WO2019053878A1

WO2019053878A1 - Knowledge generation device, knowledge generation method, and computer-readable recording medium

Info

Publication number: WO2019053878A1
Application number: PCT/JP2017/033450
Authority: WO
Inventors: 大地木村; 正明土田
Original assignee: 日本電気株式会社
Priority date: 2017-09-15
Filing date: 2017-09-15
Publication date: 2019-03-21
Also published as: JP6828823B2; JPWO2019053878A1

Abstract

A knowledge generation device 100 is provided with: a knowledge determination unit 10 for acquiring input data and knowledge data prepared in advance for generating knowledge to be used for an inference, and determining, using a statistical prediction model 30, whether knowledge generated using the acquired input data and knowledge data is appropriate or not; and a knowledge generation unit 20 for generating knowledge to be used for the inference, using the input data and knowledge data that the knowledge determination unit 10 has determined to be appropriate. The prediction model is a model for outputting the level of appropriateness of the knowledge acquired from the input data and the knowledge data in accordance with a combination of both.

Description

Knowledge generation apparatus, knowledge generation method, and computer readable recording medium

The present invention relates to a knowledge generation apparatus and a knowledge generation method for generating knowledge for use in inference, and further relates to a computer readable recording medium recording a program for realizing these.

Hypothetical reasoning is a reasoning method for deriving hypotheses that explains the observed facts (input data) based on existing knowledge, and in recent years, it has become possible to use a computer to dramatically improve processing speed. It has become. According to the hypothesis inference, for example, with respect to an abnormal state occurring in an object, a manufacturing system, a running software program, etc., related facts obtained among hypotheses based on several possibilities leading to the state are obtained. You can get the best explained hypothesis.

By the way, when it is going to prepare all the knowledge used by hypothesis inference by a static database, it is necessary to store a large amount of knowledge in the database. This is because, in order to derive a hypothesis, it is necessary to interpret the input data depending on the context, and strictly speaking, the same input data is rarely input.

Therefore, there is a limit in preparing a large amount of knowledge in advance, so an implication determination is performed on existing knowledge and input data, and an attempt is made to generate new knowledge by using the result of the implication determination. ing. For example, it is assumed that the input data is "unify the fare" and the knowledge is "adjust the price → carte". In this case, if it is determined by the implication determination that the former implies the latter, new knowledge “unify the fare → adjust the price” is generated.

Further, a technique for performing such implication determination is disclosed, for example, in Patent Document 1. In the technology disclosed in Patent Document 1, first, a common partial structure is extracted for the first text and the second text to be determined. Subsequently, any one of feature quantities based on dependencies between common partial structures and feature quantities based on dependencies between common partial structures and other structures is extracted. Thereafter, the implication relationship between the first text and the second text is determined based on the extracted feature quantity. In the technique disclosed in Patent Document 1, the implication relationship between a plurality of sentences is determined by using information representing the structure of the sentences.

International Publication No. 2015/004155

As described above, using the technology disclosed in Patent Document 1, it is possible to automatically create a large number of pieces of knowledge using data determined to be implication. However, there is a problem that the knowledge generated in this way is not all correct. And, if hypothesis inference is performed using incorrect knowledge, wrong inference results will be obtained.

An example of the object of the present invention is to provide a knowledge generation device, a knowledge generation method, and a computer readable recording medium which can solve the above problems and suppress the generation of false knowledge.

