A method and system for automatically constructing domain ontology

Abstract

The present invention discloses a method and system for automatically constructing a domain ontology, comprising the following steps: collecting domain text data from multiple data sources and preprocessing them; performing knowledge extraction and representation on the preprocessed data, including extracting entities, relationships and attributes from the domain text data, and establishing a knowledge representation model; assigning weights to each piece of knowledge according to the source and quality of the knowledge, integrating knowledge from different sources, and establishing a knowledge base; ontology modeling is based on the integrated knowledge base, and domain ontology is constructed using an ontology modeling tool; ontology reasoning engine is used to perform reasoning and inference, and in the reasoning process, the membership of fuzzy relations between entities is calculated according to the membership function of the fuzzy relations, thereby obtaining a reasoning result. The present invention combines fuzzy logic reasoning methods to handle uncertainty and fuzziness, which can improve the intelligence and robustness of the system, and enable the system to handle complex domain knowledge and reasoning tasks more flexibly.

Description

Technical Field

[0001] The present invention relates to the field of domain ontology technology, and in particular to a method and system for automatically constructing a domain ontology. Background Art

[0002] Domain ontology is a formalized model based on knowledge in a specific domain, used to represent concepts, relationships and their attributes in the domain. The automatic construction method of domain ontology needs to solve the problem of dealing with uncertainty and ambiguity. The automatic construction of domain ontology depends on multiple factors, including the complexity of the domain involved, available data, language processing technology, and the purpose of building the ontology. Therefore, the automatic construction of domain ontology includes how to deal with the uncertainty and ambiguity of knowledge in the real world, and how to represent and reason about these uncertainties and ambiguities in the ontology to improve the system's ability to deal with uncertainty and ambiguity. Summary of the invention

[0003] In order to solve the above problems, the present invention provides a method and system for automatically constructing a domain ontology.

[0004] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] On the one hand, the present application discloses a method for automatically constructing a domain ontology, comprising the following steps:

[0006] Step 1: Collect domain text data from multiple data sources and preprocess them;

[0007] Step 2: Extract and represent knowledge from the preprocessed data, including extracting entities, relations, and attributes from domain text data, and establishing a knowledge representation model;

[0008] Step 3: Assign weights to each piece of knowledge based on its source and quality, integrate knowledge from different sources, and build a knowledge base;

[0009] Step 4: Ontology modeling Based on the integrated knowledge base, use the ontology modeling tool to build the domain ontology;

[0010] Step 5: Use the ontology reasoning engine to perform reasoning and inference. During the reasoning process, the membership of the fuzzy relationship between entities is calculated according to the membership function of the fuzzy relationship, thereby obtaining the reasoning result.

[0011] Further: Step 1 comprises:

[0012] Clean and filter the data to ensure its quality and accuracy. The cleaning steps include:

[0013] Text denoising: remove HTML tags, special characters, and garbled noise information;

[0014] Text standardization: converting text into a unified format, including case conversion, word form restoration, and spelling correction;

[0015] Entity recognition: Use NLP technology to perform entity recognition on text and identify domain entities in the text, including names of people, places, and institutions;

[0016] Relationship extraction: Use NLP technology to extract the relationships between entities from the text, including cooperative relationships between people and associations between locations.

[0017] Further: Step 2 comprises:

[0018] Perform entity recognition on the preprocessed domain text to identify the entities in the text;

[0019] After identifying the entities, extract the relationships between the entities and the attribute information of the entities;

[0020] The extracted entities, relations and attributes are represented as a knowledge representation model. In the representation process, fuzzy sets are used to represent fuzzy or uncertain knowledge. Fuzzy sets are represented as:

[0021] A={(x,μ A (x)|x∈X}

[0022] Where A is a fuzzy set, X is the domain, μ A (x) is the membership function of x in the fuzzy set A.

[0023] Further: Step 3 includes:

[0024] The entities, relations and attributes obtained through preprocessing and knowledge extraction are constructed into a knowledge graph, where the nodes of the knowledge graph represent entities and the edges represent the relations between entities.

[0025] According to the source and quality of knowledge, each piece of knowledge is given a weight, and the knowledge from different sources is integrated using the following methods:

[0026]

[0027] Where F is the fused knowledge, K i is the knowledge from the ith source, ω i is the weight of the ith source, n is the number of knowledge sources;

[0028] Integrate the fused knowledge into a unified knowledge base.

