WO2017104656A1

WO2017104656A1 - Information processing device, information processing method, and recording medium

Info

Publication number: WO2017104656A1
Application number: PCT/JP2016/087046
Authority: WO
Inventors: 大地木村; 英司平尾; 俊輔河野
Original assignee: 日本電気株式会社
Priority date: 2015-12-14
Filing date: 2016-12-13
Publication date: 2017-06-22
Also published as: JPWO2017104656A1

Abstract

The present invention provides information with which it is possible to evaluate the consistency of a result in which a plurality of causal relationships are linked. An information processing device provided with a link extraction unit for extracting a link of causal relationships comprising at least two causal relationships from information that represents elements constituting an event and causal relationship information that includes information representing a causal relationship between the elements, a causal relationship calculation unit for calculating a causal relationship between the elements at both ends of the extracted link of causal relationships from at least two causal relationships constituting the extracted link of causal relationships, and a result output unit for outputting the result of the calculation by the causal relationship calculation unit.

Description

Information processing apparatus, information processing method, and recording medium

The present invention relates to a technique for analyzing a causal relationship between elements constituting an event.

There are known systems that visualize the elements that make up an event and the causal relationship between each element. For example, Patent Document 1 discloses a technique for generating a causal relationship diagram based on whether the relationship between elements constituting the system is increasing or decreasing. Specifically, the technique disclosed in Patent Document 1 visualizes a causal relationship between elements by using a coincidence rate between a change in a cause-side element and a change in a result-side element.

Patent Literature 2 collects a causal relationship pair including a cause phrase and a result phrase, and a causal relationship pair, and another causal relationship pair having a cause phrase that is causally related to the result phrase of the causal relationship pair. A technique for generating a chain of causal relationships by linking.

Patent Document 3 describes a technique for describing a structure of a causal relationship between nodes by connecting nodes (elements) constituting the system by an arc and generating a simulation program using a simulation model inheriting the structure. Is disclosed.

JP 2013-130946 A Japanese Patent Laying-Open No. 2015-121897 JP 2007-286777 A

Referring to information representing the causal relationship visualized, there are cases where current status analysis and problem solving methods are examined. In this case, if the definition of the causal relationship is not correct, there arises a problem that consistency between the causal element and the resulting element cannot be achieved in a plurality of causal relation chains. In other words, when a plurality of causal relationships are chained, if the causal relationships are not correctly interpreted, there is a possibility that the cause element and the resulting element do not match.

As an example, assume a situation (event) in which a person is identified from the crowd using image recognition. If there is a lot of time for person identification, the identification rate of the person is improved. For example, as shown in FIG. 1A, this causal relationship is expressed using nodes of “person identification time” and “person identification rate” and directional links that connect these nodes. Also, the more crowded crowds, the longer the time required for person identification. Similar to the above, this causal relationship is expressed by using nodes of “congestion” and “person identification time” and directional links connecting these nodes (FIG. 1B). FIG. 1C shows the result of combining (chaining) the causal relationships shown in FIG. 1A and FIG. 1B. The result of the chain of causal relationships shown in FIG. 1C leads to a conclusion different from the normal assumption that the specific rate of a person improves as the congestion increases. In the above example, the term “person identification time” is used in a different meaning in each causal relationship, resulting in a problem that results of chaining a plurality of causal relationships are not consistent.

On the other hand, with the techniques disclosed in the above patent documents, it is not always possible to correctly evaluate the consistency of the result of linking a plurality of causal relationships. That is, the technique disclosed in Patent Document 1 only generates a causal relationship diagram representing the relationship between elements based on past data regarding each element. The technique disclosed in Patent Document 2 is a technique for generating a scenario by selecting a causal relationship pair that maintains a predetermined causal relationship. When the cause phrase and the result phrase are the same as the character string, the technique disclosed in Patent Document 2 is considered to be regarded as the same, so that the same character string represents different meanings as described above. However, it is not always possible to evaluate causality appropriately. The technique disclosed in Patent Document 3 only generates a simulation model from a causal description format that is easy for humans to understand.

The present invention has been made in view of the above circumstances. That is, the present invention has as one of its main objects to provide an information processing apparatus or the like that provides information that can be used to evaluate the consistency of results obtained by chaining a plurality of causal relationships.

In order to achieve the above object, an information processing apparatus according to an aspect of the present invention provides at least two causal factors from causal relationship information including information representing elements constituting an event and information representing causal relationships between elements. A causal relationship between elements at both ends of the extracted chain of causal relationships is calculated from a chain extraction unit that extracts a chain of causal relationships consisting of relationships and at least two causal relationships that constitute the chain of extracted causal relationships. A causal relationship calculation unit, and a result output unit that outputs a result of the calculation by the causal relationship calculation unit.

Further, the information processing method according to one aspect of the present invention is based on causal relation information including at least two causal relations from causal relation information including information representing elements constituting an event and information representing causal relations between elements. A chain is extracted, and the causal relationship between the elements at both ends of the extracted causal relationship is calculated from at least two causal relationships constituting the extracted causal relationship chain, and the calculation result of the causal relationship is output. .

The object is also achieved by an information processing apparatus having the above-described configuration, a computer program for realizing the information processing method by a computer, a computer-readable recording medium storing the computer program, and the like.

According to the present invention, it is possible to provide information that can evaluate the consistency of the result of chaining a plurality of causal relationships.

