Knowledge model creation support device

Abstract

To provide a knowledge model creation support device capable of easily creating a knowledge model.SOLUTION: A knowledge model creation support device 1 is provided with a term-relation database creation unit 2 for creating a term-relation database 3 by using a term-relation extraction model to extract a plurality of terms and a relation of the plurality of terms from document data. The term-relation extraction model includes a document data acquisition unit 11 for acquiring the document data, a term extraction unit 13 for extracting terms from the document data, a trigger word storage unit 15 for storing a trigger word for leading to a relation between two terms, and a relation extraction unit 17 for extracting the relation between the two terms extracted by the term extraction unit 13 on the basis of the trigger word when the trigger word is included in the document data.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a knowledge model creation support device. [Background technology]

[0002] Patent Document 1 describes a system for constructing a knowledge model that describes knowledge by experts in various fields such as engineering, medicine, pharmacy, agriculture, and biology. The knowledge model represents the terms in a given field and their relationships by using the terms as factors that are keywords in the given field and interconnecting the factors in a network form. Patent Document 2 describes a knowledge model in the field of machining. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2018-147351 A [Patent Document 2] JP 2020-049606 A Summary of the Invention [Problem to be solved by the invention]

[0004] When creating a knowledge model, an operator needs to extract terms that define factors and then understand the relationships between those terms. However, since the extraction of terms and the understanding of the relationships between the extracted terms require manual setup, particularly by an expert, creating a knowledge model is not easy.

[0005] An object of the present invention is to provide a knowledge model creation support device that can easily create a knowledge model. [Means for solving the problem]

[0006] One aspect of the present invention isAn apparatus for supporting the creation of a knowledge model composed of a plurality of factors defined by terms and relationship information between the factors, comprising: a term-relationship database including a plurality of terms and relationships of the plurality of terms; a term-relationship database creation unit that creates the term-relationship database by extracting the terms and the relationships from document data using a term-relationship extraction model; a knowledge model creation unit that creates the knowledge model based on the term-relationship database, The term-relationship extraction model is a document data acquisition unit for acquiring the document data; a term extraction unit that extracts terms from the document data; a trigger word storage unit for storing a trigger word that leads to a relationship between two terms; a pair extraction unit that creates term pair information from the terms extracted by the term extraction unit; When the document data includes the trigger word, the relationship between the two terms extracted by the term extraction unit is calculated based on the trigger word. and the pair information A relationship extraction unit that extracts a relationship based on 、 the document data includes a nested structure including a plurality of terms; The nested structure is a structure that is composed of a plurality of inclusive terms and a nested outer term that combines the plurality of inclusive terms, the term extraction unit extracts, when the document data includes the nested structure, the inclusive term and the nested outer term in the nested structure from the document data; when the document data includes the nested structure, the pair extraction unit creates the pair information by using only the nested outer term in the nested structure without using the inclusive term in the nested structure; The relationship extraction unit is in a knowledge model creation support device that, when the document data includes the nested structure, treats the nested outer term in the nested structure as a relationship extraction target and does not treat the inclusive term as a relationship extraction target.

[0007] Another aspect of the present invention is an apparatus for supporting creation of a knowledge model that is configured by a plurality of factors defined by terms and relationship information between the factors, the apparatus comprising: a term-relationship database including a plurality of terms and relationships of the plurality of terms; a term-relationship database creation unit that creates the term-relationship database by extracting the terms and the relationships from document data using a term-relationship extraction model; a knowledge model creation unit that creates the knowledge model based on the term-relationship database, The term-relationship extraction model is a document data acquisition unit for acquiring the document data; a feature representation generating unit that generates a feature representation in which a feature of the document data is expressed as a vector based on each token constituting the document data acquired by the document data acquiring unit; a term extraction unit that extracts terms from the document data; a trigger word storage unit for storing a trigger word that leads to a relationship between two terms; a relationship extraction unit that extracts a relationship between two terms extracted by the term extraction unit based on the trigger word when the trigger word is included in the document data, The term extraction unit and the relationship extraction unit are in a knowledge model creation support device that extracts the terms and the relationships by sharing the characteristic expressions generated by the characteristic expression generation unit.