In order to achieve the above object, a knowledge generating device in one aspect of the present invention is:
Statistically predicts whether input data and previously prepared knowledge data are acquired for generation of knowledge used in inference, and the acquired input data and knowledge generated using the knowledge data are appropriate. A knowledge determination unit that determines using a model;
A knowledge generation unit that generates knowledge used in the inference using the input data determined by the knowledge determination unit as appropriate and the knowledge data;
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
It is characterized by

Further, in order to achieve the above object, a knowledge generation method in one aspect of the present invention is:
(A) Acquisition of input data and knowledge data prepared in advance for generation of knowledge used in inference, and statistics as to whether the acquired input data and knowledge generated using the knowledge data are appropriate Determining using a typical prediction model,
(B) generating knowledge to be used in the inference using the input data and the knowledge data determined to be appropriate in the step (a);
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
It is characterized by

Furthermore, in order to achieve the above object, a computer readable recording medium in one aspect of the present invention is:
On the computer
(A) Acquisition of input data and knowledge data prepared in advance for generation of knowledge used in inference, and statistics as to whether the acquired input data and knowledge generated using the knowledge data are appropriate Determining using a typical prediction model,
(B) generating knowledge to be used in the inference using the input data and the knowledge data determined to be appropriate in the step (a);
Record the program, including instructions to execute
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
It is characterized by

As described above, according to the present invention, generation of false knowledge can be suppressed in generation of knowledge used for hypothesis inference.

FIG. 1 is a block diagram showing a schematic configuration of a knowledge generation apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram specifically showing the configuration of the knowledge generation apparatus according to the embodiment of the present invention. FIGS. 3 (a) and 3 (b) are explanatory diagrams showing learning processing performed by the knowledge generation apparatus in the embodiment of the present invention, and FIG. 3 (a) shows learning processing in the knowledge generation unit. 3 (b) shows a learning process in the model generation unit. FIG. 4 is a flowchart showing an operation at the time of creation of a prediction model of the knowledge generation device in the embodiment of the present invention. FIG. 5 is a flow chart showing an operation at the time of knowledge generation of the knowledge generation device in the embodiment of the present invention. FIG. 6 is a block diagram showing an example of a computer for realizing the knowledge generation device in the embodiment of the present invention.

Embodiment
Hereinafter, a knowledge generation apparatus, a knowledge generation method, and a computer readable recording medium according to an embodiment of the present invention will be described with reference to FIGS. 1 to 6.

[Device configuration]
First, a schematic configuration of a knowledge generation apparatus according to the present embodiment will be described with reference to FIG. FIG. 1 is a block diagram showing a schematic configuration of a knowledge generation apparatus according to an embodiment of the present invention.

The knowledge generation device 100 according to the present embodiment shown in FIG. 1 is a device for generating knowledge used in inference. As shown in FIG. 1, the knowledge generation device 100 includes a knowledge determination unit 10 and a knowledge generation unit 20.

First, the knowledge determination unit 10 acquires input data and knowledge data 40 prepared in advance for generation of knowledge used in inference. Also, the knowledge data 40 is composed of a set of inferences that can be used for inference. Next, the knowledge determination unit 10 determines, using the statistical prediction model 30, whether the knowledge generated using the acquired input data and the knowledge data 40 is appropriate.

The prediction model 30 is a model that outputs an appropriate degree of knowledge obtained from both in accordance with a combination of input data and knowledge data. Also, the knowledge generation unit 20 generates the knowledge to be used in the inference by using the input data and the knowledge data 40 determined by the knowledge determination unit 10 to be appropriate.

As described above, in the present embodiment, it is determined in advance whether the knowledge to be created is appropriate or not, and the knowledge is generated only when it is determined that the knowledge is appropriate, and hypothesis inference is performed. For this reason, according to the present embodiment, it is possible to suppress the generation of the wrong knowledge in the generation of the knowledge used for the hypothesis inference.

Subsequently, the configuration of the knowledge generation apparatus 100 according to the present embodiment will be more specifically described with reference to FIG. FIG. 2 is a block diagram specifically showing the configuration of the knowledge generation apparatus according to the embodiment of the present invention.

As shown in FIG. 2, in the present embodiment, in addition to the knowledge determination unit 10 and the knowledge generation unit 20 described above, the knowledge generation apparatus 100 generates a model generation unit 50 that generates a prediction model 30, and input data And a data extraction unit 60 for extracting data used to generate knowledge from the above. As input data, for example, text data making up an electronic mail, a blog, writing on an electronic bulletin board, news and the like can be mentioned.

