CN114742068A - Multi-sentence correlation analysis method and system for ISO19650 standard text - Google Patents

Abstract

The invention provides a multi-sentence correlation analysis method and a multi-sentence correlation analysis system for ISO19650 standard texts, which relate to the technical field of information processing, and comprise the following steps: step S1: performing word segmentation and word change processing on sentences in an ISO19650 standard series to obtain preprocessed sentences; step S2: performing dependency syntax analysis on the preprocessed sentences to obtain dependency relationships among words in the sentences; step S3: reasoning is carried out on the dependency relationship among the words in the sentences according to the conversion rule from the dependency relationship to the semantic relationship to obtain the semantic relationship among the words in a single sentence; step S4: and importing the semantic relation in a single sentence into a graph database, importing an ontology model of the ISO standard into the graph database, establishing a link between the words in each sentence and the words in the ontology model, and reasoning the incidence relation among the sentences. The method can overcome the difficulty in extracting semantic information of the ISO19650 standard Chinese text caused by the shortage of the corpus, and is also helpful for solving the difficulty in automatic analysis of association and reference among ISO19650 sentences.

Description

Method and system for multi-statement association analysis of ISO 19650 standard text

technical field

The invention relates to the technical field of information processing, in particular, to a method for association analysis between multiple sentences in ISO19650 standard series texts based on NLP and ontology model.

Background technique

Engineering project development requires clear and timely information from all involved. In addition to the IFC file format, they also needed an information management framework to support their collaboration. The ISO 19650 family of standards provides such a framework for establishing reliable sources of information. Since the 5-part ISO 19650 family of standards forms a complex system, the construction industry hopes to capture the semantic information in these standards.

However, manual extraction of semantic information in the ISO 19650 standard series is not only time-consuming but also costly. Therefore, it is necessary to develop a method for extracting semantic information based on NLP, and then automatically analyze the association and reference relationship between various standard clauses with the help of the ontology model of the ISO 19650 standard.

The invention patent with publication number CN110096692B discloses a semantic information processing method and device. The semantic information processing method includes dividing the question stem into known conditions and conclusions according to the obtained question stem; according to the obtained known condition and conclusion, extract explicit semantic information in known conditions and conclusions; when implicit semantic information exists in known conditions and/or conclusions, extract implicit semantic information in known conditions and/or conclusions; merge the extracted Explicit semantic information and implicit semantic information, get the semantic information of the question stem.

SUMMARY OF THE INVENTION

Aiming at the defects in the prior art, the present invention provides a method and system for multi-sentence association analysis of ISO 19650 standard text.

According to a method and system for multi-statement association analysis of ISO 19650 standard text provided by the present invention, the scheme is as follows:

In a first aspect, a multi-sentence association analysis method of ISO 19650 standard text is provided, and the method includes:

Step S1: performing word segmentation and word replacement processing on the sentences in the ISO 19650 standard series to obtain preprocessed sentences;

Step S2: perform dependency syntax analysis on the preprocessed statement to obtain the dependency relationship between words in the statement;

Step S3: According to the conversion rule of the dependency relationship to the semantic relationship, inference is performed on the dependency relationship between words in the sentence, and the semantic relationship between the words in a single sentence is obtained;

Step S4: import the semantic relationship in a single sentence into the graph database, import the ISO standard ontology model into the graph database, establish the links between the words in each sentence and the words in the ontology model, and infer the association between multiple sentences.

Preferably, the step S1 includes:

Step S1.1: Obtain the text files of the Chinese version of the ISO 19650 standard series;

Step S1.2: extract sentences according to each standard entry, and perform sentence segmentation;

Step S1.3: Change words for the words obtained by word segmentation, and replace professional terms with hypernyms.

Preferably, the text file is a docx file, and the open source ZLib library is used to decompress the docx file into a set of XML files, and then from these decompressed files, the XML files are analyzed according to the entry encoding rules of the ISO 19650 standard series, and the standard entries are extracted therefrom. content, remove all font and paragraph typography, and finally generate a plain text file with a list of statements.

Preferably, the step S2 includes: performing a syntax tree analysis on the sentence through a dependency parser, marking a part of speech for each word in the sentence, finding out the central word in the sentence, and determining the non-central word associated with the central word , the non-central word is used as the central word to start the next round of related non-central word search, and finally a multi-level dependency syntax tree is obtained.