[0008] According to the above knowledge model creation support device, a term-relationship database is created using a term-relationship extraction model that extracts terms and relationships between terms from document data. First, the term-relationship extraction model automatically extracts terms from document data. Furthermore, the term-relationship extraction model pre-stores a trigger word that induces a relationship between two terms, and when the trigger word is included in the document data, extracts the relationship between the two terms based on the trigger word. In this way, the term-relationship extraction model can automatically extract terms from document data, and can automatically extract relationships between terms included in the document data by taking into account the preset trigger word. Therefore, the term-relationship database can be created without manual work. Then, if the term-relationship database can be created, a knowledge model can be created by referring to the term-relationship database, and therefore the knowledge model can be easily created. [Brief description of the drawings]

[0009] [Figure 1] A knowledge network diagram is shown. [Diagram 2] 1 is a functional block diagram showing an overall configuration of a knowledge model creation support device; [Diagram 3] FIG. 2 is a diagram showing a first database (term-relationship database) constituting the knowledge model creation support device. [Figure 4] 4 is a diagram illustrating relationship labels relating to relationships in the first database shown in FIG. 3. FIG. [Diagram 5] 2 is a functional block diagram showing a term-relationship extraction model in a first database creation unit constituting the knowledge model creation support device. FIG. [Figure 6] FIG. 13 is a diagram for explaining a nested structure applied in the term extraction unit. [Figure 7] FIG. 4 is a diagram showing trigger words stored in a trigger word storage unit. [Figure 8] FIG. 13 is a diagram showing a detailed configuration of a term-relationship extraction model. [Figure 9] FIG. 1 is a functional block diagram of a term-relationship extraction model in the first learning phase. [Figure 10] FIG. 13 is a functional block diagram of a term-relationship extraction model in the second learning phase. [Figure 11] 13 is a diagram showing a drawing GUI window showing an application example of a knowledge model creation unit constituting the knowledge model creation support device. FIG. [Figure 12] FIG. 13 is a diagram showing a drawing GUI window showing another application example of the knowledge model creation unit. [Figure 13] FIG. 13 is a diagram showing a drawing GUI window showing another application example of the knowledge model creation unit. [Figure 14] FIG. 13 is a diagram showing a drawing GUI window showing another application example of the knowledge model creation unit. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] (1. Overview of the knowledge model) A knowledge model describes and stores knowledge related to information in a given field in a specified format. That is, a knowledge model is mainly composed of multiple factors defined by the terminology of the field and information on the relationships between the factors. For example, the knowledge model may be knowledge related to a technical field.

[0011] The knowledge model is represented by, for example, data having information on factors and data having relationship information. Conceptually, the knowledge model expresses the relationships between multiple factors by connecting them to each other in the form of a network. In other words, the knowledge model stores knowledge (including know-how) in various technical fields as explicit knowledge and can be updated.

[0012] The technical fields may include engineering, medicine, pharmacy, agriculture, biology, and the like. In particular, the technical field may include the mechanical processing field, which is included in the engineering field. Here, the mechanical processing field includes, for example, cutting and grinding. In addition, it is particularly useful for the knowledge model to describe knowledge of technical information held by experts in each technical field.

[0013] For example, in the field of machining, an operator determines cutting speed, depth of cut, etc. as machining conditions, taking into consideration various information such as the material of the workpiece, the material of the tool, machining accuracy, machining cycle time, etc. In this case, the knowledge model is a model of the operator's thought process when the operator determines cutting speed, depth of cut, etc. as machining conditions, using various information such as the material of the workpiece, the material of the tool, machining accuracy, machining cycle time, etc. as input information.

[0014] In other words, in addition to the workpiece material, tool material, machining requirements (machining accuracy, machining cycle time, etc.), cutting speed, cutting depth, etc., the knowledge model defines each of the technical elements that appear in the thought process as factors, and defines the relationships between the factors.

[0015] For example, the knowledge model is used as follows: When an operator inputs necessary information about the input factors, such as the material of the workpiece, the material of the tool, and the machining requirements (such as machining accuracy and machining cycle time), into the knowledge model, information about the cutting speed, the cutting depth, and the like, as output factors, is output.

[0016] (2. Example of knowledge network diagram 100) As described above, the knowledge model is conceptually expressed in the form of a network. An example of a knowledge network diagram 100 that expresses a knowledge model in the form of a network diagram will be described with reference to Fig. 1. In this example, a knowledge network diagram 100 related to a knowledge model in the machining field is taken as an example.

[0017] As shown in FIG. 1, the knowledge network diagram 100 includes a plurality of node graphics 110 and link graphics 120 that connect the node graphics 110 to each other. The node graphics 110 are represented by any graphic such as a box, a graphic including text, an icon, or the like. The node graphics 110 represent factors in the knowledge model. The link graphics 120 are represented by straight lines, curved lines, angled lines, or the like. In this example, the link graphics 120 are represented by arrow lines to define the directionality of the relationships. The link graphics 120 represent relationships that connect factors in the knowledge model. In the knowledge network diagram 100 shown in FIG. 1, all the node graphics 110 are represented by boxes in which text can be written, and the link graphics 120 are represented by arrow lines.

[0018] Here, factors are defined by technical terms. In addition, multiple factors may have a technical inclusive relationship (also called a hierarchical relationship, parent-child relationship, or master-slave relationship) or may have a technical heterogeneous relationship. In other words, the relationships between factors are classified into the above two types.