FIG. 1A is a diagram illustrating an example of a causal relationship. FIG. 1B is a diagram illustrating another example of the causal relationship. FIG. 1C is a diagram illustrating an example of a causal chain. FIG. 2 is a block diagram illustrating a functional configuration of the information processing apparatus according to the first embodiment of this invention. FIG. 3 is a flowchart illustrating an example of processing performed by the information processing apparatus according to the first embodiment of this invention. FIG. 4 is an explanatory diagram illustrating an example of a method for representing information representing elements constituting an event and a causal relationship between the elements in the first embodiment of the present invention. FIG. 5 is an explanatory diagram showing another example of a method for representing information representing elements constituting an event and a causal relationship between the elements in the first embodiment of the present invention. FIG. 6 is an explanatory diagram illustrating an example of a chain of causal relationships extracted from information in the first embodiment of the present invention. FIG. 7 is an explanatory diagram illustrating an example of a result of calculating a causal relationship between elements arranged at both ends in a chain of causal relationships in the first embodiment of the present invention. FIG. 8 is a block diagram illustrating a functional configuration of the information processing apparatus according to the second embodiment of this invention. FIG. 9 is a flowchart showing an example of processing by the information processing apparatus according to the second embodiment of the present invention. FIG. 10 is an explanatory diagram showing an example of a consistency analysis result in the second embodiment of the present invention. FIG. 11 is an explanatory diagram illustrating a chain of causal relationships configured from a plurality of causal relationships in the modified examples of the first and second embodiments of the present invention. FIG. 12 is a block diagram illustrating a functional configuration of the information processing apparatus according to the third embodiment of this invention. FIG. 13 is a block diagram illustrating a hardware configuration capable of realizing the information processing apparatus according to each embodiment of the present invention.

Hereinafter, an information processing apparatus and the like that can solve the above-described problems will be described in detail using each embodiment. In addition, the structure of the information processing apparatus described in each following embodiment is an illustration, and the technical scope of this invention is not limited to them. The division (for example, division | segmentation by a functional unit) of the component which comprises the information processing apparatus in each following embodiment is an example which can implement | achieve the said information processing apparatus. The implementation of the information processing apparatus is not limited to the following examples, and various configurations are assumed. That is, the constituent elements constituting the information processing apparatus in each of the following embodiments may be further divided, or one or more constituent elements may be integrated.

The information processing device described below may be configured using a single device (physical or virtual device), and may be realized using a plurality of separated devices (physical or virtual devices). Also good. When the information processing apparatus includes a plurality of apparatuses, each apparatus may be connected to be communicable via a wired network, a wireless network, or a communication network (communication line) appropriately combining them. Such a communication network may be a physical communication network or a virtual communication network. An information processing apparatus described below or a hardware configuration capable of realizing the components will be described later.

<First Embodiment>
[Description of configuration]
The first embodiment of the present invention will be described below in detail with reference to the drawings. The components described in the following embodiments are merely examples, and the technical scope of the present invention is not limited thereto.

FIG. 2 is a block diagram illustrating a functional configuration of the information processing apparatus 100 according to the first embodiment of the present invention. The information processing apparatus 100 is an apparatus that operates according to a software program (computer program) such as a computer.

The information processing apparatus 100 includes an input reception unit 101, a chain extraction unit 102, a causal relationship calculation unit 103, and a result output unit 104.

The input accepting unit 101 (input accepting means) accepts, as an input, information representing the elements constituting the event and the causal relationship between the elements. Hereinafter, an element constituting an event may be simply referred to as an “element”. In addition, information representing elements and causal relationships between elements may be described as “causal relationship information”.

In the present embodiment, the above-described event represents, for example, something (such as a phenomenon) that can occur in a real environment or a virtual environment such as an information processing apparatus. The elements constituting the event may include, for example, an element indicating a cause related to the event and an element indicating a result corresponding to the cause. In the causal relationship information, the elements constituting the event are described or expressed as, for example, some form of data.

The causal relationship information is expressed by using a combination of information representing a cause element, information representing a result element, and information representing a sign of a causal relationship between these elements. In the present embodiment, the codes related to the causal relationship are determined as follows. When the causal element increases (increases), the resulting element increases, and when the causal element decreases, the causal sign that the resulting element decreases is defined as a positive sign To express. Hereinafter, a causal relationship representing such a relationship may be referred to as a “positive causal relationship”. The sign of the causal relationship that the resulting element decreases when the causal element increases and the resulting element increases when the causal element decreases is expressed as a negative sign. Hereinafter, the causal relationship representing such a relationship may be referred to as “negative causal relationship”. The above-mentioned signs of the causal relationship are considered to be information representing the causal relationship between the causal element and the resulting element. More specifically, the sign of the causal relationship described above is considered to be information representing the relationship between the increase / decrease in the causal element and the increase / decrease in the resulting element. In the causal relationship information, the causal relationship code may be expressed by using numerical values such as “+1” and “−1”, or other symbols. An increase in a certain element represents an increase (increase) in some attribute (for example, characteristics, properties, quantity, etc.) that can be expressed qualitatively or quantitatively with respect to the element. In addition, a decrease in an element indicates, for example, that an attribute related to the element is decreased (decreased).

The chain extraction unit 102 (chain extraction unit) extracts a chain of causal relationships including at least two causal relationships from the information received by the input receiving unit 101. Here, a chain of causal relationships including two causal relationships is configured as follows. That is, assuming three elements, they are described as element 1, element 2, and element 3, respectively. When there is a causal relationship (first causal relationship) that causes element 1 as a cause, and a causal relationship that results as element 2 (first causal relationship) and a cause and effect that causes element 2 as a cause (second causal relationship), The above-described chain of causal relationships is composed of these two causal relationships and three elements.

The causal relation calculation unit 103 (causal relation calculation means) derives (calculates) a causal relation between elements at both ends of the causal relation chain from the two causal relations constituting the extracted causal relation chain. More specifically, the causal relationship calculation unit 103 calculates between the elements arranged at both ends in the causal relationship chain from the two causal relationships (first and second causal relationships) that constitute the extracted causal relationship chain. Calculate the causal relationship. Here, both ends of the chain of causal relationships correspond to, for example, element 1 and element 3 among the three elements described in the above example. Hereinafter, the elements at both ends of the chain of causal relationships extracted by the chain extracting unit 102 may be referred to as “causal relationship element pairs”.