When a combination of input data for learning and knowledge data for learning is used, the model generation unit 50 applies a combination of the knowledge generated by the knowledge generation unit 10, the input data for learning, and the knowledge for learning. And the knowledge of the correct answer given is obtained as learning data. Then, the model generation unit 50 performs a learning process using the acquired learning data to generate the prediction model 30.

In the present embodiment, the knowledge determination unit 10 applies the input data and the knowledge data 40 to the prediction model 30, and extracts the data extracted by the data extraction unit 60 (hereinafter referred to as “extracted data”) and the knowledge data 40. It is determined whether the knowledge generated by the knowledge generation unit 20 is appropriate using

In the present embodiment, the knowledge generation unit 20 generates a learning model (not shown in FIG. 2) for knowledge generation in advance, and generates extraction data and knowledge data that the knowledge determination unit 10 has determined to be appropriate. Apply it to the learned model to generate the knowledge used in inference.

Here, the learning process by the model generation unit 50 and the learning process by the knowledge generation unit 20 will be specifically described using FIG. 3. FIGS. 3 (a) and 3 (b) are explanatory diagrams showing learning processing performed by the knowledge generation apparatus in the embodiment of the present invention, and FIG. 3 (a) shows learning processing in the knowledge generation unit. 3 (b) shows a learning process in the model generation unit.

Further, in the present embodiment, the knowledge generation unit 20 generates knowledge by implication determination using a learning model. Therefore, the data extraction unit 60 performs predicate term structure analysis on the sentences included in the input data so as to enable implication determination, and extracts, for example, a subject, a predicate, and an object word in the input data.

As shown in FIG. 3A, in the learning process of the knowledge generation unit 20, first, the data extraction unit 60 extracts text data (extraction data) of a subject, a predicate, and an object from input data to be learning data. The data is extracted and input to the knowledge generation unit 20.

In addition, when the knowledge generation unit 20 acquires the extraction data, for each of the knowledge constituting the knowledge data 40, the knowledge generation unit 20 determines whether the knowledge implies the extraction data and determines that the knowledge is implied. Generates new knowledge from the extracted data and the corresponding knowledge. Then, the knowledge generation unit 20 compares the generated knowledge with the correct knowledge (teacher data) prepared in advance, learns the comparison result (correct or incorrect answer), and creates a learning model.

Further, as shown in FIG. 3B, in the learning process of the model generation unit 50, the model generation unit 50, unlike the knowledge generation unit 20, acquires the entire input data to be learning data. That is, the model generation unit 50 acquires not only a subject, a predicate, an object in input data, but also a modifier.

At this time, the same input data is also input to the data extraction unit 60, and the data extraction unit 60 extracts a subject, a predicate, and an object from the input data, and passes them to the knowledge generation unit 20 as extraction data. . Then, the knowledge generation unit 20 applies the extraction data and the knowledge data 40 to a learning model to generate knowledge. After that, the generated knowledge is given a label indicating whether or not it is the correct solution based on the correct solution knowledge prepared in advance by an external system or the like.

Subsequently, the model generation unit 50 predicts, using the input data and the knowledge data 40, whether or not the label of the knowledge generated by the knowledge generation unit 20 from the same input data is correct. Then, the model generation unit 50 compares the prediction result with the label of the generated knowledge, and learns the comparison result (the prediction is correct or incorrect) to create the prediction model 30. Thereafter, the created prediction model 30 is used for determination by the knowledge determination unit 10.

[Device operation]
Next, the operation of the knowledge generation apparatus according to the embodiment of the present invention will be described using FIGS. 4 and 5. FIG. In the following description, FIGS. 1 to 3 will be referred to as appropriate. Further, in the present embodiment, the knowledge generation method is implemented by operating the knowledge generation apparatus 100. Therefore, the explanation of the knowledge generation method in the present embodiment is replaced with the following explanation of the operation of the knowledge generation apparatus 100.