[0029] Further: Step 4 includes:

[0030] Import the integrated knowledge base into the ontology modeling tool to build the domain ontology;

[0031] In ontology modeling tools, the uncertainty and ambiguity of relationships are expressed by defining fuzzy degrees and membership functions of fuzzy sets;

[0032] Embed the knowledge represented by fuzzy sets into the ontology, including embedding the definitions of fuzzy attributes and fuzzy relations into the ontology, and associating the membership function information of fuzzy sets with the corresponding attributes and relations;

[0033] Define fuzzy inference rules to perform reasoning and inference using knowledge represented by fuzzy sets.

[0034] Further: Step 5 comprises:

[0035] In the reasoning process, fuzzy logic reasoning is used to deal with uncertainty and ambiguity. The fuzzy reasoning formula is as follows:

[0036] R:A→B

[0037] μ R (x,y)=min[μ A (x),μ B (y)]

[0038] Where R is a fuzzy relation, A and B are fuzzy sets, μ R (x,y) is the degree of membership of the fuzzy relationship between x and y;

[0039] In the reasoning process, the membership of the fuzzy relationship between entities is calculated according to the membership function of the fuzzy relationship, so as to obtain the reasoning result.

[0040] Further: the calculation of the membership of the fuzzy relationship between entities according to the membership function of the fuzzy relationship, thereby obtaining the inference result includes:

[0041] For two fuzzy sets A and B, their membership functions are μ A (x) and μ B (y), the membership function indicates the degree to which element x belongs to set A and element y belongs to set B:

[0042] A={(x,μ A (x)|x∈X}

[0043] B={(y,μ B (y)|y∈Y}

[0044] The fuzzy relation R is the relation from fuzzy set A to fuzzy set B, and its membership function μ R(x,y) represents the degree of fuzziness of the relationship between elements x and y:

[0045] R={((x,y),μ R (x,y)|x∈X,y∈Y}

[0046] Membership function Membership function of fuzzy relationship μ R (x,y) is calculated by the following formula: For a given x and y, μ R (x,y) is represented as:

[0047]

[0048] Among them, Z is the set of intermediate fuzzy variables, representing all possible mediating elements, indicating that the membership between x and y is the maximum value of the minimum membership of all mediating elements z.

[0049] On the other hand, the present invention discloses a system for automatically constructing a domain ontology, comprising:

[0050] Data collection and preprocessing module: collects domain text data from multiple data sources and preprocesses it;

[0051] Knowledge extraction and representation module: extracts and represents knowledge from preprocessed data, including extracting entities, relations and attributes from domain text data, and establishing knowledge representation models;

[0052] Knowledge fusion and integration module: assign weights to each piece of knowledge based on its source and quality, integrate knowledge from different sources, and build a knowledge base;

[0053] Ontology modeling module: Ontology modeling is based on the integrated knowledge base and uses ontology modeling tools to build domain ontology;

[0054] Reasoning and inference module: Use the ontology reasoning engine to perform reasoning and inference; during the reasoning process, use fuzzy logic reasoning to deal with uncertainty and ambiguity.

[0055] Compared with the prior art, the present invention has the following technical advances:

[0056] The present invention adopts fuzzy set theory to represent and process uncertain knowledge, and takes uncertainty and fuzziness into consideration in the process of knowledge extraction, ontology modeling and reasoning, so as to better cope with the uncertainty and fuzziness of knowledge in the real world. The present invention comprehensively utilizes a variety of technical means such as natural language processing technology, knowledge graph technology and fuzzy logic reasoning method, and constructs a complete domain ontology through steps such as data collection, knowledge extraction, knowledge fusion and reasoning, so that the ontology has rich knowledge representation and reasoning capabilities. The present invention can improve the ontology's ability to express complex knowledge and more accurately reflect the complex relationships and semantics in the real world by embedding the knowledge represented by fuzzy sets into the ontology and defining concepts such as fuzzy attributes and fuzzy relationships. The present invention uses an ontology reasoning engine for reasoning and inference, and combines fuzzy logic reasoning methods to deal with uncertainty and fuzziness, which can improve the intelligence and robustness of the system, so that the system can more flexibly handle complex domain knowledge and reasoning tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0058] In the attached picture:

[0059] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0060] The following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0061] Example 1

[0062] like Figure 1 As shown in the figure, the present invention discloses a method and system for automatically constructing a domain ontology, comprising the following steps:

[0063] Step 1: Collect domain text data from multiple data sources and preprocess them.