Preferably, the step S3 includes: reasoning about the semantic relationship, designing a mapping rule between the dependency relationship and the semantic relationship, and converting the dependency syntax tree into a binary semantic relationship according to the mapping rule.

In a second aspect, a multi-sentence association analysis system for ISO 19650 standard text is provided, and the system includes:

Module M1: Perform word segmentation and word replacement processing on sentences in the ISO 19650 standard series to obtain preprocessed sentences;

Module M2: perform dependency syntax analysis on the preprocessed statement to obtain the dependency relationship between words in the statement;

Module M3: According to the conversion rule of the dependency relationship to the semantic relationship, infer the dependency relationship between the words in the sentence, and obtain the semantic relationship between the words in a single sentence;

Module M4: Import the semantic relationship in a single sentence into the graph database, import the ISO standard ontology model into the graph database, establish the link between the words in each sentence and the words in the ontology model, and infer the association between multiple sentences.

Preferably, the module M1 includes:

Module M1.1: Obtain the text files of the Chinese version of the ISO 19650 standard series;

Module M1.2: Extract sentences according to each standard entry, and perform sentence segmentation;

Module M1.3: Swap words obtained from word segmentation, and replace professional terms with hypernyms.

Preferably, the text file is a docx file, and the open source ZLib library is used to decompress the docx file into a set of XML files, and then from these decompressed files, the XML files are analyzed according to the entry encoding rules of the ISO 19650 standard series, and the standard entries are extracted therefrom. content, remove all font and paragraph typography, and finally generate a plain text file with a list of statements.

Preferably, the module M2 includes: performing a syntax tree analysis on the sentence through a dependency parser, marking a part of speech for each word in the sentence, finding out the central word in the sentence, and determining the non-central word associated with the central word , the non-central word is used as the central word to start the next round of related non-central word search, and finally a multi-level dependency syntax tree is obtained.

Preferably, the module M3 includes: semantic relationship reasoning, designing a mapping rule between dependencies and semantic relationships, and converting the dependency syntax tree into binary semantic relationships according to the mapping rules.

Compared with the prior art, the present invention has the following beneficial effects:

The invention infers the domain semantic relationship from the syntactic relationship through the mapping rules, which greatly overcomes the difficulty caused by the insufficient corpus, and verifies the feasibility and practicability of the proposed information extraction method through experiments.

Description of drawings

Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings:

Fig. 1 is the overall flow schematic diagram of the present invention;

Fig. 2 is the frame schematic diagram of the dependency syntax tree;

Figure 3 is an example of a dependency tree of a sentence;

Figure 4 is a schematic diagram of a typical semantic relationship;

Fig. 5 is the conversion schematic diagram from syntactic relation to semantic relation;

Fig. 6 is the knowledge graph stored in the Neo4J graph database;

FIG. 7 is a case of inferring the association relationship between standard sentences.

Detailed ways

The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the inventive concept. These all belong to the protection scope of the present invention.

An embodiment of the present invention provides a method for analyzing the association between multiple sentences based on NLP standard text. Referring to FIG. 1 , the method specifically includes:

Step S1: Perform word segmentation and word replacement processing on the sentences in ISO 19650 to obtain preprocessed sentences.

The step S1 specifically includes:

Step S1.1: Obtain the file of the Chinese translation text of the ISO 19650 standard series.

Step S1.2: Extract sentences according to standard items, and perform word segmentation on the sentences.

Step S1.3: Replacing the words obtained by the segmentation (named entities corresponding to the words in the sentence), and replacing the professional terms with the hypernyms. The replaced hypernyms are words with higher recognition and analysis accuracy by DDParser.

Among them, the text file is a docx file, use the open source ZLib library to decompress the docx file into a set of XML files, and then analyze the XML file from these decompressed files according to the ISO 19650 entry coding rules, extract the standard entry content from it, delete all fonts and paragraph typesetting, and finally generate a plain text file containing a list of statements.

