and based on the analysis result, text similar to the input text or input. A natural language processing technique that infers the text following the text and outputs the analogy result is known.

Japanese Unexamined Patent Publication No. 2015-28625

"Statistical Latent Semantics Analysis by Tobic Model", by Kazumasa Sato Corona Publishing ISBN 978-4-339-02758-7

However, when the above-mentioned prior art is applied as an aid to a rally, the rally may not be carried out efficiently.

For example, it is conceivable to use the above-mentioned conventional technique to determine a participant to make the next remark based on the remarks of the participants, and to have the determined participant perform the moderator. However, in the above-mentioned prior art, only the information that the participant can predict, such as the text similar to the input text and the text following the input text, is output. For this reason, for example, in the above-mentioned conventional technique, as a result of requesting a participant who makes a similar statement, there is a possibility that the participant who efficiently advances the meeting cannot be requested to make a statement.

The present application has been made in view of the above, and an object of the present application is to facilitate the meeting efficiently.

The information providing device according to the present application regards the acquisition unit for acquiring a word set of a utterance or a sentence and the degree of abstraction of the word space to which the word set content is applied as a single complex, and each dimension in the word space. It is characterized by including an analysis unit that analyzes the state between dimensions of the above by indexing by cohomology, and an output unit that outputs information generated based on the analysis result by the analysis unit.

According to one aspect of the embodiment, the rally can be carried out efficiently.

FIG. 1 is a diagram showing an example of analysis processing and determination processing executed by the information providing apparatus according to the embodiment. FIG. 2 is a diagram showing a configuration example of the information providing device according to the embodiment. FIG. 3 is a diagram showing an example of the word space according to the embodiment. FIG. 4 is a diagram showing an example of information registered in the participant attribute database according to the embodiment. FIG. 5 is a flowchart illustrating an example of the flow of the analysis process and the determination process according to the embodiment. FIG. 6 is a diagram showing an example of a hardware configuration.

Hereinafter, the information providing device, the information providing method, and the mode for carrying out the information providing program (hereinafter, referred to as “the embodiment”) according to the present application will be described in detail with reference to the drawings. The information providing device, the information providing method, and the information providing program according to the present application are not limited by this embodiment.

[Embodiment]
[1. An example of an information providing device]
First, an example of an information providing device that executes analysis processing and determination processing will be described with reference to FIG. FIG. 1 is a diagram showing an example of analysis processing and determination processing executed by the information providing apparatus according to the embodiment. In FIG. 1, the information providing device 10 is an information processing device that executes analysis processing and determination processing described below, and is realized by, for example, a server device, a cloud system, or the like.

More specifically, the information providing device 10 includes an input device 100, an output device 200 (see, for example, FIG. 2), and a terminal device 300 via a predetermined network N (see, for example, FIG. 2) such as the Internet. It is possible to communicate with any device (see, for example, FIG. 2).

Here, the input device 100 acquires the remarks of the participants by using a voice acquisition device such as a microphone that acquires the voice. Then, the input device 100 converts the remark into text data by using an arbitrary voice recognition technique, and transmits the converted text data to the information providing device 10. Further, the output device 200 reads out the text data received from the information providing device 10 by using a device that outputs voice such as a speaker. The output device 200 can display the text data received from the information providing device 10 on a predetermined display device. Further, the terminal device 300 is a terminal device used by each participant in the meeting, and is an information processing device capable of outputting arbitrary information received from the information providing device 10. For example, the terminal device 300 can display the information received from the information providing device 10 or output it as voice.

The input device 100, the output device 200, and the terminal device 300 are realized by smart devices such as smartphones and tablets, and information processing devices such as desktop PCs (Personal Computers), notebook PCs, and server devices. The input device 100, the output device 200, and the terminal device 300 may be realized by, for example, the same information processing device, or may be realized by, for example, a device such as a robot.

In the following description, the information providing device 10 can convert the remarks of the participants into text data by using an arbitrary voice recognition technique instead of the input device 100. That is, when the voice data of the participant's speech is acquired from the input device 100, the participant's speech can be converted into text data. Further, the information providing device 10 performs linguistic analysis for analyzing the content of the remarks converted into text data by using techniques such as morphological analysis, w2v (word2vec), and deep learning. As for the language analysis technology, any technology can be applied.

[2. Analysis processing and decision processing executed by the information provider]
Here, a technique is conceivable in which a remark corresponding to a remark in a meeting or the like is output from a robot or the like to facilitate the meeting. However, in the above-mentioned prior art, only the information that the participant can predict, such as the text similar to the input text and the text following the input text, is output. For this reason, for example, the above-mentioned conventional technique may not be able to efficiently advance the meeting.