The result output unit 104 (result output unit) outputs the result of the calculation by the causal relationship calculation unit 103. The method of outputting the calculation result by the result output unit 104 may be a method of outputting to a suitable output device (display device) such as a display (not shown) or a method of outputting to a file. The result output unit 104 may transmit data representing the calculation result to another information processing apparatus or the like. The method by which the result output unit 104 outputs the calculation result is not limited to the above, and an appropriate method may be adopted.

The information processing apparatus 100 may acquire the causal relationship information from an input device (not shown) such as a keyboard, may be acquired from a database or the like (not shown), or another information processing apparatus (for example, a server). Etc.). The method by which the information processing apparatus 100 acquires the causal relationship information is not limited to the above, and an appropriate method may be adopted.

In the present embodiment, the information processing apparatus 100 includes a control unit, a storage unit, and an input / output unit as a hardware configuration. The control unit is configured using an arithmetic device such as a CPU (Central Processing Unit), for example. The storage unit is configured using, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), an HDD (Hard Disk Drive), and the like. The input / output unit is configured using various interfaces such as an operation unit, a display unit, and a communication interface. The functions of the information processing apparatus 100 described above are realized by the cooperation of these hardware and various programs stored in the storage unit. Note that a hardware configuration capable of realizing the information processing apparatus 100 will be described later.

[Description of operation]
Next, an example of a flow of processing by the information processing apparatus 100 will be described with reference to FIG. The flowchart illustrated in FIG. 3 is an example of the operation of the information processing apparatus 100, and the present embodiment is not limited to this. The execution order of the processing steps in the flowchart may be changed within a range that does not affect the processing result, and one or more processing steps may be executed in parallel. Also, the control of the process illustrated in FIG. 3 is performed, for example, by the control unit configuring the information processing apparatus 100 developing and executing a program stored in the storage unit.

First, in step S301, the input receiving unit 101 receives input of information (causal relationship information) representing elements constituting an event and a causal relationship between elements.

FIG. 4 shows an example of information representing the elements constituting the event and the causal relationship between the elements. In the specific example shown in FIG. 4, the causal relationship information 400 is expressed using a table format including a column 401 indicating a causal element, a column 402 indicating a sign of the causal relationship, and a column 403 indicating a result element. The In each of the above columns, data expressed in an appropriate format such as a character string, a numerical value, a code, and a symbol is appropriately set. The causal relationship information may be expressed using other methods (forms). For example, as such an expression method, an expression method that uniquely defines the graph structure as shown in FIG. 5 may be used. In the graph structure shown in FIG. 5, elements constituting an event are represented using graph nodes, and the causal relationship between elements is a directed link from the cause element (node) to the result element (node). It is expressed using The causal relationship sign is added to the link by a plus sign (“+”) when positive and by a minus sign (“−”) when negative.

Next, in step S302, the chain extraction unit 102 extracts a chain of causal relationships including two causal relationships from the received causal relationship information. Hereinafter, a method for extracting the causal chain will be described with reference to a specific example shown in FIG.

FIG. 6 shows a list of causal relation chains extracted from the causal relation information exemplified in FIG. The causal relation list 600 illustrated in FIG. 6 includes a column 601 indicating the element 1, a column 602 indicating the sign of the causal relationship between the element 1 and the element 2, a column 603 indicating the element 2, and the causality of the element 2 and the element 3. A column 604 indicating the sign of the relationship and a column 605 indicating the element 3 are included. In each of the above columns, data expressed in an appropriate format such as a character string, a numerical value, a code, and a symbol is appropriately set.

To extract the chain of causality shown in FIG. 6, for example, for the causal relationship information illustrated in FIG. 4, the element that is the result (403) in one row and the element that is the cause (401) in another row Search for a combination of lines that matches. Alternatively, a pair of nodes whose path length is “2” is searched from the graph structure illustrated in FIG. 5, and three nodes and two links constituting the path are extracted. Good. The method of extracting the causal relationship chain is not limited to the method exemplified above, and other appropriate methods may be adopted.

Next, in step S303, the causal relation calculation unit 103 calculates the causal relation between the elements at both ends of the causal relation chain from the two causal relations constituting the extracted causal relation chain. More specifically, the causal relationship calculation unit 103 derives the causal relationship between these elements by calculating the sign of the causal relationship between the elements at both ends of the chain of causal relationships.

The method for calculating the causal relationship code of element 1 and element 3 from the chain of causal relationships shown in FIG. 6 is as follows. That is, the result of multiplying the sign of the causal relationship between element 1 and element 2 and the sign of the causal relation between element 2 and element 3 is the calculation of the causal relationship between element 1 and element 3. Results (signs of causality). In other words, the sign of the causal relationship between element 1 and element 2 and the sign of the result of causal relation between element 2 and element 3 are the causality between element 1 and element 3. Treated as a relationship sign.

For example, the first line (600A in FIG. 6) of the causal relation list 600 in FIG. 6 is an element in which element 1 indicates “security”, element 2 indicates “city appeal”, and element 3 indicates “population”. Yes, the sign of the causal relationship between element 1 and element 2 is positive, and the sign of the causal relation between element 2 and element 3 is positive. Since the result of multiplying positive and positive is positive, the calculation result of the causal relationship between “security” as element 1 and “population” as element 3 is positive. In the third line (600B in FIG. 6), element 1 represents “population”, element 2 represents “crowded”, element 3 represents “security”, and the causal relationship between element 1 and element 2 is positive, element The sign of the causal relationship between 2 and element 3 is negative. Since the result of multiplying positive and negative is negative, the calculation result of the causal relationship between “population” as element 1 and “security” as element 3 is negative. Although not illustrated in FIG. 6, if both the sign of the causal relationship between element 1 and element 2 and the sign of the causal relation between element 2 and element 3 are both negative, the product of negative and negative Since the combination becomes positive, the calculation result of the causal relationship between element 1 and element 3 is positive.