[0019] For example, the workpiece specifications have an inclusive relationship with the thermal properties of the workpiece material, the hardness of the workpiece material, the elongation of the workpiece material, etc. In other words, as factors having a technical inclusive relationship, the workpiece specifications are considered to be higher-level conceptual factors, and the thermal properties of the workpiece material, the hardness of the workpiece material, the elongation of the workpiece material, etc. are considered to be lower-level conceptual factors. For example, an example of a factor having a technically heterogeneous relationship is the thermal properties of the workpiece material and the required heat resistance of a tool. In the following, the relationship between two factors having a technical inclusive relationship will be simply referred to as an inclusive relationship, and the relationship between two factors having a technically heterogeneous relationship will be simply referred to as a heterogeneous relationship.

[0020] As for the link graphic 120, a first link graphic 121 representing an inclusive relationship and a second link graphic 122 representing a heterogeneous relationship are displayed separately. That is, the first link graphic 121 and the second link graphic 122 are displayed in different display methods.

[0021] 1, the first link graphic 121 representing the inclusion relationship is represented by a frame showing the area of ​​the superordinate concept factor, and the subordinate concept factor is arranged within the frame showing the first link graphic 121. Note that the first link graphic 121 may be represented as an L-shape connecting the node graphics 110 arranged close to each other vertically and slightly shifted to the left and right, other than as a frame. In this case, of the two node graphics 110 connected by the first link graphic 121, the node graphic 110 located at the top corresponds to the superordinate concept factor.

[0022] 1, the second link graphic 122 representing a heterogeneous relationship is represented by a straight line, a broken line, or the like, connecting the node graphics 110 arranged at any positions (up, down, left, right) apart. The second link graphic 122 is represented by an arrow line indicating the direction of the definition of the factors.

[0023] (3. Configuration of knowledge model creation support device 1) The knowledge model creation support device 1 is a support device for creating the above-mentioned knowledge model. The configuration of the knowledge model creation support device 1 will be described with reference to FIG.

[0024] The knowledge model creation support device 1 includes a first DB creation unit 2, a first DB 3, a second DB creation unit 4, a second DB 5, a third DB creation unit 6, a third DB 7, and a knowledge model creation unit 8. DB is an abbreviation for database. The knowledge model creation support device 1 includes three DB creation units 2, 4, 6 and three DBs 3, 5, 7, but may include one each, two each, or four or more each.

[0025] The first DB creation unit 2 creates a first DB 3, which is a term-relationship DB, by extracting multiple terms and relationships between the terms from document data (text data) using a term-relationship extraction model described below. That is, when document data (text data) is input to the first DB creation unit 2, the term-relationship extraction model is executed to automatically extract multiple terms and relationships between the terms. The first DB 3 stores the terms and relationships created by the first DB creation unit 2. The terms here are information that can be used to create factors in the above-mentioned knowledge model, and the relationships are information that can be used to create relationship information between factors in the above-mentioned knowledge model.

[0026] The second DB creation unit 4 is configured by a model, for example, word2vec, different from the term-relationship extraction model configuring the first DB creation unit 2. The second DB creation unit 4 creates a second DB 5 by extracting relationships between multiple terms from document data using word2vec. The second DB 5 stores the relationships created by the second DB creation unit 4.

[0027] The third DB creation unit 6 is configured by a model, for example, a known knowledge graph, different from the models configuring the first DB creation unit 2 and the second DB creation unit 4. The third DB creation unit 6 creates a third DB 7 by extracting relationships between multiple terms from document data using the knowledge graph. The third DB 7 stores the relationships created by the third DB creation unit 6.

[0028] The knowledge model creation unit 8 creates a knowledge model using the terms and the relationships between the terms stored in the first DB 3, and further using the relationships between the terms stored in each of the second DB 5 and the third DB 7 as auxiliary terms. For example, the knowledge model creation unit 8 sets a plurality of terms stored in the first DB 3 as factor candidates. Then, the knowledge model creation unit 8 uses each DB 3, 5, 7 to provide candidates for locations where factors to be registered are placed, provide factor candidates having a relationship with a factor of interest, and verify already created knowledge models. The knowledge model creation unit 8 can perform some operations automatically and other operations manually. However, it is also possible for the knowledge model creation unit 8 to fully automate the creation of knowledge models.

[0029] (4. First DB3 example) An example of the first DB (term-relationship DB) will be described with reference to Fig. 3 and Fig. 4. As shown in Fig. 3, the first DB3 stores a reference term A, a related term B having a relationship with the reference term A, and a relationship label indicating the relationship between the reference term A and the related term B. For example, W1 as the reference term A has relationships with W2, W4, and W10, and the respective relationships (relation labels) are Positive, Negative, and Positive.

[0030] As shown in FIG. 4, four types of relationship labels are defined: Positive, Negative, Sub, and Relation. Positive represents a relationship in which the larger the reference term A, the larger the related term B. Negative represents a relationship in which the larger the reference term A, the smaller the related term B. Sub represents a relationship in which reference term A is a type of related term B. Relation represents that reference term A has some qualitative relationship with related term B. Note that relationship labels are not limited to the above four types, and may include other types and can be freely set.