Therefore, the information providing device 10 executes the following analysis process. First, the information providing device 10 acquires the content of remarks at the meeting. Then, the information providing device 10 regards the degree of abstraction of the word space to which the statement content is applied as a simplex complex, and indexes the state between the dimensions of each dimension in the word space to which the statement content is applied by cohomology. Analyze by.

For example, the information providing device 10 applies the content of remarks to a word space in which an LDA (Latent Dirichlet Allocation) is geometrically formed and a single concept is introduced. As a result, the degree of abstraction of the word space can be regarded as a simplex complex. For example, consider three words as points, the triangular coordinates formed by the lines connecting these points are used as a single word coordinate, and the degree of abstraction of the word space is determined by connecting these single units to make them multidimensional. It can be regarded as a single complex. The word space includes, for example, a semantic space or a probability space which is also called a topic space. For details, refer to pages 31-39 of Non-Patent Document 1.

In the example shown in FIG. 1 (A), three words W1, W2, and W3 are used as points, and the triangular coordinates connecting these points are used as a single word coordinate. Then, a plurality of triangular simple substances are connected to form a single unit having a higher dimension as a triangular pyramid-shaped simple substance. In addition, a plurality of triangular pyramid-shaped simple substances are connected to form a single unit having a higher dimension. In this way, the simple substances are connected, and the word space is formed by the multidimensional simple substances. The information providing device 10 can change the fitting of words in the word space, the combination of simple substances to be connected, and the like for each type of meeting, whereby an appropriate word space can be used for each type of meeting. ..

The word W1, the word W2, and the word W3 are words such as "music", "game", and "play". As shown in (A) of FIG. 1, the state of the word space changes by applying the content of the statement to a single unit of each dimension. The state of the word space to which the content of the remark is applied indicates a state in which the remark history, which is the history of a plurality of remarks made at the meeting, is applied to the word space. There is.

The information providing device 10 analyzes the state between the dimensions of each dimension in the word space by using a boundary operator whose degree of abstraction is regarded as a single complex and lowers the dimension of the word space to which the content of the statement is applied. For example, assume that the number of dimensions of the word space is k (k is an integer). In this case, as shown in FIG. 1B, the information providing device 10 has an abelian group C _k (K), C _k-1 (K), ..., C ₂ (K), C ₁ (K). ), C ₀ (K) the boundary operator _{∂ k, ∂ k-1,} ···, ∂ 2, ∂ 1, the connected chain-complexes in _{∂ 0 C = (C * (} K), ∂ *) operations To do.

The abelian groups C _k (K), C _k-1 (K), ..., C ₂ (K), C ₁ (K), and C ₀ (K) make it possible to connect various words in the meeting speech space. You can grasp the state. The larger the dimension, the higher the degree of abstraction.

Then, a partial vector space represented by the following equations (1) and (2) is formed from the chain complex C. The element of Z _k is called k cycle, and the element of B _k is called k boundary. Relationship between Ker∂ _k and Im∂ _{k + 1} can be expressed as shown in FIG. 1 (C).

Here, cohomology is represented by Ker ∂ _k / Im ∂ _{k + 1} = Z _k / B _k . It should be noted that Im (∂ _k · ∂ _{k + 1} ) = 0, and B _k ⊂ Z _k ⊂ C _k .

The homology group of the chain complex C is a sequence of vector spaces represented by the following equation (4) defined by the quotient vector space represented by the following equation (3). Z _k indicates that there is no boundary, and B _k indicates the boundary of a figure (single substance) that is one-dimensionally higher. Therefore, Z _k / B _k indicates that there is no boundary and is not the boundary of a figure one-dimensionally higher, and it can be said that cohomology corresponds to a hole or cavity of a connection between dimensions in a simple substance of each dimension. Hereinafter, a hole or cavity in which a unit of each dimension is connected between dimensions is referred to as a “dimensional hole”. The _{larger Z k} / B _k , the larger the dimensional hole.

In this way, by analyzing the connections between the dimensions of each dimension in the word space where the degree of abstraction is regarded as a simplex complex using cohomology as an index, and detecting holes or cavities in the connections between the dimensions of each dimension. , The lack of connections between dimensions can be grasped. Therefore, it is possible to grasp the deficient part (hereinafter referred to as deficient element) in the word space, and fill the hole in each dimension by compensating for the deficient element.

For example, if the assembly is brainstorming, the assembly word space described above is an idea space formed by brainstorming. Each participant changes his / her own idea space by listening to the remarks of other participants, and the change in the participant's idea space affects the idea space of the meeting.