FIG. 7 shows a list of the results of calculating the causal relationship between the elements at both ends of the causal relationship chain illustrated in FIG. In FIG. 7, the calculation result list 700 includes a column 701 indicating element 1, a column 702 indicating element 3, and a column 703 indicating the sign of the calculation result of the causal relationship between element 1 and element 3. In each of the above columns, data expressed in an appropriate format such as a character string, a numerical value, a code, and a symbol is appropriately set.

Finally, in step S304, the result output unit 104 outputs the result of the calculation.

The result output method may be in the form of a table as exemplified in FIG. Further, a table in which a column indicating the element 2 is added to the table of FIG. 7 may be displayed.

As described above, the information processing apparatus 100 according to the present embodiment accepts input of information representing elements constituting events and causal relationships between elements, and extracts a chain of causal relationships including two causal relationships. To do. Then, the information processing apparatus 100 calculates a causal relationship between elements at both ends of the causal relationship chain from the two causal relationships and outputs the calculation result. According to such an information processing apparatus 100, the calculation result of the causal relationship between the elements at both ends of the chain of causal relationships can be provided to the user. The user can confirm whether or not the result of the chain of causality is consistent based on the provided calculation result.

For example, from the first line of the calculation result list 700 in FIG. 7 (700A in FIG. 7), the user has a causal relationship that “the population increases when security is improved” or “the population decreases when security is worse”. The result of chaining can be confirmed. Therefore, the user can determine that the causal relationship is consistent. On the other hand, from the sixth line of the calculation result list 700 in FIG. 7 (700B in FIG. 7), the user indicates that “the criminal identification rate increases when congestion is severe” or “criminal when congestion is alleviated” The result of the chain of causal relations that “the specific rate goes down” can be confirmed. In this case, since the result is usually difficult to assume, the user can find a problem that the chain of causal relationships is not consistent. As a result, the user can correct a causal error, for example. As described above, the information processing apparatus 100 according to the present embodiment can provide the user with information that can evaluate the consistency of the result of linking a plurality of causal relationships.

<Second Embodiment>
Next, a second embodiment of the present invention will be described.

[Constitution]
FIG. 8 is a block diagram illustrating a functional configuration example of an information processing apparatus 800 according to the second embodiment of the present invention. Hereinafter, of the configuration of the information processing apparatus 800, a configuration different from the information processing apparatus 100 will be described. About the structure similar to the said 1st Embodiment, detailed description is abbreviate | omitted by attaching | subjecting the same referential mark. Note that the hardware configuration of the information processing apparatus 800 may be the same as that of the first embodiment.

The information processing apparatus 800 includes an input reception unit 101, a chain extraction unit 102, and a causal relationship calculation unit 103. The information processing apparatus 800 further includes a document storage unit 801, a causal relationship extraction unit 802, a consistency analysis unit 803, and a result output unit 804.

Document storage unit 801 (document storage means) holds (stores) one or more documents. The document stored in the document storage unit 801 is not particularly limited, and may be any document that is expected to describe the elements constituting the event and the causal relationship between the elements. The document may be a document that can be acquired from various information sources such as a document published on the WEB (World Wide Web), a newspaper article, or a white paper. The document is not limited to the above example, and other appropriate documents can be selected. Note that the document storage unit 801 is not limited to documents composed of text data, but includes information that can be documented using appropriate analysis techniques (speech recognition, image analysis, etc.), such as voice, image, and video data. Various data may be held. The document storage unit 801 can store the document using, for example, a known file system or a database.

The causal relationship extraction unit 802 (causal relationship extraction means) extracts the causal relationship regarding the elements (causal relationship element pairs) at both ends of the chain of the causal relationship extracted by the chain extraction unit 102 from the document stored in the document storage unit 801. To do.

Specifically, the causal relationship extraction unit 802 analyzes a document stored in the document storage unit 801 using a well-known natural language analysis method (for example, morphological analysis) and includes a sentence including a causal relationship element pair. To extract. And the causal relationship extraction part 802 extracts the causal relationship between the elements which comprise the causal relationship element pair represented in the extracted sentence. The causal relationship extraction unit 802 may extract the causal relationship using, for example, a well-known natural language analysis (syntax analysis, semantic analysis, etc.) or a data mining (text mining) method. The causal relationship extraction unit 802 may assign the code to the extracted causal relationship and provide the code (specifically, data representing the code) to the consistency analysis unit 803 described later.

The causal relationship extraction unit 802, for example, when voice, image, video data, or the like is stored in the document storage unit 801, uses an appropriate analysis technique such as voice recognition or image analysis to generate a document from the data. May be extracted (character information). Then, the causal relationship extraction unit 802 may extract the causal relationship from the extracted documentable information.

The consistency analysis unit 803 (consistency analysis unit) analyzes the consistency between the causal relationship calculated by the causal relationship calculation unit 103 and the causal relationship extracted by the causal relationship extraction unit 802. Hereinafter, the causal relationship calculated by the causal relationship calculating unit 103 may be described as “calculated causal relationship”, and the causal relationship extracted by the causal relationship extracting unit 802 may be described as “extracted causal relationship”.

The result output unit 804 (result output means) outputs the result of consistency analysis. The result output unit 804 may output the consistency analysis result to an output device (display device) such as a display (not shown) or a file. The method for outputting the consistency analysis result is not limited to the above, and another appropriate method may be selected.

[Operation]
Next, processing by the information processing apparatus 800 will be described. FIG. 9 is a flowchart illustrating an example of processing of the information processing apparatus 800. The flowchart illustrated in FIG. 9 is an example of the operation of the information processing apparatus 800, and the present embodiment is not limited to this. The execution order of the processing steps in the flowchart may be changed within a range that does not affect the processing result, and one or more processing steps may be executed in parallel. Control of the process illustrated in FIG. 9 is performed, for example, by the control unit of the information processing device 800 developing and executing a program stored in the storage unit.