Step S2: Perform dependency syntax analysis on the preprocessed sentence to obtain the dependency relationship between words in the sentence. Syntax tree analysis is performed on the sentence through the dependency parser, each word in the sentence is marked with a part of speech, the central word in the sentence is found, the non-central word associated with the central word is determined, and the non-central word is used as the central word again Start the next round of searching for related non-center words, and finally obtain a multi-level dependency syntax tree.

Step S3: According to the conversion rule of the dependency relationship to the semantic relationship, inference is performed on the dependency relationship between words in the sentence, and the semantic relationship between the words in a single sentence is obtained. Semantic relationship reasoning, designing a mapping rule between dependencies and semantic relationships, and converting the dependency syntax tree into binary semantic relationships according to the mapping rules.

Step S4: import the semantic relationship in a single sentence into the graph database, import the ISO standard ontology model into the graph database, establish the links between the words in each sentence (named entities corresponding to the words in the sentence) and the words in the ontology model, and reason multiple relationship between sentences.

The present invention also provides an association analysis system between multiple sentences based on NLP standard text, the system specifically includes:

Module M1: Perform word segmentation and word replacement processing on sentences in ISO 19650 to obtain preprocessed sentences.

Specifically included in module M1:

Module M1.1: Obtain files for Chinese translation texts of the ISO 19650 standard series.

Module M1.2: Extract sentences according to standard items and segment the sentences.

Module M1.3: Replace the words obtained by the segmentation (named entities corresponding to the words in the sentence), and replace the professional terms with the hypernyms. The replaced hypernyms are words with high recognition and analysis accuracy by DDParser.

Among them, the text file is a docx file, use the open source ZLib library to decompress the docx file into a set of XML files, and then analyze the XML file from these decompressed files according to the ISO 19650 entry coding rules, extract the standard entry content from it, delete all fonts and paragraph typesetting, and finally generate a plain text file containing a list of statements.

Module M2: Perform dependency syntax analysis on the preprocessed statement to obtain the dependency relationship between words in the statement. Syntax tree analysis is performed on the sentence through the dependency parser, each word in the sentence is marked with a part of speech, the central word in the sentence is found, the non-central word associated with the central word is determined, and the non-central word is used as the central word again Start the next round of searching for related non-center words, and finally obtain a multi-level dependency syntax tree.

Module M3: According to the conversion rule of the dependency relationship to the semantic relationship, infer the dependency relationship between the words in the sentence, and obtain the semantic relationship between the words in a single sentence. Semantic relationship reasoning, designing a mapping rule between dependencies and semantic relationships, and converting the dependency syntax tree into binary semantic relationships according to the mapping rules.

Module M4: Import the semantic relationship in a single sentence into the graph database, import the ISO standard ontology model into the graph database, establish the link between the words in each sentence and the words in the ontology model, and infer the association between multiple sentences.

Next, the present invention will be described in more detail.

The present invention provides an association analysis method among multiple sentences in ISO 19650 standard series texts based on NLP and ontology model, in particular a semantic information extraction and sentence association analysis method of ISO 19650 standard series Chinese translation texts based on NLP and ontology model, The method includes:

1. Preprocessing of ISO 19650 text:

Referring to Figure 1, it shows the process of preprocessing sentences in the ISO 19650 standard series. First, the sentences are segmented, and then the specialized terms are replaced with their hypernyms.

The original standard text in English is first translated into Chinese and stored in Microsoft docx format. All sentences in each part of the ISO standard are then extracted using the preprocessing procedure shown in Figure 1.

A docx file is actually a zip file of a collection of XML files, containing textual content and markup for formatting or typesetting definitions. Use the open source ZLib library to extract docx files into a set of XML files, then extract standard text content from these unzipped files, removing all fonts and paragraph typography. In this way, the program generates a plain text file containing a list of sentences.

Unlike English sentences, Chinese sentences have no space to split characters into words, which makes it difficult to evaluate the relationship between words, so Jieba library is used to split each sentence in ISO 19650 standard text into an array of multiple Chinese words , each Chinese word contains one or more Chinese characters. In addition, since Jieba's ability to recognize new words is limited, in order to enhance Jieba's ability to recognize new words, it is necessary to use ISO 19650 professional vocabulary. For example, in the case of not using the ISO19650 professional vocabulary, the proprietary vocabulary "public data environment" is usually divided into three words "public", "data" and "environment" by Jieba, which is obviously not the expected result. Therefore, after incorporating the ISO 19650 professional vocabulary containing the professional term "common data environment" into the Jieba special dictionary, the correct word segmentation results can be obtained.