Therefore, the information providing device 10 analyzes the connection between the dimensions of each dimension in the idea space using cohomology as an index, and detects a hole in the dimension in the idea space. By grasping such a dimensional hole, it is possible to make a statement to fill the dimensional hole, which makes it possible to efficiently generate an idea that can be generated without the dimensional hole at a meeting.

In addition, the information providing device 10 can analyze changes in states between dimensions in the word space. For example, as shown in FIG. 1D, the state of the word space changes to states ST1, ST2, ST3, and so on. In this case, the information providing device 10 analyzes how the dimensional holes are reduced by comparing the cohomology of each dimension in the word space of the state ST1 with the cohomology of each dimension in the word space of the state ST2. can do. As a result, it is possible to easily grasp whether or not the meeting is being held efficiently.

For example, when the meeting is brainstorming, it is possible to grasp whether the idea is properly developed by grasping the state in which the number of dimensional holes in the idea space is changing. The information providing device 10 can identify, for example, a participant who has made a statement that reduces a dimensional hole in the idea space, and can provide the information of the identified participant to the participants of the meeting.

The information providing device 10 analyzes a state or change between dimensions of each dimension in the word space each time a participant of the meeting speaks, but the embodiment is not limited to this. For example, the information providing device 10 can perform the above-mentioned analysis process at regular time intervals, and can also perform the above-mentioned analysis process at intervals according to the amount of change in the number of dimensional holes.

When the information providing device 10 executes the above-mentioned analysis process, the information providing device 10 then executes a determination process. Based on the result of the analysis process, the information providing device 10 determines a speaker who makes up for the lacking element in the meeting.

For example, the information providing device 10 determines a speaking subject to supplement the deficient element based on the relationship between the dimensional hole in the word space and the attributes of the participants, and outputs information for making the determined speaking subject speak. Can be done. As a result, the possibility of compensating for the deficiency can be increased, and the meeting can be carried out efficiently.

Specifically, the information providing device 10 has demographic attributes such as the gender and age of the candidate, psychographic attributes such as preference, and the academic field in which each candidate is specialized in research, that is, each candidate. Collect the specialty of the person as an attribute. Then, the information providing device 10 calculates the similarity between the attribute of the dimensional hole in the word space and the attribute of the participant, and the calculated similarity determines the participant as a speaker who supplements the above-mentioned deficiency element. The attributes of the dimensional holes are represented by the relationships between the dimensions. For example, k dimensional hole shown in (C) of FIG. 1 is shown in relation to the abelian group Abelian group _{C k (K) C k-} 1 (K).

Further, the information providing device 10 calculates the similarity with the attributes of the participants for each of the two dimensions related to the hole of the dimension, and the participants having a high degree of similarity for each dimension of the two dimensions are described above. It can also be decided as a speaker to make up for the deficiency.

Further, the information providing device 10 stores the change of the state between the dimensions of each dimension in the word space as a state change history, and is insufficient so that the hole of the dimension can be smoothly filled based on the state change history. It can be determined as a speaker who supplements the elements. For example, the information providing device 10 determines the position of the word space whose state is updated according to the content of the statement from the latest predetermined time before to the present, or the hole of the dimension near the area as the hole of the dimension to be filled next. It can be determined as a speaker to supplement the deficient element corresponding to the determined dimensional hole. As a result, it is possible to prevent the topic from suddenly changing to a completely different field at the meeting, and the meeting can proceed smoothly and efficiently.

The method of determining the speaker who supplements the deficient element is not limited to the above-mentioned example, and various methods can be used so that the deficient element can be supplemented. For example, the information providing device 10 can analyze the remarks of the participants in the past meetings and determine the participants who have made remarks that fill the dimensional holes as the remark subject that supplements the above-mentioned deficiency elements.

In addition, the information providing device 10 can output information indicating the result of the above-mentioned analysis process to the participants. For example, the information providing device 10 can output information indicating a state between dimensions of each dimension in the word space and information indicating a change in the state between dimensions of each dimension in the word space. As a result, the participant can grasp the state between the dimensions of each dimension in the word space or the change of such a state, can grasp the state of the assembly and the change of such a state, and can advance the assembly efficiently. be able to.

Further, the information providing device 10 can determine a deficient element in the assembly as a state between the dimensions of each dimension in the word space, and output support information for supplementing the hole of the dimension corresponding to the determined deficient element. As a result, support information can be presented to the participants, and the meeting can be promoted more efficiently.

The information providing device 10 stores, for example, information useful for compensating for missing elements (for example, one or more keywords or sentences) in the storage unit 30 (for example, see FIG. 2) as support information for each hole in each dimension. be able to. The information providing device 10 can read out from the storage unit 30 and output the support information corresponding to the dimensional hole among the insufficient information stored in the storage unit 30.