[0064] 1.1. Data collection: Use web crawler technology to collect domain text-related data from multiple data sources, which can include professional websites, academic papers, news reports, social media, etc.

[0065] 1.2. Data cleaning: The collected data often contains a lot of noise and irregular information. In the data cleaning stage, the data needs to be cleaned and filtered to ensure the quality and accuracy of the data. The cleaning steps include:

[0066] Text denoising: remove noise information such as HTML tags, special characters, and garbled characters.

[0067] Text standardization: Convert text into a unified format, including case conversion, word form restoration, spelling correction, etc.

[0068] Entity recognition: Use NLP technology to perform entity recognition on text and identify domain entities in the text, such as names of people, places, and institutions.

[0069] Relationship extraction: Use NLP technology to extract the relationship between entities from the text, such as the cooperative relationship between people, the association relationship between places, etc.

[0070] 1.3. Data format standardization: During the preprocessing process, the data also needs to be converted into a unified format to facilitate subsequent processing and analysis. This includes converting text data into structured data formats such as JSON, XML, etc., and defining unified entity and relationship types.

[0071] In summary, in the data collection and preprocessing stage, NLP technology and data cleaning technology are used to collect and preprocess domain-related data from multiple data sources. Through cleaning and standardization, the quality and accuracy of the data can be improved, laying the foundation for subsequent knowledge extraction and ontology construction.

[0072] Step 2: Extract and represent knowledge from the preprocessed data, including extracting entities, relations, and attributes from domain text data, and establishing a knowledge representation model.

[0073] 2.1. Entity recognition and named entity recognition (NER): Perform entity recognition on the preprocessed text to identify entities in the text, including people, places, organizations, etc.

[0074] 2.2. Relationship extraction: After identifying the entities, NLP technology is used to extract the relationships between the entities. In this embodiment, this can be achieved through semantic analysis and grammatical analysis in the text.

[0075] 2.3. Attribute extraction: In addition to the relationship between entities, the attribute information of the entities can also be extracted.

[0076] 2.4. Knowledge representation: The extracted entities, relations and attributes are represented as a knowledge representation model. In the representation process, uncertainty and fuzziness are taken into consideration, and fuzzy set theory is used to represent fuzzy or uncertain knowledge. Fuzzy sets are represented as:

[0077] A={(x,μ A (x)|x∈X}

[0078] Where A is a fuzzy set, X is the domain, μ A (x) is the membership function of x in the fuzzy set A. For example, the extracted entity attributes can be represented as fuzzy sets, where the membership function represents the degree of certainty or credibility of the attribute.

[0079] Through the above steps, knowledge is extracted and represented on the preprocessed data, and uncertainty and ambiguity are handled through fuzzy set theory to establish a representation model of domain knowledge.

[0080] Step 3: Assign weights to each piece of knowledge based on its source and quality, integrate knowledge from different sources, and establish a knowledge base.

[0081] In this step, the knowledge from different data sources and knowledge extraction technologies is fused and integrated to establish a comprehensive knowledge base. During the integration process, it is necessary to consider the weight and credibility of knowledge from different sources, as well as the methods for dealing with uncertainty and ambiguity.

[0082] 3.1. Construct a knowledge graph: The entities, relationships, and attributes obtained through preprocessing and knowledge extraction are constructed into a knowledge graph. The nodes in the knowledge graph represent entities, and the edges represent the relationships between entities. The knowledge graph provides a structured representation to facilitate knowledge query and reasoning.

[0083] 3.2. Knowledge fusion: Fusion of knowledge from different data sources and knowledge extraction technologies. The fusion process includes the following aspects:

[0084] Consistency check: Check and eliminate conflicts and duplications between knowledge from different sources. For example, if different data sources have different descriptions of the same entity, these descriptions need to be consistent.

[0085] Weighted average: Considering the weight and credibility of knowledge from different sources, a weight is assigned to each piece of knowledge. The fusion formula is as follows:

[0086]

[0087] Where F is the fused knowledge, K i is the knowledge from the ith source, ω i is the weight of the ith source, and n is the number of knowledge sources.