[0031] (5. Configuration of First DB Creation Unit 2) The configuration of the first DB creation unit 2 will be described with reference to Fig. 5 to Fig. 7. The first DB creation unit 2 is configured by a term-relationship extraction model. As shown in Fig. 5, the first DB creation unit 2 includes a document data acquisition unit 11, a feature expression generation unit 12, a term extraction unit 13, a term output unit 14, a trigger word storage unit 15, a pair extraction unit 16, a relationship extraction unit 17, and a relationship output unit 18.

[0032] The document data acquisition unit 11 acquires document data as text data. The characteristic representation generation unit 12 generates a characteristic representation that expresses the characteristics of the document data as a vector, based on each token that constitutes the document data acquired by the document data acquisition unit 11. The term extraction unit 13 extracts terms from the document data. In detail, the term extraction unit 13 uses the characteristic representation generated by the characteristic representation generation unit 12 to extract terms included in the document data. The term output unit 14 outputs the terms extracted by the term extraction unit 13.

[0033] Here, the term extraction unit 13 extracts terms taking into consideration a nested structure 20. The nested structure will be described with reference to Fig. 6. The nested structure 20 is a structure in which a term contains another term. The nested structure 20 is composed of a plurality of contained terms 22, 23 and a nested outer term 21 which combines the contained terms 22, 23.

[0034] For example, the machining term "cutting" includes the machining terms "cutting" and "processing." In this case, "cutting" is the nested outer term 21, and "cutting" and "processing" are the inclusive terms 22, 23. The nested outer term 21 and the inclusive terms 22, 23 that make up the nested structure have, for example, a superior-subordinate relationship, an attribute relationship, a subject-predicate relationship, etc.

[0035] As an example of a sentence from the acquired document data, "When the cutting speed increases, the cutting temperature increases." will be described. In this example sentence, the term extraction unit 13 extracts the terms "cutting speed," "cutting," "speed," "increase," "cutting temperature," "temperature," and "increase." That is, "cutting speed" and "cutting temperature" are extracted as nested outer terms 21, and "cutting," "speed," and "temperature" are extracted as contained terms 22 and 23.

[0036] The trigger word storage unit 15 stores preset trigger words. A trigger word is a keyword that leads to a relationship between two terms. A trigger word is a keyword that represents a change in a term that represents a physical quantity, etc. For example, as shown in FIG. 7, the trigger word storage unit 15 stores "increase," "decrease," "increase," "decrease," "rise," "fall," "contain," etc.

[0037] As an example of a sentence from the acquired document data, "When the cutting speed increases, the cutting temperature increases." will be described. In this example sentence, the trigger words are "increase" and "increase." The trigger words are keywords that can derive a relationship (positive relationship label) between two terms, "cutting speed" and "cutting temperature," in which if one increases, the other increases (see Figure 4).

[0038] The pair extraction unit 16 creates pairs from the terms extracted by the term extraction unit 13, and formats the data to be used by the relationship extraction unit 17 described below. When constructing a nested structure 20, the term extraction unit 13 extracts a nested outer term 21 and inclusion terms 22, 23, but the pair extraction unit 16 uses only the largest nested outer term 21 and does not use the inclusion terms 22, 23. When the number of terms to be paired is n, the pair extraction unit 16 n C2 pairs are created.

[0039] When the document data includes a trigger word, the relationship extraction unit 17 extracts the relationship between the two terms extracted by the term extraction unit 13 based on the trigger word. In extracting the relationship, the relationship extraction unit 17 uses the trigger word stored in the trigger word storage unit 15 and the term pair information created by the pair extraction unit 16. Therefore, when the document data includes a nested structure 20 (shown in FIG. 6 ), the relationship extraction unit 17 will extract the relationship from a nested outer term 21 in the nested structure 20, but will not extract the relationship from inclusive terms 22, 23.

[0040] Furthermore, the relationship extraction unit 17 extracts the relationship between two terms by using the characteristic expression generated by the characteristic expression generation unit 12. That is, the above-mentioned term extraction unit 13 and the relationship extraction unit 17 share the characteristic expression generated by the characteristic expression generation unit 12 to extract terms and relationships. The relationship output unit 18 outputs the relationship between the terms extracted by the relationship extraction unit 17.

[0041] (6. Detailed structure of term-relationship extraction model) A detailed configuration of the term-relationship extraction model constituting the first DB creation unit 2 will be described with reference to Fig. 8. Fig. 8 shows the feature representation generation unit 12, the term extraction unit 13, the pair extraction unit 16, and the relationship extraction unit 17. An explanation will be given using an example sentence, "When the cutting speed increases, the cutting temperature increases,...".