In the flowchart illustrated in FIG. 9, the same reference numerals are assigned to the same steps as those in FIG. 3 in the first embodiment. About these steps (step S301 thru | or S303), since the process similar to 1st Embodiment is performed, detailed description is abbreviate | omitted.

In step S901, the causal relationship extraction unit 802 extracts the causal relationship regarding the elements at both ends of the causal relationship chain extracted by the chain extraction unit 102 from the accumulated document. Here, as a method of extracting the causal relationship from the accumulated document, for example, a known method such as natural language processing or data mining may be used. As such a method, for example, a technique disclosed in the following reference may be used.

[Reference Document] Japanese Patent Application Laid-Open No. 2013-175097 Note that the method for extracting the causal relationship from the accumulated document is not limited to the above example.

Suppose, for example, that the following text is described in the accumulated document.

“Preventing crime will increase the attractiveness of the city.”
From the above sentence, with regard to the last line of the causal chain list 600 illustrated in FIG. 6, as the “crime prevention rate” that is element 1 increases, the “city appeal” that is element 3 also increases. The causal relationship of the positive sign is extracted.

As another example, it is assumed that the following text is described in the accumulated document.

“Congestion makes it difficult to identify criminals.”
From the above sentence, regarding the sixth line (600C in FIG. 6) of the causal relation list 600 illustrated in FIG. 6, element 1 “congestion” becomes severe (increases) and element 3 “ A causal relationship with a negative sign that “criminal identification rate” decreases is extracted. The causal relationship extraction unit 802 may assign a code to the extracted causal relationship and provide the code to the consistency analysis unit 803.

In step S902, the consistency analysis unit 803 extracts the causal relationship calculated by the causal relationship calculation unit 103 in step S303 for the elements at both ends of the extracted causal relationship chain, and the causal relationship extraction unit 802 extracts the document from the document in step S901. Analyze the consistency between the causal relationship.

Specifically, the consistency analysis unit 803 determines that the calculated causal relationship and the causal relationship extracted from the document are consistent if the codes are the same, and determines that they are not consistent if the codes are different. To do.

FIG. 10 is an explanatory diagram showing an example of the consistency analysis result. 10, the consistency analysis result list 1000 includes a column 1001 indicating element 1, a column 1002 indicating element 3, a column 1003 indicating the sign of the calculated causal relation, and a column indicating the sign of the causal relation extracted from the document. 1004 includes a column 1005 indicating the analysis result of consistency. In each of the above columns, data expressed in an appropriate format such as a character string, a numerical value, a code, and a symbol is appropriately set.

For example, the first line (1000A in FIG. 10) of the consistency analysis result list 1000 illustrated in FIG. 10 is an element in which element 1 indicates “security” and element 3 indicates “population”. Is positive, and the causal relationship extracted from the document is positive. Since the signs of the calculated causal relationship and the causal relationship extracted from the document are the same, the consistency analysis unit 803 determines that these causal relationships are consistent. Therefore, data indicating “match” is set in the analysis result.

The sixth line (1000B in FIG. 10) of the consistency analysis result list 1000 is an element in which element 1 indicates “congestion” and element 3 indicates “criminal identification rate”. Is positive, and the sign of the causal relationship extracted from the document is negative. Since the calculated causal relationship and the causal relationship extracted from the document have different signs, the consistency analysis unit 803 determines that these causal relationships are not consistent. Therefore, data indicating “inconsistency” is set in the analysis result.

In step S903, the result output unit 804 outputs the result of the consistency analysis.

The result output method by the result output unit 804 may be, for example, a table format as exemplified in FIG. 10 or may be in another format. In addition, the result output unit 804 combines the element 1 and the element 3 whose consistency analysis result is “inconsistent” (that is, the calculated causal relationship and the causal relationship extracted from the document are not consistent). May be output only.

As described above, the information processing apparatus 800 according to the present embodiment extracts a causal relationship between elements (causal relationship element pairs) at both ends of a chain of causal relationships from the accumulated document, Analyzing consistency with causal relationships derived from calculations. Then, the information processing apparatus 800 outputs the analysis result.

According to the information processing apparatus 800 in the present embodiment, it is possible to provide a user with a result of analyzing whether or not the results of chained causal relationships are consistent. Since the information processing apparatus 800 extracts the causal relationship between the causal relationship element pairs from the documents acquired and stored from various information sources, the causal relationship extracted from the document is considered to represent a general causal relationship. . That is, the user can compare and consider the general causal relationships described in these documents and the results of the chain of causal relationships calculated based on information input by the user. As a result, the user can correct a causal error. As described above, the information processing apparatus 800 according to the present embodiment can provide the user with information used for evaluation of consistency related to a result of chaining a plurality of causal relationships.

<Modification of Second Embodiment>
Next, a modification of the second embodiment of the present invention described above will be described. The configuration of the information processing apparatus in the present embodiment may be the same as that in the second embodiment.

In this modification, the chain extraction unit 102 extracts a chain of causal relationships including three or more causal relationships from the information received by the input receiving unit 101.

Here, a chain of causal relationships composed of n (n is an integer of 3 or more) causal relationships is expressed as follows, for example. That is, as illustrated in FIG. 11, there are (n + 1) elements from element 1 to element (n + 1), the causal relationship that results from element (m + 1) due to element (m), and element (m + 1) ) To cause the element (m + 2) as a result. In the above, m is an integer satisfying (1 ≦ m <n).

The method for extracting n causal relationships from the causal relationship information may be the same as in the second embodiment.

The causal relationship calculation unit 103 calculates a causal relationship between elements at both ends in a chain of causal relationships composed of n causal relationships extracted by the chain extraction unit 102. For example, the causal relationship calculation unit 103 calculates the sign of the causal relationship between the element (m) and the element (m + 1) and the sign of the causal relationship between the element (m + 1) and the element (m + 2) (m = 1 to m = (n−1)). Thereby, the causal relationship calculation part 103 can calculate the causal relationship between the both ends (element 1 and element (n + 1)) of a causal relationship.