[0088] 3.3. Dealing with uncertainty and fuzziness: In the process of knowledge fusion, uncertainty and fuzziness are dealt with by using fuzzy set theory to represent uncertain knowledge. Fuzzy sets are represented as:

[0089] A={(x,μ A (x)|x∈X}

[0090] Where A is a fuzzy set, X is the domain, μ A (x) is the membership function of x in the fuzzy set A. For each fuzzy attribute or relationship, a corresponding membership function is defined to represent its degree of fuzziness.

[0091] 3.4. Knowledge base integration: Integrate the fused knowledge into a unified knowledge base. In this embodiment, the knowledge base can be an RDF-based triple storage system or a graph database such as Neo4j.

[0092] Through the above steps, a comprehensive knowledge base can be established to effectively fuse and integrate knowledge from different sources and deal with uncertainty and ambiguity. The knowledge base not only provides a rich data foundation for ontology modeling, but also provides the necessary knowledge support for subsequent reasoning and inference.

[0093] Step 4: Ontology modeling Based on the integrated knowledge base, use ontology modeling tools to build domain ontology.

[0094] 4.1. Importing knowledge base: First, import the integrated knowledge base into the ontology modeling tool as the basis for ontology construction. These knowledge will serve as entities, relationships and attributes of the ontology.

[0095] 4.2. Define classes and attributes: In the ontology modeling tool, define the classes and attributes in the domain. For fuzzy attributes, they can be represented by defining the fuzziness degree and the membership function of the fuzzy set. For example, you can define a fuzzy attribute "temperature" and represent the fuzziness degree and membership function of temperature through fuzzy sets.

[0096] 4.3. Define relationships: In the ontology, it is also necessary to define the relationships between entities. For fuzzy relationships, the uncertainty and ambiguity of the relationship can be represented by defining fuzzy sets. For example, a fuzzy relationship "similarity" can be defined, and the uncertainty and membership of the similarity between entities can be represented by fuzzy sets.

[0097] 4.4. Embedding fuzzy set representation: In the ontology modeling process, the knowledge represented by fuzzy sets is embedded into the ontology, including embedding the definitions of fuzzy attributes and fuzzy relations into the ontology, and associating information such as the membership function of the fuzzy set with the corresponding attributes and relations.

[0098] 4.5. Define fuzzy reasoning rules: In the ontology, fuzzy reasoning rules can also be defined to use the knowledge represented by fuzzy sets for reasoning and inference. The reasoning rules can be defined according to the needs of specific fields to achieve the handling of fuzziness and uncertainty.

[0099] Through the above steps, the ontology modeling tool can be used to build the domain ontology, and the knowledge represented by fuzzy sets can be embedded into the ontology. In the ontology modeling process, fuzzy attributes and fuzzy relationships can be defined, and fuzzy reasoning rules can be used for reasoning and inference to achieve the processing of uncertainty and ambiguity.

[0100] Step 5: Use the ontology reasoning engine to perform reasoning and inference. During the reasoning process, the membership of the fuzzy relationship between entities is calculated according to the membership function of the fuzzy relationship, thereby obtaining the reasoning result.

[0101] 5.1. Using ontology reasoning engine: First, use ontology reasoning engine (such as Pellet, HermiT, etc.) to reason and infer the constructed ontology. The ontology reasoning engine can reason and infer based on the concepts, relationships and rules defined in the ontology, so as to obtain new knowledge or reasoning results.

[0102] 5.2. Fuzzy logic reasoning: In the reasoning process, consider using fuzzy logic reasoning methods to deal with uncertainty and ambiguity. Fuzzy logic reasoning is based on fuzzy set theory and can handle fuzzy or uncertain knowledge and improve the system's ability to handle uncertainty and ambiguity.

[0103] 5.3. Fuzzy reasoning formula: The fuzzy reasoning formula is as follows:

[0104] R:A→B

[0105] μ R (x,y)=min[μ A (x),μ B (y)]

[0106] Where R is a fuzzy relation, A and B are fuzzy sets, μ R (x,y) is the degree of membership of the fuzzy relationship between x and y.

[0107] In the reasoning process, the membership of the fuzzy relationship between entities is calculated according to the membership function of the fuzzy relationship, so as to obtain the reasoning result.