First, the process of creating a prediction model, that is, the learning process by the model creating unit 50 will be described using FIG. 4. FIG. 4 is a flowchart showing an operation at the time of creation of a prediction model of the knowledge generation device in the embodiment of the present invention. As a premise, it is assumed that learning processing is executed in the knowledge generation unit 20 (see FIG. 3A), and a learning model has already been created.

As shown in FIG. 4, first, the model generation unit 50 acquires the entire input data to be learning data (step A1). Next, the model generation unit 50 predicts, using the input data and the knowledge data 40, whether the label of the knowledge generated by the knowledge generation unit 20 from the same input data is a correct answer (step A2).

The same input data is also input to the data extraction unit 60 in synchronization with steps A1 and A2, and the data extraction unit 60 extracts a subject, a predicate, and an object from the input data, and uses these as extraction data. Pass it to the knowledge generation unit 20. Then, the knowledge generation unit 20 applies the extraction data and the knowledge data 40 to a learning model to generate knowledge. Thereafter, a label indicating whether or not the correct answer is given is added to the generated knowledge based on the correct answer knowledge prepared in advance by an external system or the like (see FIG. 3B).

After execution of step A2, the model generation unit 50 compares and contrasts the prediction result (correct or incorrect answer) of step A2 with the label (correct or incorrect answer) given to the knowledge generated by the knowledge generation unit 20. The result is learned to create a prediction model 30 (step A3). The prediction model 30 created in step A3 is used for the determination by the knowledge determination unit 10.

Subsequently, the process of creating knowledge will be described with reference to FIG. FIG. 5 is a flow chart showing an operation at the time of knowledge generation of the knowledge generation device in the embodiment of the present invention.

As shown in FIG. 5, first, the data extraction unit 60 acquires input data, and inputs the acquired input data to the knowledge determination unit 10 (step B1).

Subsequently, the data extraction unit 60 extracts data used to generate knowledge from the acquired input data, and inputs the extracted data (extracted data) to the knowledge generation unit 20 (step B2). Specifically, for example, when the input data is an e-mail, the data extraction unit 60 executes a predicate term structure analysis for each sentence constituting the e-mail, and forms a subject of each sentence, The predicate and the object are extracted, and these are input to the knowledge generation unit 20.

Next, the knowledge determination unit 10 applies the input data and knowledge data 40 acquired in step B1 to the prediction model 30, and uses the extraction data and knowledge data 40 extracted in step B2 to generate the knowledge generation unit 20. It is determined whether the knowledge generated at step S is proper (step B3).

As a result of the determination in step B3, when it is determined not to be appropriate, step B5 is executed. On the other hand, if it is determined as a result of the determination in step B3 that the knowledge generation unit 20 determines that the extraction data extracted in step B2 and the knowledge data 40 are learning models (FIGS. 3A and 3B) ) (See B.4) to generate knowledge to be used in inference. The generated knowledge is used, for example, for inference performed in another system.

Next, in step B5, the knowledge determination unit 10 determines whether or not step B3 is performed on all the texts included in the input data acquired in step B1. As a result of the determination in step B5, when step B3 is not executed for all the text, step B3 is executed again. On the other hand, as a result of the determination in step B5, when step B3 is executed for all the texts, the processing in the knowledge generation device 100 ends.

Here, steps B1 to B5 shown in FIG. 5 will be described using a specific example. First, it is assumed that the input data acquired in step B1 is an e-mail, and that the sentence “We will unify the fare as the assumption” is included.

In this case, the data extraction unit 60 performs a predicate term structure analysis on "assuming that we have unified the fare as an assumption," and obtains the following result.
Predicate term structure: We unify fares (to cases) into (to cases) (predicates)
(In addition, there is no "case" in this sentence.)