In addition, the information providing device 10 can output information that identifies a participant who has made a statement that reduces the number of dimensional holes in the word space by a predetermined number or more. As a result, the participants can be motivated to speak, and the meeting can be carried out efficiently.

[3. An example of calculation in the analysis process executed by the information providing device]
Here, an example of the calculation in the analysis process executed by the information providing device 10 will be described. FIG. 3 is a diagram showing an example of the word space according to the embodiment. In the following, in order to make the explanation easy to understand, the word space to which the remarks of the meeting are applied is assumed to be the word space 50 shown in FIG. 3, but in principle, the word space in the information providing device 10 is shown in the figure. It is higher than the word space 50 shown in 3.

3 represents the coordinate position of the words W1, W2, W3, W4 as the position _{P 1, P 2, P 3} , P 4 , connecting two of the coordinate positions _{P 1, P 2, P 3} , P 4 The line is represented as the line l ₁ , l ₂ , l ₃ , l ₄ , l ₅ , and the surface formed by the three lines l ₁ , l ₂ , l _{3 is represented as the surface S.} The information providing device 10 regards the degree of abstraction of the word space 50 as a single complex, and analyzes the state between the dimensions of each dimension in the word space 50 by indexing by cohomology.

With respect to the word space 50, the abelian groups C ₂ (K), C ₁ (K), and C ₀ (K) can be expressed as shown in the following equations (5) to (7). The information providing device 10 performs the operations shown in the following equations (5) to (7). In addition, "Z" in equations (5) to (7) represents a set Z.

Further, the surface S can be expressed as shown in the following equations (8) to (12) by using the boundary operators ∂ ₂ , ∂ ₁ , and ∂ _0. The information providing device 10 performs the operations shown in the following equations (8) to (12).

_{The cohomologies H 2} (C _* ), H ₁ (C _* ), and H ₀ (C _* ) in the homology group H _* (C _* ) represented by the following formula (13) are expressed in the following formulas (13) to (15). It can be expressed as shown in. The information providing device 10 performs the operations shown in the following equations (13) to (15).

In addition, "A" and "B" in the above formulas (13) to (15) can be expressed as shown in the following formulas (16) and (17).

In this way, the information providing device 10 can determine the hole of the dimension by the _{calculated cohomology H 2} (C _* ), H ₁ (C _* ), H ₀ (C _{*) in each dimension.} For example, in the example shown in the above equation (13), H ₂ (C _* ) is zero, so there is no two-dimensional hole. Further, in the example shown in the above equation (14), since H ₁ (C _* ) is not zero, there is a one-dimensional hole. The information providing device 10 _{can determine that there is a k-dimensional hole when, for example, the absolute value of the cohomology H k} (C _* ) is equal to or greater than the threshold value Hth. The threshold value Hth is, for example, a value larger than 0, and can be changed for each dimension.

[4. An example of processing executed by the information providing device]
Next, an example of the analysis process and the determination process executed by the information providing device 10 will be described with reference to FIG. First, an example of the analysis process executed by the information providing device 10 will be described.

First, the information providing device 10 generates a word space to which the content of the statement is applied (step S1). For example, as shown in FIG. 1A, the information providing device 10 geometrically forms an LDA to generate a word space in which a single concept is introduced. It is only necessary to be able to generate a word space in which the degree of abstraction can be regarded as a simplex complex, and the word space may be formed by an analysis method other than LDA.

The information providing device 10 receives the remarks of the participants (step S2). Then, the information providing device 10 updates the state of the word space based on the content of the acquired remark (step S3). For example, the information providing device 10 analyzes the content of the remark by using techniques such as morphological analysis, w2v, and deep learning, assigns the content of the remark to the word space, and updates the state of the word space.

Further, the information providing device 10 regards the degree of abstraction of the word space as a single complex, and analyzes the state between the dimensions of each dimension in the word space by indexing by cohomology (step S4). In step S4, the information providing device 10 uses cohomology to detect holes in each dimension in the word space.

For example, in the state (C) shown in FIG. 1, since | Z _k / B _k |> 0, there is a hole in the k dimension. In this case, the information providing device 10 creates a hole in the k dimension. Detected. The information providing device 10 performs a process of detecting a dimensional hole for all dimensions, but detects a dimensional hole only between the dimensions corresponding to the dimensional hole detected in the previous process. Processing can also be performed. As a result, the processing load in the analysis process can be reduced.

Next, an example of the determination process executed by the information providing device 10 will be described. The information providing device 10 determines a speaking subject to supplement the deficient element based on the result of the analysis process described above (step S5). For example, the information providing device 10 makes a statement to supplement the deficiency element among the participants # 1 to # 3, when the participants # 1 to # 3 are participating in the meeting. Determine as the subject.