The causal relationship extraction unit 802 is a causal factor relating to elements (causal relationship element pairs) at both ends of a chain of causal relationships composed of n number of causal relationships extracted by the chain extraction unit 102 from the documents stored in the document storage unit 801. Extract relationships. The consistency analysis unit 803 analyzes the consistency between the causal relationship calculated by the causal relationship calculation unit 103 and the causal relationship extracted by the causal relationship extraction unit 802. These processes may be the same as those in the second embodiment.

For example, in the document stored in the document storage unit 801, there may be a case where the causal relationship about the elements at both ends of the chain composed of two causal relationships is not sufficiently described. In such a case, it is conceivable to determine the consistency of the chain consisting of the two causal relationships and the chain consisting of three or more causal relationships having the same elements at both ends using the document. That is, for example, if it is possible to determine the consistency of a chain composed of three or more causal relationships using the document, the determination result can be used to determine the consistency of the chain composed of the two causal relationships. Is possible.

Alternatively, it is conceivable to use the information processing apparatus in the present modification to determine the consistency of a chain composed of three or more causal relationships including the chain composed of the two causal relationships. For example, it is assumed that a determination result is obtained that the causal relationship is consistent with respect to a chain composed of three or more causal relationships using the document. In this case, it can be considered that the causal relationship of the above two causal relationships included in the chain of three or more causal relationships is also consistent. Further, a case is assumed in which a determination result that the causal relationship is inconsistent is obtained for a chain including three or more causal relationships. In this case, it is considered that there is a possibility that the causal relationship is not consistent with respect to any two causal relationships included in the chain including the three or more causal relationships. In this case, the information processing apparatus according to this modification can issue a warning or the like to the user, for example. As described above, the information processing apparatus according to the present modification can provide the user with information that can evaluate the consistency of the result of chaining three or more causal relationships.

<Third Embodiment>
Next, a third embodiment that is a basic embodiment of the present invention will be described. FIG. 12 is a block diagram illustrating a functional configuration of an information processing apparatus according to the third embodiment of this invention.

As illustrated in FIG. 12, the information processing apparatus 1200 includes a chain extraction unit 1201, a causal relationship calculation unit 1202, and a result output unit 1203. These components constituting the information processing apparatus 1200 are communicably connected using an appropriate communication method.

The chain extraction unit 1201 (chain extraction unit) extracts a chain of causal relationships including at least two causal relationships from the causal relationship information that is information representing the elements constituting the event and the causal relationships between the elements. The causal relation information and the chain of causal relations including at least two causal relations may be the same as those in the above embodiments. The chain extraction unit 1201 may be provided with the causal relationship information from a user (not shown) of the information processing apparatus 1200 or another information processing apparatus, and is held in a storage device (not shown) in the information processing apparatus 1200. The causal relationship information may be acquired.

The chain extraction unit 1201 may be configured, for example, in the same manner as the chain extraction unit 102 in each of the above embodiments, and at least two causal information is obtained from the causal relationship information by the same processing as the chain extraction unit 102 in each of the above embodiments. A chain of causal relationships consisting of relationships may be extracted.

The causal relationship calculation unit 1202 (causal relationship calculation means) calculates the causal relationship between the elements at both ends of the extracted chain of causal relationships from at least two causal relationships constituting the chain of causal relationships extracted by the chain extraction unit 1201. calculate. Specifically, the causal relationship calculation unit 1202 may calculate the relationship between the increase / decrease of the cause element and the increase / decrease of the element as a result among the elements at both ends of the extracted chain of causal relationships. . The causal relationship calculation unit 1202 may be configured, for example, in the same manner as the causal relationship calculation unit 103 in each of the above-described embodiments. You may calculate the causal relationship between the elements arrange | positioned to.

The result output unit 1203 (result output means) outputs the result of the calculation by the causal relationship calculation unit 1202. The result output unit 1203 may output the calculation result to a display device or the like (not shown), for example, similarly to the result output unit 104 in the first embodiment, and outputs the calculation result in the form of a file or the like. May be. The method by which the result output unit 1203 outputs the calculation result is not limited to the above, and an appropriate method may be selected.

As described above, the information processing apparatus 1200 according to the present embodiment extracts a chain of causal relationships including at least two causal relationships from the causal relationship information indicating the elements constituting the events and the causal relationships between the elements. To do. Then, the information processing apparatus 1200 calculates a causal relationship between elements at both ends of the causal relationship chain from the two causal relationships, and outputs the calculation result. According to such an information processing apparatus 1200, the calculation result of the causal relationship between the elements at both ends of the chain of causal relationships can be provided to the user. The user can confirm whether or not the result of the chain of causality is consistent based on the provided calculation result. As described above, the information processing apparatus 1200 according to the present embodiment can provide the user with information that can evaluate the consistency of the result of linking a plurality of causal relationships.

<Configuration of hardware and software program (computer program)>
Hereinafter, a hardware configuration capable of realizing each of the above-described embodiments will be described.

In the following description, the information processing apparatuses (100, 800, 1200) described in the above embodiments are collectively referred to simply as “information processing apparatus”. Each component of the information processing apparatus may be simply referred to as “component of the information processing apparatus”.

The information processing apparatus described in each of the above embodiments may be configured by one or a plurality of dedicated hardware devices. In that case, each component shown in the above figures (FIGS. 2, 8, and 12) uses hardware (an integrated circuit or a storage device on which processing logic is mounted) that is partially or entirely integrated. It may be realized.