[0042] (6-1. Feature Representation Generation Unit 12) The feature representation generating unit 12 constitutes a shared unit for the term extracting unit 13 and the relationship extracting unit 17. The feature representation generating unit 12 divides the acquired document data into tokens and acquires a token string x. For example, each of "cut", "shave", and "speed" is one token. Next, a token representation H1 is acquired from the token string x. The token representation H1 is expressed by a string of vectors corresponding to the token string x. To acquire the token representation H1, a pre-training model for the token string x is used. For example, the token representation H1 is acquired as shown in formula (1) using BERT (Bidirectional Encoder Representations from Transformers), which is one of the pre-training models.

[0043]

number

[0044] Next, the acquired token representation H1 is used as an input to a convolutional neural network (CNN) to acquire an intermediate representation H2 as shown in formula (2). In this example, the intermediate representation H2 corresponds to the feature representation generated by the feature representation generation unit 12.

[0045]

number

[0046] The intermediate representation H2 is generated using the set parameters. Here, the intermediate representation H2 is a representation used for both term extraction and relationship extraction. Therefore, as will be described later, in the learning phase, learning is performed based on the term extraction loss generated in the term extraction by the term extraction unit 13 and the relationship extraction loss generated in the relationship extraction by the relationship extraction unit 17. In other words, the parameters for generating the intermediate representation H2 are updated using the losses generated in each of the term extraction and relationship extraction.

[0047] (6-2.Term extraction section 13) The term extraction unit 13 focuses on the fact that the difference between the nested outer terms 21 and the inclusive terms 22, 23 that make up the nested structure 20 is the number of tokens that make up the term, and is a model that outputs two values ​​for each number of tokens (token length).

[0048] First, the intermediate representation H2 is used as input, and a convolution operation is performed to obtain a representation of a token sequence of length i, as shown in equation (3). In equation (3), spanCNNi represents a convolution operation using a filter with kernel size i, and n entity corresponds to the number of tokens that make up a term, and is a hyperparameter given before learning. The intermediate representation Vi obtained is n entity It is an intermediate representation of individuals.

[0049]

number

[0050] Next, as shown in equation (4), n entity A common fully connected layer is applied to each of the intermediate representations Vi to generate an intermediate representation Li.

[0051]

number

[0052] The output of all connections is n entity The output is a sequence of token length corresponding to each kernel size. As shown in FIG. 8, the contents of the output are a 1 at the start position of the tokens that make up the kernel size term. For example, the term "Cutting speed" is made up of three tokens, and the start position of "Cutting speed" is the position of the "Cut" token, so the token length is "3" and a 1 (black circle) is set at the token position of "Cut". Similarly, the term "Speed" is made up of one token, and the start position is the "Speed" token, so the token length is "1" and a 1 (black circle) is set at the token position of "Speed".

[0053] (6-3. Pair Extraction Unit 16) As described above, the pair extraction unit 16 creates pairs of terms extracted by the term extraction unit 13. However, when the extracted terms form a nested structure 20, the pair extraction unit 16 only targets the largest nested external term 21 for relationship extraction. For example, as shown in FIG. 8, "cutting speed", "cutting", and "speed" are extracted by the term extraction unit 13, and in this case, only "cutting speed" is targeted for relationship extraction. Therefore, in FIG. 8, the token number length is "3" and a 1 is set at the position of the token "cut", and the positions corresponding to "cutting" and "speed", which are the inclusive terms 22 and 23, are set to 0. The same is true for "cutting temperature". In other words, here, a pair of "cutting speed" and "cutting temperature" is created. When n terms are extracted from one sentence of document data, two terms are selected to create a pair, so n C2 pairs are created.

[0054] Next, a relational label (see FIG. 4) is assigned to each of the created pairs by referring to the annotation file, and a triplet of "relational label, term A, term B" is created. Furthermore, the pair extraction unit 16 also creates a term position vector PEi for term A and term B. The term position vector is a vector of the number of tokens with a length of one sentence, in which 1 is placed at the position where a term exists and 0 is placed at the position where a term does not exist. In FIG. 8, for one term of the pair, "cutting speed," 1 is placed at the positions of "cut," "cut," and "speed," and 0 is placed at the remaining positions. For the other term of the pair, "cutting temperature," 1 is placed at the positions of "cut," "cut," and "temperature," and 0 is placed at the remaining positions. In this manner, the term position vector PEi is created by the pair extraction unit 16.

[0055] (6-4. Relationship Extraction Unit 17) The relationship extraction unit 17 is a model that takes trigger words into consideration as described above. First, trigger word information is added to the intermediate representation H2 generated by the feature representation generation unit 12 as a sharing unit, and is used as the input of the Multi Head Attention layer. First, Vtrig is generated by referring to the trigger word storage unit 15 (shown in FIG. 5). Vtrig is a vector whose length is the token length (SeqLen) in one sentence and whose dimension is 1. For the corresponding token, 1 is defined if the token is a trigger word, and 0 is defined if the token is not a trigger word. Then, as shown in formula (5), an embedding process is performed on Vtrig using a matrix initialized with weights based on a normal distribution with a mean of 0 and a variance of 1.