Here, it is assumed that the specialized field of participant # 1 is field # 1, the specialized field of participant # 2 is field # 2, and the specialized field of participant # 3 is field # 3. In this case, the information providing device 10 calculates the similarity between the attribute of the hole in the dimension and the attribute of each participant # 1 to # 3, and supplements the participants with high similarity in each dimension with the lacking element in the word space. Determine as the speaker.

Then, the information providing device 10 prompts the determined participant to speak (step S6). For example, the information providing device 10 causes an output device 200 such as a robot to make a statement that prompts a determined participant to speak. As a result, the information providing device 10 can prompt the speaker to make up for the lacking element in the word space, and can efficiently advance the meeting.

[5. Configuration of information providing device]
Hereinafter, an example of the functional configuration of the information providing device 10 that realizes the above-mentioned analysis processing and determination processing will be described. FIG. 2 is a diagram showing a configuration example of the information providing device according to the embodiment. As shown in FIG. 2, the information providing device 10 includes a communication unit 20, a storage unit 30, and a control unit 40.

The communication unit 20 is realized by, for example, a NIC (Network Interface Card) or the like. Then, the communication unit 20 is connected to the network N by wire or wirelessly, and transmits / receives information to / from the input device 100, the output device 200, and the terminal device 300.

The storage unit 30 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. In addition, the storage unit 30 stores the participant attribute database 31.

The attributes of the participants of the meeting are registered in the participant attribute database 31. For example, FIG. 4 is a diagram showing an example of information registered in the participant attribute database according to the embodiment. As shown in FIG. 4, information having items such as "participant ID (Identifier)" and "specialty field" is registered in the participant attribute database 31.

Here, the "participant ID" is an identifier for identifying a participant. Further, the "specialty field" is information indicating the specialty field of the participant indicated by the associated participant ID. In addition to the items shown in FIG. 4, various attribute information such as demographic attributes and psychographic attributes of participants may be registered in the participant attribute database 31.

For example, in the example shown in FIG. 4, the participant ID “participant # 1” and the specialized field “field # 1” are registered in association with each other in the participant attribute database 31. Such information indicates that the specialized field of the participant indicated by the participant ID "participant # 1" is the specialized field "field # 1". In the example shown in FIG. 3, conceptual values such as "field # 1" and the like are described, but in reality, text data and numerical values indicating the field are registered.

Returning to FIG. 2, the explanation will be continued. The control unit 40 is a controller, and for example, various programs stored in a storage device inside the information providing device 10 by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) are stored in a RAM or the like. Is realized by executing as a work area. Further, the control unit 40 is a controller, and may be realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 2, the control unit 40 has an acquisition unit 41, an analysis unit 42, a determination unit 43, and an output unit 44, and executes the analysis process and the determination process described above. More specifically, the acquisition unit 41 and the analysis unit 42 execute the above-mentioned analysis process, and the determination unit 43 executes the above-mentioned determination process.

The acquisition unit 41 acquires various types of information necessary for executing the analysis process. More specifically, the acquisition unit 41 acquires a word set of an utterance or a sentence. For example, the acquisition unit 41 acquires information about a participant or a meeting of the information providing device 10 as a parameter. For example, the acquisition unit 41 acquires the attribute information of the participant from the input device 100 or the terminal device 300 and registers it in the participant attribute database 31. Further, the acquisition unit 41 acquires information (for example, voice data or text data) indicating the remarks of the participants of the meeting from the input device 100 as a word set.

The analysis unit 42 updates the state of the word space based on the content of the statement acquired by the acquisition unit 41. The analysis unit 42 regards the degree of abstraction of the word space as a single complex, and analyzes the state between the dimensions of each dimension in the word space by indexing by cohomology.

Specifically, the analysis unit 42, for example, geometrically transforms the LDA and arranges it in a word space in which a single concept is introduced by vector-converting the content of the statement using a language analysis technique such as w2v, and arranges the state of the word space. To update. It suffices if the content of the statement can be applied to the word space, and the analysis unit 42 can update the state of the word space by a process other than the above-mentioned process.

Further, the analysis unit 42 analyzes the state between the dimensions of each dimension of the word space by using the boundary operator that lowers the dimension of the word space. Then, the analysis unit 42 calculates the cohomology of each dimension to determine the hole in the dimension.

The analysis unit 42 can analyze the change in the state between the dimensions of each dimension in the word space each time the participants of the meeting speak. The analysis unit 42 can perform such an analysis at regular time intervals, and can perform such an analysis at intervals according to the amount of change in the number of dimensional holes.