After the word segmentation is completed, the professional term is then replaced with its hypernym, and the replaced hypernym is a vocabulary with high recognition and analysis accuracy by DDParser. In order to realize the above-mentioned replacement, in the ISO 19650 professional vocabulary, each professional term vocabulary is defined with its hypernym. For example, the hypernym of "common data environment" is "information source". Because many specialties in ISO 19650 will make the next step in the analysis of dependency syntactic relations difficult, they are replaced by hypernyms that are easier to identify and analyze.

2. Semantic information extraction from ISO 19650 standard text:

2.1. Obtaining the dependency syntactic relationship:

The structure of a sentence can be obtained through dependency syntax analysis. This process will mark the part-of-speech of each Chinese word, and at the same time determine its dependencies with other components in the sentence, and finally interpret the sentence as a syntactic dependency tree, which is more convenient for follow-up. Computer comprehension of sentences. Use the open source Baidu Dependency Parser (DDParser) to analyze the dependencies between a series of words.

Referring to Figure 2, DDParser develops a syntactic analysis model based on the DeepBiaffineAttention model. The input vector of the ith word e _i is the concatenation of its embedding vector e ^word _i and the character-level LSTM vector charLSTM( _wi ), as follows:

e _i =e ^word _i ⊕charLSTM(wi _i )

where charLSTM( _wi ) is the concatenated vector resulting from sequentially feeding each character in the ith word to the BiLSTM layer. Then, each e _i is input into a three-layer BiLSTM, resulting in a high-dimensional vector _ri . Subsequently, the dimensionality of each _ri vector is reduced by Multilayer Perception (MLP). This dimensionality reduction can exclude information that has little effect on the dependencies between words. Finally, the dependencies and their syntactic types are identified using bi-affine attention.

Each word can be regarded as a head item or a dependency, and after its word vector is input into the deep bi-affine attention model, the dependency arc S _i ^arc and the relation score S _i ^rel are calculated respectively. Based on these two scores, the dependency relationship between the corresponding two words can be calculated. For the example in Figure 2, the syntactic relationship between "Is" and "Informationsource" is evaluated as Verb-Object(VOB).

DDParser has been trained on the manually labeled dataset DuCTB, which annotates the annotated dependencies between two Chinese words. DuCBT has a total of 24 part-of-speech tags and 14 kinds of dependencies. Table 1 below lists some of the dependencies.

Table 1

dependencies explain HED the center of the entire sentence SBV The relationship between subjects and predictions (Head) VOB The relationship between the prediction (Head) and the target ATT The relationship between attribute and noun (Head) ADV The relationship between adverbial and verb (Head) POB The relationship between preposition and object (Head) COO Co-occurrence between two words

Figure 3 shows an example of a dependency syntax tree. A directed arc link is issued from the head word to the modifier, each directed arc link represents a grammatical aspect dependency. For example, there is an arc between the subject "common data environment" and the predicate "is", indicating the dependency "SBV", and the object "information source" also has a dependency "VOB" with the same predicate "is". Further, the semantic relationship between two words can be inferred from these two dependencies.

2.2. Semantic relational reasoning:

DDParser mainly deals with the syntactic structure of sentences, not semantic relations. Syntactic relations focus on conveying information about the grammatical structure of sentences, while semantic information extraction focuses on semantic connections between entities (concepts). Therefore, semantic relations are more important for understanding the text, and the associations between multiple sentences can be inferred from these semantic relations later.

In order to infer the semantic information from the grammatical structure of the sentence, we design a set of mapping rules from one or more dependencies to semantic relations. Table 2 lists typical mapping rules. Using these mapping rules, the dependency syntactic relations derived by DDParser can be transformed into semantic relations. The letters A, I, F, O in the table represent entity type, information, function and others, respectively.

Table 2

In addition, in order to better explain the semantic relationship, through the manual analysis of the ISO 19650 series, an ontology model of ISO 19650 is proposed, which defines three core concepts of information, role and behavior, and defines one of the three core concepts. The semantic relationship between them is shown in Figure 4.