[0056]

number

[0057] Next, as shown in equation (6), the intermediate representation H2 is concat- ed against Htrig in the dimension direction, and input to the fully connected layer to generate Q (query).

[0058]

number

[0059] Moreover, K (key) is generated according to equation (7), and V (value) is generated according to equation (8).

[0060]

number

[0061]

number

[0062] Next, the generated Q, K, and V are input to the multi-head attention layer to generate intermediate representations H3 and W. In this attention layer, Q (query) is the search source (target), K (key) is the search destination (source), and V (value) is the score.

[0063]

number

[0064] Here, Multi Head Attention adds trigger word information to Q (query). Since information is added to the query, the trigger word is added as "something about which we want to know the relevance in the sentence." Also, considering that K (key) and V (value) contain information from the intermediate representation H2, the obtained intermediate representation H3 can be interpreted as "an expression that pays strong attention to tokens related to the trigger word among the tokens in the entire sentence."

[0065] Next, as shown in equations (10) and (11), the intermediate representation H3 is passed through two skip connections and two convolutional layers to generate intermediate representation H4.

[0066]

number

[0067]

number

[0068] Next, as shown in formula (12), the term position vector PEi and intermediate representation H4 created by the pair extraction unit 16 are used to extract expressions E1 and E2 of two terms to be used for relationship extraction. Here, PEi is the i-th term position vector, and * is the element product.

[0069]

number

[0070] Next, as shown in formula (13), the obtained E1 and E2 are subjected to a Maxpooling operation to convert them into one vector each, where E1' is a term composed of multiple tokens.

[0071]

number

[0072] Next, as shown in equation (14), E1' is input to a fully connected layer to generate a terminological representation E1".

[0073]

number

[0074] Next, as shown in equation (15), the vectors of E1" and E2" are summed, and the probability of each relationship existing between terms is output through the fully connected layer as shown in equation (16). In other words, Lrel is a vector whose length is the number of types of relationships, the probability of the i-th type of relationship is represented by the value of the i-th element of Lrel, and the relationship with the highest probability in the Lrel vector represents the relationship between the two terms.

[0075]

number

[0076]

number

[0077] (7. Learning Phase) (7-1. Overview) The above-mentioned term-relationship extraction model needs to acquire a large amount of document data and perform machine learning. In particular, in this example, in the term-relationship extraction model, parameters in the feature representation generation unit 12 shared by the term extraction unit 13 and the relationship extraction unit 17 are learned.

[0078] (7-2. First learning phase) The first learning phase as the first learning will be described with reference to Fig. 9. In the first learning phase, the term extraction unit 13 uses the term extraction loss Lenency generated in the extraction of terms to update the generation parameters of the intermediate representation H2 in the feature representation generation unit 12, thereby performing the first learning. That is, in the first learning phase, learning is performed by a single-task learning process that updates the generation parameters based on the term extraction loss Lenency. In Fig. 9, functions that are not used in the first learning phase are indicated by dashed lines.

[0079] Here, for example, cross entropy loss is used as the term extraction loss Lenty. In this case, the term extraction loss Lenty is expressed by formula (17). n represents the number of model outputs, i.e., the number of filters prepared, and l represents the number of tokens in one sentence. The difference between the predicted value of l for each token and the correct answer y is defined as cross entropy. The term extraction loss Lenty is then calculated by adding up the outputs corresponding to each filter. Note that, in addition to the cross entropy loss, the term extraction loss Lenty can also use losses obtained by other similar loss calculation formulas.

[0080]

number

[0081] (7-3. Second learning phase) The second learning phase, which is the second learning step following the first learning phase, will be described with reference to Fig. 10. In the second learning phase, a term extraction loss Lenity generated in the extraction of terms by the term extraction unit 13 and a relationship extraction loss Lrelation generated in the extraction of relationships by the relationship extraction unit 17 are used. In the second learning phase, the second learning step is performed by updating the generation parameters of the intermediate representation H2 in the feature representation generation unit 12 based on the term extraction loss Lenity and the relationship extraction loss Lrelation. That is, in the second learning phase, learning is performed by a multitask learning process that updates the generation parameters based on the term extraction loss Lenity and the relationship extraction loss Lrelation.

[0082] Here, the relationship extraction loss Lrelation is expressed by formula (18). r is the number of types of relation labels (see FIG. 4). Since a loss occurs for each relation, the sum of all of these is the relationship extraction loss Lrelation. Furthermore, Nentity is the number of constituent terms of the nested structure 20 targeted for term extraction.

[0083]

number

[0084] Then, the term extraction loss Lenity expressed by the above formula (17) and the relationship extraction loss Lrelation shown in formula (18) are used to generate the overall loss Lall as shown in formula (19).