In addition, the analysis unit 42 can determine the support information that supplements the lacking element in the meeting based on the state between the dimensions of each dimension in the word space. For example, the analysis unit 42 can determine as support information an element necessary for filling a hole in the determined dimension. The support information is, for example, one or more keywords or sentences that can fill in the lack of connection between dimensions corresponding to the holes in the dimensions. Instead of keywords and sentences, images such as still images and moving images can be used as support information.

The analysis unit 42 can determine the support information based on the information stored in the storage unit 30. For example, the storage unit 30 can store a database in which a combination of two consecutive dimensions (for example, k-dimensional and k-1 dimension) and support information are associated with each other. The analysis unit 42 can determine the support information by acquiring the support information associated with the combination of dimensions corresponding to the dimensional holes from the storage unit 30.

Further, when the dimensional holes are continuous over three or more consecutive dimensions, the analysis unit 42 can also determine the elements necessary for filling the holes of the continuous dimensions as support information. In this case, the storage unit 30 can store a database in which a combination of three or more consecutive dimensions (for example, k + 1 dimension, k dimension, and k-1 dimension) and support information are associated with each other. The analysis unit 42 can determine the support information by acquiring the support information associated with the combination of three or more dimensions corresponding to the holes of the continuous dimension from the storage unit 30.

The determination unit 43 determines a speaker who makes up for the lacking element. For example, the determination unit 43 determines a speaker who supplements the deficient element based on the relationship between the attribute of the dimensional hole in the word space and the attribute of the participant, and outputs information for making the determined speaker speak. be able to. The determination unit 43 acquires the attributes of each participant from the participant attribute database 31, calculates the similarity between the attribute of the dimensional hole in the word space and the attribute of the participant, and calculates the similarity of 1 or more. Participants can be determined as the speakers who make up for the deficiencies.

The attributes of the holes in each dimension are registered in the dimension attribute database (not shown) stored in the storage unit 30, and the determination unit 43 is the attributes of the dimension having holes from the dimension attribute database stored in the storage unit 30. Can be obtained. The attribute of the hole of the dimension is an attribute of information necessary for connecting two consecutive dimensions, and is, for example, an attribute of a combination of two consecutive dimensions (for example, k dimension and k-1 dimension). is there.

In the above-mentioned example, the determination unit 43 determines the similarity of the participants for each hole, but can determine the similarity of the participants for two or more consecutive holes. For example, the determination unit 43 calculates the similarity between the attributes corresponding to two or more consecutive holes and the attributes of the participants, and the calculated similarity determines the participants as the speakers who supplement the above-mentioned deficiency factors. You can also do it.

In addition, the determination unit 43 can identify the participants who have made statements that reduce the dimensional holes in the word space. The determination unit 43 can also specify the dimension of the hole filled by the specified participant. The determination unit 43 can store the identified participant and dimension information in the storage unit 30. The determination unit 43 can identify, for example, the participant who made a statement that reduces the hole in the dimension and the dimension of the hole filled by the participant, based on the change in the state between the dimensions of each dimension in the word space. ..

The output unit 44 outputs information for making the speaking subject determined by the determining unit 43 speak. For example, in the output unit 44, when the determination unit 43 selects participant # 1 as a speaker among the participants, participant # 1 such as "Participant # 1, do you have any thoughts?" The output device 200 is made to utter a remark that urges the output device 200 to remark. As a result, the information providing device 10 can increase the possibility of obtaining a statement that fills a dimensional hole.

Further, the output unit 44 can output the information for making the speaking subject determined by the determining unit 43 speak, including the support information determined by the determining unit 43. For example, if the support information is "○○○" ("○○○" is a keyword or sentence), participation such as "Participant # 1, do you have any thoughts about" ○○○ "?" The output device 200 can utter a statement prompting the person # 1 to speak.

Further, the output unit 44 can make the output device 200 utter support information without including the information for making the speaking subject determined by the determination unit 43 speak. Further, the support information may include an image such as a still image or a moving image, and may not include text information. In this case, the output unit 44 can provide the support information to the participants by displaying the support information on the output device 200.

In addition, the output unit 44 can output information (for example, information on missing elements) indicating a state between dimensions of each dimension in the word space from the output device 200. As a result, the participants of the rally can grasp the state of speech from the start of the rally to the present. Further, the output unit 44 can also output information indicating a change in the state between the dimensions of each dimension in the word space (for example, information indicating a change in the missing element) from the output device 200. As a result, the participants of the rally can grasp the degree of decrease in the number of dimensional holes as the remarks in the rally progress, and if the degree of decrease in the number of dimensional holes is small, fill the dimensional holes. It can be easily grasped that there are few remarks for.