Referring to FIG. 5 , the conversion of the dependencies in FIG. 3 into semantic relationships is illustrated. For example, following the mapping rules described in Table 2, two grammatical relations SBV ("Common Data Context" and "is") and VOB ("source of information" and "is") are mapped to a semantic between subject and object Relationship "underlying". That is, "common data environment" is a synonym for "information source". At the same time, the ATT (attributive relationship) between "information source" and "approval" is converted into a semantic relationship "negotiation" between the two entities, and the central word "information source" of "negotiation" is "information" type, The modifier "approval" is of the "functional" type and is the trigger word for the "negotiation" of the semantic relationship. Furthermore, two other syntactic relations ATT and ADV are transformed into two semantic relations "attribute" and "restriction".

3. Article association reasoning based on graph database:

When referring to the ISO 19650 standard text contains many interrelated sentences, these sentences are either related to each other through shared concepts, or related to each other through the ISO 19650 ontology model. When reviewing the ISO 19650 standard series of texts, it is often necessary to determine the references or dependencies between individual statements. Therefore, the graph database Neo4J is used to describe the semantic relations obtained by the analysis in the aforementioned standard sentences, the ISO 19650 ontology model and the links between the two. Compared with SQL relational databases, Neo4J can define rich attribute descriptions for edges, which can greatly reduce the connection operations required by SQL databases, thereby improving the speed of query and reasoning.

Specifically, a node is used to represent a named entity, and a directed edge is used to represent a binary relationship between two nodes. Both nodes and edges can have multiple properties. The association of sentences with ontology models can also be represented by the graph database Neo4J. In Neo4J, a sequence of words in the same sentence is represented by the "AFTER" relation from the previous word to the next word. For example, Figure 6 shows that the syntactic relationship is represented by the "Depend_On" relationship of modifiers to head words. The type of syntactic relationship is represented as an attribute of the "Depend_On" edge.

The graph structure of Neo4J database can be further used for knowledge reasoning. Figure 7 shows a typical case of inference using the graph database Neo4J. The sentence in Section 3.3.15 of ISO 19650 Part 1 indicates that "common" modifies the central word "approval", and the hypernym of approval is "behavior". Through the ontology model, it can be inferred that "behavior" must have an "executor". A "role" is an actor as defined in ISO 19650 Part 1, Section 3.2.1. Meanwhile, in the second sentence of Figure 7, there are three hyponyms of "executor", "person", "organization" and "unit". This means that these people, organizations and units involved in the project construction process should agree on what constitutes a common data environment as it is a shared source of information. The above reasoning process can be implemented by the following Cypher query code:

MATCH (a: Common Data Environment)

WHERE(a)-[*]-(b: common)

MATCH (c: role)

WHERE(b)-[:Trigger_Word]->(c)

MATCH(d)

WHERE(c)-[:Hyponymy]-[d]

RETURNd

Engineering infrastructure project management requires the sharing of BIM information among all project participants. Project participants need a widely recognized BIM information management collaboration framework. The ISO 19650 family of standards provides concepts, principles and processes for establishing a reliable source of shared information. But the 5-part ISO 19650 standard series constitutes a very complex text system, and infrastructure builders hope to have an efficient tool to capture the semantic information in these standards, so as to better discover and infer the content of the standard provisions. relationships and references.

Those skilled in the art know that, in addition to implementing the system provided by the present invention and its various devices, modules, and units in the form of purely computer-readable program codes, the system provided by the present invention and its various devices can be implemented by logically programming the method steps. , modules, and units realize the same function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system provided by the present invention and its various devices, modules and units can be regarded as a kind of hardware components, and the devices, modules and units included in it for realizing various functions can also be regarded as hardware components. The device, module and unit for realizing various functions can also be regarded as both a software module for realizing the method and a structure within a hardware component.

Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above-mentioned specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essential content of the present invention. The embodiments of the present application and features in the embodiments may be combined with each other arbitrarily, provided that there is no conflict.