[0085]

number

[0086] Here, as in the first learning phase, for example, cross entropy loss can be used to express the loss. The cross entropy loss is expressed by Equation (20). Class represents a label that is a target for relationship extraction.

[0087]

number

[0088] (8. Example of Processing by Knowledge Model Creation Unit 8) Next, examples of processing by the knowledge model creation unit 8 will be described with reference to Figs. 11 to 14. However, the processing by the knowledge model creation unit 8 is not limited to the following, and various processing is possible. Furthermore, the knowledge model creation unit 8 can be fully automated.

[0089] (8-1, first example) The processing of the first example will be described with reference to Fig. 11. The knowledge model creation unit 8 draws a knowledge network diagram consisting of node figures representing factors and link figures representing relationships between the factors in a drawable drawing GUI window 30. The drawing operation can be performed by the operator.

[0090] When an operator draws a knowledge network diagram in the drawing GUI window 30, first, the operator determines a focus factor. Then, the knowledge model creation unit 8 displays, in the drawing GUI window 30, arrangement candidates for node shapes of terms having a relationship with the focus factor, based on information on the relationship in the DBs 3, 5, and 7. At this time, a plurality of arrangement candidates can be displayed. In particular, by using a plurality of DBs 3, 5, and 7, it becomes possible to output arrangement candidates from a plurality of viewpoints. Then, by determining the arrangement candidate, the operator can draw (arrange) the focus factor in the knowledge network diagram in the drawing GUI window 30.

[0091] (8-2. Second example) The processing of the second example will be described with reference to Fig. 12. In the drawing GUI window 30, the operator selects an already drawn factor of interest. Then, the knowledge model creation unit 8 displays terms having a relationship with the factor of interest as factor candidates based on the information on the relationships in the DBs 3, 5, and 7. Then, the operator selects from among the factor candidates, and the selected factor candidate is placed in the drawing GUI window 30 in a state associated with the factor of interest.

[0092] (8-3. Third example) The processing of the third example will be described with reference to Fig. 13. The worker inputs document data into the DB creation units 2, 4, and 6. Then, the knowledge model creation unit 8 displays a plurality of terms obtained based on the first DB 3 (term-relationship DB) in the GUI window 30. Then, when the worker selects one of the displayed plurality of terms, the knowledge model creation unit 8 displays, in the GUI window 30, terms having a relationship with the selected term based on the information on the relationship in the DBs 3, 5, and 7.

[0093] That is, the term selected by the operator and terms related to the selected term are displayed in the GUI window 30. When the operator selects multiple displayed terms, the knowledge model creation unit 8 draws a knowledge network diagram in the GUI window 30.

[0094] (8-4. Fourth example) The processing of the fourth example will be described with reference to Fig. 14. An operator inputs document data into the DB creation units 2, 4, and 6. Then, the knowledge model creation unit 8 displays existing similar knowledge model candidates in the GUI window 30. When there are multiple similar knowledge model candidates, the similar knowledge model candidate selected by the operator is displayed in the GUI window 30.

[0095] Next, the knowledge model creation unit 8 displays a plurality of terms obtained based on the first DB3 (term-relationship DB) in the GUI window 30 based on the input document data. Then, when the operator selects one of the displayed plurality of terms, the knowledge model creation unit 8 displays, in the GUI window 30, terms having a relationship with the selected term based on the information on the relationship in the DB3, 5, 7.

[0096] That is, the GUI window 30 displays the similar knowledge model, the term selected by the operator, and terms related to the selected term. In this case, when the term selected by the operator is set as a factor of interest, the knowledge model creation unit 8 displays, for example, a cross mark or the like as a link figure that does not have a relationship with the factor of interest in the similar knowledge model displayed in the GUI window 30. Furthermore, when the term selected by the operator is not included in the similar knowledge model, the knowledge model creation unit 8 may newly draw the selected term.

[0097] (9. Effects) According to the knowledge model creation support device 1, DB3 is created using a term-relationship extraction model that extracts terms and relationships between terms from document data. First, the term-relationship extraction model automatically extracts terms from document data. Furthermore, the term-relationship extraction model pre-stores trigger words that lead to relationships between two terms, and when the trigger words are included in the document data, extracts the relationship between the two terms based on the trigger words. In this way, the term-relationship extraction model can automatically extract terms from document data, and can automatically extract relationships between terms included in the document data by taking into account the preset trigger words. Therefore, DB3 can be created without manual work. Then, if DB3 can be created, a knowledge model can be created by referring to DB3, so that a knowledge model can be easily created. [Explanation of symbols]