The output unit 44 can output the support information corresponding to the hole of a higher dimension from the output device 200 with priority. For example, suppose that there are a k + 1 dimensional hole, a k dimensional hole, and a k-1 dimensional hole as continuous dimensional holes. In this case, the output unit 44 preferentially outputs the support information corresponding to the k + 1 dimensional hole from the output device 200. As a result, the concept in the word space can be presented to the participants in the order of the support information corresponding to the higher dimension to the support information corresponding to the lower dimension, and the meeting can be efficiently advanced.

Further, when there are two or more holes in the same dimension, the output unit 44 can preferentially output the support information corresponding to the upper dimension having many holes in the lower dimension from the output device 200. As a result, it is possible to give priority to the support information for the part where there are many insufficient elements in the word space, and it is possible to efficiently proceed with the meeting.

The continuous dimension is, for example, a combination of dimensions having a relationship of _{"S" and "l 1} ", "S" and "l ₂ ", or "S" and "l _{3" shown in FIG.} .. Further, the three or more consecutive dimensions have, for example, the relationship of _{"S", "l 1} ", and "P ₁ ", or "S", "l ₁ ", and "P _{2" shown in FIG.} It is a combination of certain dimensions.

The output unit 44 may output the information of the participant specified by the determination unit 43 to reduce the dimensional hole and the dimensional information of the hole filled by the specified participant from the output device 200. it can. As a result, the participants can be motivated to speak, and the meeting can be carried out efficiently. In the above-mentioned example, an example of a process of acquiring the content of a statement at a meeting and outputting support information for efficiently advancing the meeting has been described as a word set, but the embodiment is limited to this. It's not a thing. For example, the information providing device 10 executes the above-described processing for an arbitrary number of arbitrary contents that are a set of words, such as a treatise, news, and commentary, and provides support for a part having many missing elements in the word space of the contents. Information may be provided.

[6. An example of the flow of processing executed by the information providing device]
Next, an example of the flow of the analysis process and the determination process executed by the information providing device 10 will be described with reference to FIG. FIG. 5 is a flowchart illustrating an example of the flow of the analysis process and the determination process according to the embodiment.

First, the information providing device 10 acquires the remarks of the participants at the meeting and updates the state of the word space (step S101). The process of updating the state of the word space is performed, for example, every time a participant's remark is acquired.

The information providing device 10 regards the degree of abstraction of the word space to which the content of the statement is applied as a single complex, and analyzes the state between the dimensions of each dimension in the word space by indexing by cohomology (step S102). The information providing device 10 determines the speaking subject based on the analysis result in step S102 (step S103). Then, the information providing device 10 outputs information prompting the determined speaker to speak (step S104), and ends the process.

[7. Others]
Further, among the processes described in the above-described embodiment, all or a part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed can be performed. All or part of it can be done automatically by a known method. In addition, the processing procedure, specific name, and information including various data and parameters shown in the above text and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

Further, each component of each of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of them may be functionally or physically distributed / physically in arbitrary units according to various loads and usage conditions. Can be integrated and configured.

Further, the above-described embodiments can be appropriately combined as long as the processing contents do not contradict each other.

[8. program〕
Further, the information providing device 10 according to the above-described embodiment is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 6 is a diagram showing an example of a hardware configuration. The computer 1000 is connected to the output device 1010 and the input device 1020, and the arithmetic unit 1030, the primary storage device 1040, the secondary storage device 1050, the output IF (Interface) 1060, the input IF 1070, and the network IF 1080 are connected by the bus 1090. Has.

The arithmetic unit 1030 operates based on a program stored in the primary storage device 1040 or the secondary storage device 1050, a program read from the input device 1020, or the like, and executes various processes. The primary storage device 1040 is a memory device that temporarily stores data used by the arithmetic unit 1030 for various calculations, such as a RAM. Further, the secondary storage device 1050 is a storage device in which data used by the calculation device 1030 for various calculations and various databases are registered, and is realized by a ROM (Read Only Memory), an HDD, a flash memory, or the like.

The output IF 1060 is an interface for transmitting information to be output to an output device 1010 that outputs various information such as a monitor and a printer. For example, USB (Universal Serial Bus), DVI (Digital Visual Interface), and the like. It is realized by a connector of a standard such as HDMI (registered trademark) (High Definition Multimedia Interface). Further, the input IF 1070 is an interface for receiving information from various input devices 1020 such as a mouse, a keyboard, and a scanner, and is realized by, for example, USB.

The input device 1020 is, for example, an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), or a tape. It may be a device that reads information from a medium, a magnetic recording medium, a semiconductor memory, or the like. Further, the input device 1020 may be an external storage medium such as a USB memory.