In addition, assuming that “adjust price → cartel generation” is included in the knowledge data 40 as knowledge, the knowledge generation unit 20 normally calculates “unify the fare” from the knowledge and the extraction data. It is judged that it is implication to "adjust the price", and the following knowledge is newly generated.
Knowledge: Unify fares → adjust prices

However, in the present embodiment, the knowledge determination unit 10 uses the entire input data, that is, a portion that is not extracted as extraction data to determine whether the generated knowledge is appropriate. In the above-described example, the data extraction unit 60 applies not only the extraction data but also the word “assumption” to the prediction model.

At this time, it is assumed that the prediction model outputs a low value (for example, 10% or less) as the appropriate degree when the word “assumption” is included. The knowledge determination unit 10 determines that the knowledge generated by the knowledge generation unit 20 from the above-described input data is not appropriate.

As described above, in the present embodiment, it is determined in advance whether knowledge predicted to be created from input data is appropriate. Then, the knowledge generation unit generates knowledge from input data only when it is determined that the data is appropriate.

Therefore, according to the present embodiment, it is possible to suppress the generation of the wrong knowledge in the generation of the knowledge used for the hypothesis inference, and to improve the accuracy of the hypothesis inference. Further, since the knowledge determination unit 10 makes the determination using the entire input data, the context is reflected in the knowledge generation by the knowledge generation unit 20, and the validity of the knowledge is enhanced. In addition, even when the accuracy in the knowledge generation unit 20 is not very high, that is, even when the learning model is simple, it is possible to generate appropriate knowledge.

[program]
The program in the present embodiment may be a program that causes a computer to execute steps A1 to A3 shown in FIG. By installing this program in a computer and executing it, the knowledge generation device and the knowledge generation method in the present embodiment can be realized. In this case, the processor of the computer functions as the knowledge determination unit 10, the knowledge generation unit 20, the model generation unit 50, and the data extraction unit 60 to perform processing.

Further, in the present embodiment, the prediction model 30 and the knowledge data 40 are realized by storing data files that configure these in a storage device such as a hard disk provided in a computer.

Also, the program in the present embodiment may be executed by a computer system constructed by a plurality of computers. In this case, for example, each computer may function as any of the knowledge determination unit 10, the knowledge generation unit 20, the model generation unit 50, and the data extraction unit 60. Further, the prediction model 30 and the knowledge data 40 may be stored in a storage device of a computer different from the computer that executes the program in the present embodiment.

Here, a computer for realizing the knowledge generation device 100 by executing the program according to the present embodiment will be described with reference to FIG. FIG. 6 is a block diagram showing an example of a computer for realizing the knowledge generation device in the embodiment of the present invention.

As shown in FIG. 6, the computer 110 includes a CPU 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. These units are communicably connected to each other via a bus 121.

The CPU 111 develops the program (code) in the present embodiment stored in the storage device 113 in the main memory 112 and executes various operations by executing these in a predetermined order. The main memory 112 is typically a volatile storage device such as a dynamic random access memory (DRAM). In addition, the program in the present embodiment is provided in the state of being stored in computer readable recording medium 120. The program in the present embodiment may be distributed on the Internet connected via communication interface 117.

Further, as a specific example of the storage device 113, besides a hard disk drive, a semiconductor storage device such as a flash memory may be mentioned. The input interface 114 mediates data transmission between the CPU 111 and an input device 118 such as a keyboard and a mouse. The display controller 115 is connected to the display device 119 and controls the display on the display device 119.

The data reader / writer 116 mediates data transmission between the CPU 111 and the recording medium 120, and executes reading of a program from the recording medium 120 and writing of the processing result in the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and another computer.

Further, specific examples of the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as flexible disk (Flexible Disk), or CD- An optical recording medium such as a ROM (Compact Disk Read Only Memory) may be mentioned.

The knowledge generation device 100 in the present embodiment can also be realized by using hardware corresponding to each unit, not a computer on which a program is installed. Furthermore, the knowledge generation device 100 may be partially realized by a program, and the remaining portion may be realized by hardware.

A part or all of the embodiment described above can be expressed by (Appendix 1) to (Appendix 9) described below, but is not limited to the following description.