[0098] 1: knowledge model creation support device, 2: first database creation unit, 3: first database, 4: second database creation unit, 5: second database, 6: second database creation unit, 7: third database, 8: knowledge model creation unit, 11: document data acquisition unit, 12: feature expression generation unit, 13: term extraction unit, 14: term output unit, 15: trigger word storage unit, 16: pair extraction unit, 17: relationship extraction unit, 18: relationship output unit, 20: nested structure, 21: nested outer term, 22, 23: contained term, 100: knowledge network diagram, 110: node diagram, 120: link diagram, 121: first link diagram, 122: second link diagram, Lentity: term extraction loss, Lrelation: relationship extraction loss

Claims

1. An apparatus for supporting the creation of a knowledge model composed of a plurality of factors defined by terms and relationship information between the factors, comprising: a term-relationship database including a plurality of terms and relationships of the plurality of terms; a term-relationship database creation unit that creates the term-relationship database by extracting the terms and the relationships from document data using a term-relationship extraction model; a knowledge model creation unit that creates the knowledge model based on the term-relationship database, The term-relationship extraction model is a document data acquisition unit for acquiring the document data; a term extraction unit that extracts terms from the document data; a trigger word storage unit for storing a trigger word that leads to a relationship between two terms; a pair extraction unit that creates term pair information from the terms extracted by the term extraction unit; a relationship extraction unit that extracts a relationship between the two terms extracted by the term extraction unit based on the trigger word and the pair information when the trigger word is included in the document data, the document data includes a nested structure including a plurality of terms; The nested structure is a structure that is composed of a plurality of inclusive terms and a nested outer term that combines the plurality of inclusive terms, the term extraction unit extracts, when the document data includes the nested structure, the inclusive term and the nested outer term in the nested structure from the document data; when the document data includes the nested structure, the pair extraction unit creates the pair information by using only the nested outer term in the nested structure without using the inclusive term in the nested structure; A knowledge model creation support device, wherein when the document data includes the nested structure, the relationship extraction unit treats the nested outer term in the nested structure as a relationship extraction target, and does not treat the contained term as a relationship extraction target.

2. 2. The knowledge model creation support device according to claim 1, wherein the trigger word is a keyword that indicates a change in a term that represents a physical quantity.

3. The term-relationship extraction model further comprises: a feature representation generating unit that generates a feature representation in which a feature of the document data is expressed by a vector based on each token constituting the document data acquired by the document data acquiring unit, 3 . The knowledge model creation support device according to claim 1 , wherein the term extraction unit and the relationship extraction unit extract the terms and the relationships by sharing the characteristic expressions generated by the characteristic expression generation unit.

4. An apparatus for supporting the creation of a knowledge model composed of a plurality of factors defined by terms and relationship information between the factors, comprising: a term-relationship database including a plurality of terms and relationships of the plurality of terms; a term-relationship database creation unit that creates the term-relationship database by extracting the terms and the relationships from document data using a term-relationship extraction model; a knowledge model creation unit that creates the knowledge model based on the term-relationship database, The term-relationship extraction model is a document data acquisition unit for acquiring the document data; a feature representation generating unit that generates a feature representation in which a feature of the document data is expressed as a vector based on each token constituting the document data acquired by the document data acquiring unit; a term extraction unit that extracts terms from the document data; a trigger word storage unit for storing a trigger word that leads to a relationship between two terms; a relationship extraction unit that extracts a relationship between two terms extracted by the term extraction unit based on the trigger word when the trigger word is included in the document data, A knowledge model creation support device, wherein the term extraction unit and the relationship extraction unit share the characteristic representation generated by the characteristic representation generation unit to extract the terms and the relationships.

5. The term-relationship extraction model is 5. The knowledge model creation support device according to claim 3, wherein, as a learning phase, learning is performed by a multi-task learning process that updates generation parameters of the feature representation in the feature representation generation unit based on a term extraction loss generated by the term extraction unit in extracting the terms and a relationship extraction loss generated by the relationship extraction unit in extracting the relationships.

6. The term-relationship extraction model is In a first learning phase, learning is performed by a single-task learning process that updates the generation parameters based on the term extraction loss; 6. The knowledge model creation support device according to claim 5, wherein learning is performed by the multitask learning process as a second learning phase following the first learning phase.

7. The knowledge model creation support device comprises: a first database, which is the term-relationship database created using the term-relationship extraction model; a second database that is created using a model different from the term-relationship extraction model and includes relationships between at least a plurality of terms; Equipped with The knowledge model creation unit 7. The knowledge model creation support device according to claim 1, wherein the knowledge model is created based on the first database and the second database.

8. The knowledge model creation unit Draw a knowledge network diagram in a drawing GUI window, the knowledge network diagram being composed of node figures representing the factors and link figures representing relationships between the factors; A knowledge model creation support device as described in any one of claims 1 to 7, which performs at least one of the following when drawing the knowledge network diagram: displaying placement candidates for the node figures of terms that have a relationship with the factor of interest based on the term-relationship database; displaying terms that have a relationship with the factor of interest based on the term-relationship database as factor candidates; and displaying link figures that have no relationship with the factor of interest based on the term-relationship database.

Images