The network IF1080 receives data from another device via the network N and sends it to the arithmetic unit 1030, and also transmits the data generated by the arithmetic unit 1030 to the other device via the network N.

The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output IF 1060 and the input IF 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040, and executes the loaded program.

For example, when the computer 1000 functions as the information providing device 10, the arithmetic unit 1030 of the computer 1000 realizes the function of the control unit 40 by executing the program loaded on the primary storage device 1040.

[9. effect〕
As described above, the information providing device 10 regards the acquisition unit 41 that acquires the word set of the utterance or the sentence and the degree of abstraction of the word space to which the word set content is applied as a single complex, and in the word space. It includes an analysis unit 42 that analyzes the state between dimensions of each dimension by indexing by cohomology, and a 44 output unit that outputs information generated based on the analysis result by the analysis unit 42. For example, the information providing device 10 acquires the content of remarks at the meeting as a word set. As a result, the participants of the rally can grasp the lack of connection between the dimensions and can carry out the rally efficiently.

Further, the analysis unit 42 determines the support information that supplements the lacking element in the meeting based on the state between the dimensions of each dimension in the word space, and the output unit 44 outputs the information of the support information determined by the analysis unit 42. To do. As a result, support information can be presented to the participants, and the meeting can be promoted more efficiently.

In addition, the information providing device 10 includes a determination unit 43 that determines a speaking subject that supplements the deficient element. The output unit 44 outputs information for making the speaking subject determined by the determining unit 43 speak. As a result, the possibility of compensating for the deficiency can be increased, and the meeting can be carried out efficiently.

Further, the analysis unit 42 determines the support information that supplements the lacking element in the meeting based on the state between the dimensions of each dimension in the word space, and the output unit 44 includes the support information in the information for making the speaker speak. And output. As a result, the possibility of compensating for the deficiency element can be further increased, and the meeting can be carried out efficiently.

Further, the analysis unit 42 analyzes the change in the state between the dimensions of each dimension in the word space, and the output unit 44 outputs information indicating the change in the state between the dimensions analyzed by the analysis unit 42. As a result, it is possible to grasp whether or not the rally is held so that the number of dimensional holes is reduced, and it is possible to easily grasp whether or not the rally is held efficiently.

Although some of the embodiments of the present application have been described in detail with reference to the drawings, these are examples, and various modifications are made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure column of the invention. It is possible to practice the present invention in other improved forms.

Further, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the analysis unit can be read as an analysis means or an analysis circuit.

10 Information providing device 20 Communication unit 30 Storage unit 31 Participant attribute database 40 Control unit 41 Acquisition unit 42 Analysis unit 43 Decision unit 44 Output unit 100 Input device 200 Output device 300 Terminal device

Claims (8)

An acquisition unit that acquires a word set of an utterance or a sentence,
An analysis unit that considers the degree of abstraction of the word space to which the word set content is applied as a single complex and analyzes the state between dimensions of each dimension in the word space by indexing by cohomology.
An information providing device including an output unit that outputs information generated based on an analysis result by the analysis unit. The information providing device according to claim 1, wherein the word set is the content of a statement at a meeting. The analysis unit
Based on the state between the dimensions of each dimension, the support information that supplements the lacking element in the meeting is determined.
The output unit
The information providing device according to claim 2, wherein the information of the support information determined by the analysis unit is output. It is equipped with a decision-making unit that determines the speaker who makes up for the shortage in the meeting.
The output unit
The information providing device according to claim 2, wherein the decision unit outputs information for making the determined speaker speak. The analysis unit
Based on the state between the dimensions of each dimension, the support information that supplements the deficiency element is determined.
The output unit
The information providing device according to claim 4, wherein the support information is included in the information for making the speaking subject speak. The analysis unit
Analyzing the change of state between the dimensions of each dimension,
The output unit
The information providing device according to any one of claims 1 to 5, wherein information indicating a change in a state between the dimensions of each dimension is output. A computer-executed method of providing information
The acquisition process to acquire the word set of utterances or sentences,
An analysis process in which the degree of abstraction of the word space to which the word set content is applied is regarded as a single complex, and the state between the dimensions of each dimension in the word space is analyzed by indexing by cohomology.
An output process that outputs information generated based on the analysis results of the analysis process, and
A method of providing information, which is characterized by including. The acquisition procedure to acquire the word set of utterances or sentences, and
An analysis procedure in which the degree of abstraction of the word space to which the word set content is applied is regarded as a single complex, and the state between dimensions of each dimension in the word space is analyzed by indexing by cohomology, and an analysis procedure.
An output procedure that outputs information generated based on the analysis result of the analysis procedure, and an output procedure.
An information providing program that causes a computer to execute.

Images