[0108] For two fuzzy sets A and B, their membership functions are μ A (x) and μ B (y), the membership function indicates the degree to which element x belongs to set A and element y belongs to set B:

[0109] A={(x,μ A (x)|x∈X}

[0110] B={(y,μ B (y)|y∈Y}

[0111] 5.4. Define fuzzy relation R: Fuzzy relation R is the relation from fuzzy set A to fuzzy set B, and its membership function μ R (x,y) represents the degree of fuzziness of the relationship between elements x and y.

[0112] R={((x,y),μ R (x,y)|x∈X,y∈Y}

[0113] 5.5. Calculate the membership function of the fuzzy relationship:

[0114] Membership function μ of fuzzy relation R (x,y) is calculated by the following formula: For a given x and y, μ R (x,y) can be expressed as:

[0115]

[0116] Among them, Z is the set of intermediate fuzzy variables, representing all possible mediating elements, indicating that the membership between x and y is the maximum value of the minimum membership of all mediating elements z.

[0117] 5.6. Calculation of reasoning results: In the fuzzy reasoning process, the membership function of the fuzzy relationship is used to calculate the reasoning results. Assume that there is a fuzzy rule R:A→B, where μ R (x, y) represents the membership function of the rule, and μ R (x, y) to obtain the fuzzy membership of y.

[0118] If the membership degree μ of x is known A (x), the membership degree μ of y can be calculated by fuzzy reasoning B (y):

[0119]

[0120] This formula means that for all possible x, we calculate μ A (x) and μ R The minimum value of (x, y) is then taken as the maximum value of these minimum values. B (y).

[0121] For example, consider the following fuzzy sets and fuzzy relations:

[0122] A={(x1,0.7),(x2,0.4)}

[0123] B={(x2,0.6),(y2,0.8)}

[0124] R={((x1,y1),0.5),((x1,y2),0.9,((x2,y1),0.3),((x2,y2),0.7)}

[0125] According to the calculation and reasoning results:

[0126] μ B (y1)=max[min(0.7,0.5),min(0.4,0.3)]=max[0.5,0.3]=0.5

[0127] μ B (y2)=max[min(0.7,0.9),min(0.4,0.7)]=max[0.7,0.4]=0.7

[0128] Finally, the fuzzy membership of y1 and y2 are 0.5 and 0.7 respectively.

[0129] Through the above calculation steps, using the fuzzy logic reasoning method, the membership of the fuzzy relationship between entities is calculated through the membership function of the fuzzy relationship, and finally the reasoning result is obtained.

[0130] Example 2

[0131] A domain ontology automatic construction system, comprising:

[0132] Data collection and preprocessing module: collects domain text data from multiple data sources and preprocesses it;

[0133] Knowledge extraction and representation module: extracts and represents knowledge from preprocessed data, including extracting entities, relations and attributes from domain text data, and establishing knowledge representation models;

[0134] Knowledge fusion and integration module: assign weights to each piece of knowledge based on its source and quality, integrate knowledge from different sources, and build a knowledge base;

[0135] Ontology modeling module: Ontology modeling is based on the integrated knowledge base and uses ontology modeling tools to build domain ontology;

[0136] Reasoning and inference module: Use the ontology reasoning engine to perform reasoning and inference; during the reasoning process, use fuzzy logic reasoning to deal with uncertainty and ambiguity.

[0137] Each module in Example 2 is used to implement the corresponding function in the example.

[0138] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of protection of the claims of the present invention.

Claims

1. A method for automatically constructing a domain ontology, characterized in that: The steps include: Step 1: Collect domain text data from multiple data sources and preprocess them; Step 2: Extract and represent knowledge from the preprocessed data, including extracting entities, relations, and attributes from domain text data, and establishing a knowledge representation model; Step 3: Assign weights to each piece of knowledge based on its source and quality, integrate knowledge from different sources, and build a knowledge base; Step 4: Ontology modeling Based on the integrated knowledge base, use ontology modeling tools to build domain ontology, including: 4.

1. Importing knowledge base: First, import the integrated knowledge base into the ontology modeling tool as the basis for ontology construction; 4.

2. Define classes and attributes: In the ontology modeling tool, define the classes and attributes in the domain. For fuzzy attributes, define the fuzziness degree and the membership function of the fuzzy set to represent them, and use the fuzzy set to represent the fuzziness degree and membership function of temperature. 4.