(Supplementary Note 1)
Statistically predicts whether input data and previously prepared knowledge data are acquired for generation of knowledge used in inference, and the acquired input data and knowledge generated using the knowledge data are appropriate. A knowledge determination unit that determines using a model;
A knowledge generation unit that generates knowledge used in the inference using the input data determined by the knowledge determination unit as appropriate and the knowledge data;
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
A knowledge generation apparatus characterized in that.

(Supplementary Note 2)
The apparatus further comprises a model generation unit that generates the prediction model.
The model generation unit is a combination of knowledge generated by the knowledge generation unit, input data for learning, and knowledge for learning when using a combination of input data for learning and knowledge data for learning. The learning process is performed using the knowledge of the correct answer given to the symbol as learning data to generate the prediction model.
The knowledge generation device according to appendix 1.

(Supplementary Note 3)
It further comprises a data extraction unit for extracting data used for generating knowledge from the input data,
The knowledge determination unit determines, using the prediction model, whether the data extracted by the data extraction unit and the knowledge generated using the knowledge data are appropriate.
The knowledge generation unit generates knowledge used in the inference by using the data extracted by the data extraction unit and the knowledge data, which the knowledge determination unit has determined to be appropriate.
The knowledge generation device according to appendix 1 or 2.

(Supplementary Note 4)
(A) Acquisition of input data and knowledge data prepared in advance for generation of knowledge used in inference, and statistics as to whether the acquired input data and knowledge generated using the knowledge data are appropriate Determining using a typical prediction model,
(B) generating knowledge to be used in the inference using the input data and the knowledge data determined to be appropriate in the step (a);
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
A knowledge generation method characterized by

(Supplementary Note 5)
(C) generating the prediction model;
When a combination of input data for learning and knowledge data for learning is used in the step (c), the knowledge generated by the execution of the step (b), the input data for the learning and the data Using the knowledge of the correct answer given to the combination of knowledge for learning as learning data, the learning process is performed to generate the prediction model.
The knowledge generation method according to appendix 4.

(Supplementary Note 6)
(D) extracting the data used to generate knowledge from the input data;
In the step (a), it is determined using the prediction model whether the data extracted in the step (d) and the knowledge generated using the knowledge data are appropriate;
In the step (b), using the data extracted in the step (d) and the knowledge data determined to be appropriate in the step (a), the knowledge used in the inference is generated.
The knowledge generation method according to Appendix 4 or 5.

(Appendix 7)
On the computer
(A) Acquisition of input data and knowledge data prepared in advance for generation of knowledge used in inference, and statistics as to whether the acquired input data and knowledge generated using the knowledge data are appropriate Determining using a typical prediction model,
(B) generating knowledge to be used in the inference using the input data and the knowledge data determined to be appropriate in the step (a);
Record the program, including instructions to execute
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
A computer readable recording medium characterized in that.

(Supplementary Note 8)
The program is stored in the computer
(C) generating an instruction to further execute the step of generating the prediction model;
When a combination of input data for learning and knowledge data for learning is used in the step (c), the knowledge generated by the execution of the step (b), the input data for the learning and the data Using the knowledge of the correct answer given to the combination of knowledge for learning as learning data, the learning process is performed to generate the prediction model.
The computer-readable recording medium according to appendix 7.

(Appendix 9)
The program is stored in the computer
(D) extracting from the input data data to be used for generating knowledge, further including instructions for further executing steps;
In the step (a), it is determined using the prediction model whether the data extracted in the step (d) and the knowledge generated using the knowledge data are appropriate;
In the step (b), using the data extracted in the step (d) and the knowledge data determined to be appropriate in the step (a), the knowledge used in the inference is generated.
The computer-readable recording medium according to appendix 7 or 8.

As described above, according to the present invention, generation of false knowledge can be suppressed in generation of knowledge used for hypothesis inference. The present invention is useful for the efficiency of work to clarify the situation or cause of an accident, a crime, a cyber attack or the like. Similarly, the present invention is also useful for cause analysis and countermeasure investigation of disasters and system failures.