When the information processing apparatus is realized by dedicated hardware, the components of the information processing apparatus may be realized by, for example, a circuit configuration capable of providing each function. Such a circuit configuration includes, for example, an integrated circuit such as SoC (System on a Chip), a chip set realized using the integrated circuit, and the like. In this case, the data held by the components of the information processing apparatus is, for example, a RAM (Random Access Memory) area integrated as SoC, a flash memory area, or a storage device (such as a semiconductor storage device) connected to the SoC. May be stored. The data includes, for example, causal relationship information received by the input receiving unit 101, a chain of causal relationships extracted by the chain extracting unit (102, 1201), a calculation result by the causal relationship calculating unit (103, 1202), and the like. May be. The data may include the causal relationship extracted by the causal relationship extraction unit 802, the analysis result by the consistency analysis unit 803, the document stored in the document storage unit 801, and the like. In addition, the data may include processing data generated by the components of the information processing apparatus during the processing.

In this case, a well-known communication network (for example, a communication bus) may be employed as a communication line that connects each component of the information processing apparatus. Further, the communication line connecting each component may be connected between each component by peer-to-peer.

Further, the information processing apparatus described above may be configured by general-purpose hardware exemplified in FIG. 13 and various software programs (computer programs) executed by the hardware. In this case, the information processing apparatus may be configured by an arbitrary number of general-purpose hardware devices and software programs. That is, an individual hardware device may be assigned to each component configuring the information processing apparatus, and a plurality of components may be realized using a single hardware device.

The arithmetic device 1301 in FIG. 13 is an arithmetic processing device such as a general-purpose CPU (Central Processing Unit) or a microprocessor. For example, the arithmetic device 1301 may read various software programs stored in a non-volatile storage device 1303, which will be described later, into the storage device 1302, and execute processing according to the software programs. In this case, the function of the component of the information processing apparatus in each of the above embodiments is realized using a software program executed by the arithmetic device 1301.

The storage device 1302 is a memory device such as a RAM or a ROM that can be referred to from the arithmetic device 1301, and stores software programs and various data. Note that the storage device 1302 may be a volatile memory device or a nonvolatile memory device.

The storage device 1302 may temporarily store data held by the components of the information processing device. The data includes, for example, causal relationship information received by the input receiving unit 101, a chain of causal relationships extracted by the chain extracting unit (102, 1201), a calculation result by the causal relationship calculating unit (103, 1202), and the like. May be. The data may include the causal relationship extracted by the causal relationship extraction unit 802, the analysis result by the consistency analysis unit 803, the document stored in the document storage unit 801, and the like. In addition, the data may include processing data generated by the components of the information processing apparatus during the processing.

The nonvolatile storage device 1303 is a nonvolatile storage device such as a magnetic disk drive or a semiconductor storage device using a flash memory. The nonvolatile storage device 1303 can store various software programs, data, and the like. For example, various documents stored in the document storage unit 801 may be stored in the nonvolatile storage device 1303.

The network interface 1306 is an interface device connected to a communication network, and for example, a wired and wireless LAN connection interface device may be employed. For example, the information processing apparatus may acquire a document from various information sources via the network interface 1306 and various communication networks. Further, the information processing apparatus may accept causal relationship information via the network interface 1306.

The drive device 1304 is, for example, a device that processes reading and writing of data with respect to a recording medium 1305 described later.

The recording medium 1305 is an arbitrary recording medium capable of recording data, such as an optical disk, a magneto-optical disk, and a semiconductor flash memory.

The input / output interface 1307 is a device that controls input / output with an external device. The input receiving unit 101 may receive input of causal relationship information from an input device (such as a keyboard) connected via the input / output interface 1307, for example. In addition, the result output unit (104, 804, 1203) may output the calculation result of the causal relationship or the determination result of the consistency to the display device connected via the input / output interface 1307.

The information processing apparatus according to the present invention described with the above-described embodiments as an example, or a component thereof, for example, software that can realize the functions described in the above-described embodiments with respect to the hardware apparatus illustrated in FIG. -It may be realized by supplying a program. More specifically, for example, the present invention may be realized by causing the arithmetic device 1301 to execute a software program supplied to the hardware device. In this case, an operating system running on the hardware device, database management software, network software, middleware such as a virtual environment platform, etc. may execute part of each process.

In each of the above-described embodiments, each unit illustrated in each drawing can be realized as a software module, which is a function (processing) unit of a software program executed by the above-described hardware. However, the division of each software module shown in these drawings is a configuration for convenience of explanation, and various configurations can be assumed for implementation.

For example, when each component of the information processing apparatus illustrated in FIGS. 2, 8, and 12 is realized as a software module, these software modules are stored in the nonvolatile storage device 1303. Then, when the arithmetic device 1301 executes each process, these software modules are read out to the storage device 1302.

Further, these software modules may be configured to transmit various data to each other by an appropriate method such as shared memory or inter-process communication. With such a configuration, these software modules are connected so as to communicate with each other.

Furthermore, the software program may be recorded on the recording medium 1305. In this case, the software program may be stored in the non-volatile storage device 1303 through the drive device 1304 as appropriate at the shipping stage or operation stage of the components of the information processing apparatus.

In the above case, the method of supplying various software programs to the hardware is installed in the apparatus using an appropriate jig in the manufacturing stage before shipment or the maintenance stage after shipment. A method may be adopted. As a method for supplying various software programs, a general procedure may be adopted at present, such as a method of downloading from the outside via a communication line such as the Internet.

In such a case, the present invention can be understood to be constituted by a code constituting the software program or a computer-readable recording medium on which the code is recorded. In this case, the recording medium is not limited to a medium independent of the hardware device, but includes a recording medium in which a software program transmitted via a LAN or the Internet is downloaded and stored or temporarily stored.

Further, the components of the information processing apparatus described above are configured by a virtualized environment in which the hardware device illustrated in FIG. 13 is virtualized and various software programs (computer programs) executed in the virtualized environment. May be. In this case, the components of the hardware device illustrated in FIG. 13 are provided as virtual devices in the virtual environment. In this case as well, the present invention can be realized with the same configuration as the case where the hardware device illustrated in FIG. 13 is configured as a physical device.