Claims (10)

1. A multi-sentence correlation analysis method of ISO19650 standard texts is characterized by comprising the following steps:

step S1: performing word segmentation and word change processing on sentences in an ISO19650 standard series to obtain preprocessed sentences;

step S2: performing dependency syntax analysis on the preprocessed sentences to obtain dependency relationships among words in the sentences;

step S3: reasoning is carried out on the dependency relationship among the words in the sentences according to the conversion rule from the dependency relationship to the semantic relationship to obtain the semantic relationship among the words in a single sentence;

step S4: and importing the semantic relation in a single sentence into a graph database, importing an ontology model of the ISO standard into the graph database, establishing a link between the words in each sentence and the words in the ontology model, and reasoning the incidence relation among the sentences.

2. The method for multiple sentence correlation analysis according to ISO19650 standard text, wherein the step S1 comprises:

step S1.1: acquiring a text file of a Chinese edition ISO19650 standard series;

step S1.2: extracting sentences according to each standard entry, and performing sentence segmentation;

step S1.3: and (5) changing words obtained by word segmentation, and replacing the professional terms with hypernyms.

3. The method for multi-sentence correlation analysis of ISO 19650-standard text according to claim 2, wherein the text file is a docx file, the docx file is decompressed into a set of XML files by using an open source ZLib library, and then the XML files are analyzed from the decompressed files according to the entry coding rules of ISO 19650-standard series, standard entry contents are extracted from the XML files, all font and paragraph layouts are deleted, and finally a plain text file containing a sentence list is generated.

4. The method for multiple sentence correlation analysis according to ISO19650 standard text, wherein the step S2 comprises: and performing syntax tree analysis on the sentence through a dependency relationship analyzer, marking a part of speech for each word in the sentence, finding out a central word in the sentence, determining a non-central word associated with the central word, starting the next round of searching for the relevant non-central word by taking the non-central word as the central word, and finally obtaining a multilayer dependency syntax tree.

5. The method for multiple sentence correlation analysis according to ISO19650 standard text, wherein the step S3 comprises: and semantic relation reasoning, namely designing a mapping rule from the dependency relation to the semantic relation, and converting the dependency syntax tree into a binary semantic relation according to the mapping rule.

6. A multi-sentence correlation analysis system for ISO19650 standard text, comprising:

module M1: performing word segmentation and word change processing on sentences in an ISO19650 standard series to obtain preprocessed sentences;

module M2: performing dependency syntax analysis on the preprocessed sentences to obtain dependency relationships among words in the sentences;

module M3: reasoning is carried out on the dependence relationship among the words in the sentences according to the conversion rule from the dependence relationship to the semantic relationship to obtain the semantic relationship among the words in a single sentence;

module M4: and importing the semantic relation in a single sentence into a graph database, importing an ontology model of the ISO standard into the graph database, establishing a link between the words in each sentence and the words in the ontology model, and reasoning the incidence relation among the sentences.

7. The ISO19650 multi-sentence correlation analysis system of standard text according to claim 6, wherein the module M1 comprises:

module M1.1: acquiring text files of a Chinese edition ISO19650 standard series;

module M1.2: extracting sentences according to each standard entry, and performing sentence segmentation;

module M1.3: and (5) changing words obtained by word segmentation, and replacing the professional terms with hypernyms.

8. The ISO19650 multi-lingual correlation analysis system for standard text according to claim 7, wherein the text file is a docx file, the open source ZLib library is used to decompress the docx file into a set of XML files, and then the XML files are analyzed from the decompressed files according to the entry encoding rules of the ISO19650 standard series, so as to extract the standard entry contents from the decompressed files, delete all font types and paragraph types, and finally generate a plain text file containing the sentence list.

9. The ISO19650 multi-sentence correlation analysis system of standard text according to claim 6, wherein the module M2 comprises: and performing syntax tree analysis on the sentence through a dependency relationship analyzer, marking a part of speech for each word in the sentence, finding out a central word in the sentence, determining a non-central word associated with the central word, starting the next round of searching for the relevant non-central word by taking the non-central word as the central word, and finally obtaining a multilayer dependency syntax tree.

10. The ISO19650 multi-sentence correlation analysis system of standard text according to claim 6, wherein the module M3 comprises: and semantic relation reasoning, namely designing a mapping rule from the dependency relation to the semantic relation, and converting the dependency syntax tree into a binary semantic relation according to the mapping rule.

Images