DESCRIPTION OF REFERENCE NUMERALS 10 knowledge determination unit 20 knowledge generation unit 30 prediction model 40 knowledge data 50 model generation unit 60 data extraction unit 100 knowledge generation device 110 computer 111 CPU
112 main memory 113 storage device 114 input interface 115 display controller 116 data reader / writer 117 communication interface 118 input device 119 display device 120 recording medium 121 bus

Claims

Statistically predicts whether input data and previously prepared knowledge data are acquired for generation of knowledge used in inference, and the acquired input data and knowledge generated using the knowledge data are appropriate. A knowledge determination unit that determines using a model;
A knowledge generation unit that generates knowledge used in the inference using the input data determined by the knowledge determination unit as appropriate and the knowledge data;
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
A knowledge generation apparatus characterized in that.
The apparatus further comprises a model generation unit that generates the prediction model.
The model generation unit is a combination of knowledge generated by the knowledge generation unit, input data for learning, and knowledge for learning when using a combination of input data for learning and knowledge data for learning. The learning process is performed using the knowledge of the correct answer given to the symbol as learning data to generate the prediction model.
The knowledge generation device according to claim 1.
It further comprises a data extraction unit for extracting data used for generating knowledge from the input data,
The knowledge determination unit determines, using the prediction model, whether the data extracted by the data extraction unit and the knowledge generated using the knowledge data are appropriate.
The knowledge generation unit generates knowledge used in the inference by using the data extracted by the data extraction unit and the knowledge data, which the knowledge determination unit has determined to be appropriate.
The knowledge generation device according to claim 1.
(A) Acquisition of input data and knowledge data prepared in advance for generation of knowledge used in inference, and statistics as to whether the acquired input data and knowledge generated using the knowledge data are appropriate Determining using a typical prediction model,
(B) generating knowledge to be used in the inference using the input data and the knowledge data determined to be appropriate in the step (a);
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
A knowledge generation method characterized by
(C) generating the prediction model;
When a combination of input data for learning and knowledge data for learning is used in the step (c), the knowledge generated by the execution of the step (b), the input data for the learning and the data Using the knowledge of the correct answer given to the combination of knowledge for learning as learning data, the learning process is performed to generate the prediction model.
The knowledge generation method according to claim 4.
(D) extracting the data used to generate knowledge from the input data;
In the step (a), it is determined using the prediction model whether the data extracted in the step (d) and the knowledge generated using the knowledge data are appropriate;
In the step (b), using the data extracted in the step (d) and the knowledge data determined to be appropriate in the step (a), the knowledge used in the inference is generated.
The knowledge generation method according to claim 4 or 5.
On the computer
(A) Acquisition of input data and knowledge data prepared in advance for generation of knowledge used in inference, and statistics as to whether the acquired input data and knowledge generated using the knowledge data are appropriate Determining using a typical prediction model,
(B) generating knowledge to be used in the inference using the input data and the knowledge data determined to be appropriate in the step (a);
Record the program, including instructions to execute
The prediction model is a model that outputs an appropriate degree of knowledge obtained from both according to a combination of input data and knowledge data.
A computer readable recording medium characterized in that.
The program is stored in the computer
(C) generating an instruction to further execute the step of generating the prediction model;
When a combination of input data for learning and knowledge data for learning is used in the step (c), the knowledge generated by the execution of the step (b), the input data for the learning and the data Using the knowledge of the correct answer given to the combination of knowledge for learning as learning data, the learning process is performed to generate the prediction model.
The computer readable recording medium according to claim 7.
The program is stored in the computer
(D) extracting from the input data data to be used for generating knowledge, further including instructions for further executing steps;
In the step (a), it is determined using the prediction model whether the data extracted in the step (d) and the knowledge generated using the knowledge data are appropriate;
In the step (b), using the data extracted in the step (d) and the knowledge data determined to be appropriate in the step (a), the knowledge used in the inference is generated.
A computer readable recording medium according to claim 7 or 8.