The present invention has been described above as an example applied to the exemplary embodiment described above. However, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to such embodiments. In such a case, new embodiments to which such changes or improvements are added can also be included in the technical scope of the present invention. Furthermore, the embodiments described above, or embodiments obtained by combining new embodiments with such changes or improvements may be included in the technical scope of the present invention. This is clear from the matters described in the claims.

This application claims priority based on Japanese Patent Application No. 2015-243100 filed on Dec. 14, 2015, the entire disclosure of which is incorporated herein.

DESCRIPTION OF SYMBOLS 100 Information processing apparatus 101 Input reception part 102 Chain extraction part 103 Causal relation calculation part 104 Result output part 800 Information processing apparatus 801 Document storage part 802 Causal relation extraction part 803 Consistency analysis part 804 Result output part 1200 Information processing apparatus 1201 Chain extraction Unit 1202 causal relation calculation unit 1203 result output unit 1301 arithmetic unit 1302 storage unit 1303 nonvolatile storage unit 1304 drive unit 1305 recording medium 1306 network interface 1307 input / output interface

Claims

A chain extracting means for extracting a chain of causal relations composed of at least two causal relations from causal relation information including information representing elements constituting the event and information representing causal relations between the elements;
A causal relationship calculating means for calculating a causal relationship between elements at both ends of the extracted causal relationship from at least two causal relationships constituting the extracted causal relationship; and
An information processing apparatus comprising: result output means for outputting a result of calculation by the causal relation calculation means.
Document storage means capable of holding one or more documents;
Causal relation extraction means for extracting causal relations about elements at both ends of a chain of causal relations extracted by the chain extraction means from the document held in the document storage means;
Consistency analyzing means for analyzing consistency between the causal relation calculated by the causal relation calculating means and the causal relation extracted by the causal relation extracting means, further comprising:
The result output means outputs the consistency analysis result.
The information processing apparatus according to claim 1.
The information indicating the causal relationship between the elements included in the causal relationship information is the relationship between the increase / decrease in the element causing the event and the increase / decrease in the element resulting in the event. Represent,
The causal relationship calculation means, based on the information representing the two extracted causal relationships, out of the elements at both ends of the extracted chain of causal relationships, the increase and decrease of the cause elements, and the resulting element The information processing apparatus according to claim 1, wherein a causal relationship between increase and decrease is calculated.
Information representing the causal relationship between the elements included in the causal relationship information is as follows:
When the factor causing the event increases, the factor resulting from the event increases, and when the factor causing the event decreases, the event results. The causal relationship that the element decreases is represented by a positive sign,
When the factor causing the event increases, the factor resulting from the event decreases, and when the factor causing the event increases, the event results It is information that represents the causal relationship that the element decreases by a negative sign,
The causal relationship calculating means calculates a product of codes representing the extracted two causal relationships, thereby increasing or decreasing the causal elements among the elements at both ends of the extracted causal relationship chain, and the result. The information processing apparatus according to claim 3, wherein a sign representing a causal relationship between increase and decrease of the element is calculated.
Information representing the causal relationship between the elements included in the causal relationship information is as follows:
When the factor causing the event increases, the factor resulting from the event increases, and when the factor causing the event decreases, the event results. The causal relationship that the element decreases is represented by a positive sign,
When the factor causing the event increases, the factor resulting from the event decreases, and when the factor causing the event increases, the event results It is information that represents the causal relationship that the element decreases by a negative sign,
The causal relationship calculating means calculates a product of codes representing the extracted two causal relationships, thereby increasing or decreasing the causal elements among the elements at both ends of the extracted causal relationship chain, and the result. Calculating a sign representing a causal relationship between the increase and decrease of the element
The causal relationship extracting unit extracts the causal relationship between the elements at both ends of the chain of causal relationships extracted by the chain extracting unit by analyzing the document, and sets the code representing the causal relationship to the matching Provided to the sex analysis means,
The consistency analysis means is based on a result of comparing the code representing the causal relation extracted by the causal relation extracting means with the code representing the causal relation calculated by the causal relation calculating means. The information processing apparatus according to claim 2, wherein the information processing apparatus determines whether or not the causal relationship calculated by the calculating unit matches the causal relationship extracted by the causal relationship extracting unit.
The chain extraction means, from the causal relationship information, the first element is the element that causes the event, and the second element is the element that results as the event. The causal relationship and the second causal relationship in which the second element is the element that causes the event and the third element is the element that results in the event are extracted. The information according to any one of claims 1 to 5, wherein a chain of causal relationships including the two causal relationships in which the first causal relationship and the second causal relationship are chained is extracted. Processing equipment.
Extracting a chain of causal relationships consisting of at least two causal relationships from causal relationship information including information representing the elements constituting the event and information representing the causal relationships between the elements,
From at least two causal relationships constituting the extracted causal relationship chain, calculate the causal relationship between the elements at both ends of the extracted causal relationship chain,
An information processing method for outputting a calculation result of the causal relationship.
The information processing method according to claim 7, further comprising:
For elements at both ends of the chain of causal relationships, extract the causal relationship between them from one or more documents;
Analyze the consistency between the causal relationship extracted from the document and the calculated causal relationship;
An information processing method for outputting the consistency analysis result.
A process of extracting a chain of causal relationships composed of at least two causal relationships from causal relationship information including information representing elements constituting the events and information representing causal relationships between the elements;
A process of calculating a causal relationship between elements at both ends of the extracted causal relationship chain from at least two causal relationships constituting the extracted causal relationship chain;
A recording medium on which a computer program for causing a computer to execute a process for outputting a calculation result of the causal relationship is recorded.
The computer program according to claim 9, further comprising:
Processing to hold the document;
A process of extracting causal relationships between the elements at both ends of the chain of causal relationships from one or more of the documents;
A process for analyzing consistency between the causal relationship extracted from the document and the calculated causal relationship;
A recording medium on which a computer program for causing a computer to execute processing for outputting the analysis result of the consistency is recorded.