3. Define relationships: In the ontology, define the relationships between entities. For fuzzy relationships, define fuzzy sets to represent the uncertainty and ambiguity of the relationship, and use fuzzy sets to represent the uncertainty and membership of the similarity between entities. 4.

4. Embedding fuzzy set representation: In the ontology modeling process, the knowledge represented by fuzzy sets is embedded into the ontology, including embedding the definitions of fuzzy attributes and fuzzy relations into the ontology, and associating the membership function information of fuzzy sets with the corresponding attributes and relations; 4.

5. Define fuzzy reasoning rules: In the ontology, define fuzzy reasoning rules to use the knowledge represented by fuzzy sets for reasoning and inference; Step 5: Use the ontology reasoning engine to perform reasoning and inference. During the reasoning process, the membership of the fuzzy relationship between entities is calculated according to the membership function of the fuzzy relationship, thereby obtaining the reasoning result.

2. The method for automatically constructing a domain ontology according to claim 1, characterized in that: The step 1 comprises: Clean and filter the data to ensure its quality and accuracy. The cleaning steps include: Text denoising: remove HTML tags, special characters, and garbled noise information; Text standardization: converting text into a unified format, including case conversion, word form restoration, and spelling correction; Entity recognition: Use NLP technology to perform entity recognition on text and identify domain entities in the text, including names of people, places, and institutions; Relationship extraction: Use NLP technology to extract the relationships between entities from the text, including cooperative relationships between people and associations between locations.

3. The method for automatically constructing a domain ontology according to claim 2, characterized in that: The step 2 comprises: Perform entity recognition on the preprocessed domain text to identify the entities in the text; After identifying the entities, extract the relationships between the entities and the attribute information of the entities; The extracted entities, relations and attributes are represented as a knowledge representation model. In the representation process, fuzzy sets are used to represent fuzzy or uncertain knowledge. Fuzzy sets are represented as: A={(x,μ A (x)|x∈X} Where A is a fuzzy set, X is the domain, μ A (x) is the membership function of x in the fuzzy set A.

4. The method for automatically constructing a domain ontology according to claim 3, characterized in that: The step 3 comprises: The entities, relations and attributes obtained through preprocessing and knowledge extraction are constructed into a knowledge graph, where the nodes of the knowledge graph represent entities and the edges represent the relations between entities. According to the source and quality of knowledge, each piece of knowledge is given a weight, and the knowledge from different sources is integrated using the following methods: Where F is the fused knowledge, K i is the knowledge from the ith source, ω i is the weight of the ith source, n is the number of knowledge sources; Integrate the fused knowledge into a unified knowledge base.

5. The method for automatically constructing a domain ontology according to claim 4, characterized in that: The step 5 comprises: In the reasoning process, fuzzy logic reasoning is used to deal with uncertainty and ambiguity. The fuzzy reasoning formula is as follows: R:A→B m R (x,y)=min[μ A (x),μ B (y)] Where R is a fuzzy relation, A and B are fuzzy sets, μ R (x,y) is the degree of membership of the fuzzy relationship between x and y; In the reasoning process, the membership of the fuzzy relationship between entities is calculated according to the membership function of the fuzzy relationship, so as to obtain the reasoning result.

6. The method for automatically constructing a domain ontology according to claim 5, characterized in that: The calculation of the membership of the fuzzy relationship between entities according to the membership function of the fuzzy relationship, thereby obtaining the inference result includes: For two fuzzy sets A and B, their membership functions are μ A (x) and μ B (y), the membership function indicates the degree to which element x belongs to set A and element y belongs to set B: A={(x,μ A (x)|x∈X} B={(y,μ B (y)|y∈Y} The fuzzy relation R is the relation from fuzzy set A to fuzzy set B, and its membership function μ R (x,y) represents the degree of fuzziness of the relationship between elements x and y: R={((x,y),μ R (x,y)|x∈X,y∈Y} Membership function Membership function of fuzzy relationship μ R (x,y) is calculated by the following formula: For a given x and y, μ R (x,y) is represented as: Among them, Z is the set of intermediate fuzzy variables, representing the mediating elements, indicating that the membership between x and y is the maximum value of the minimum membership of all mediating elements z.

7. A domain ontology automatic construction system, characterized in that: include: Data collection and preprocessing module: collects domain text data from multiple data sources and preprocesses it; Knowledge extraction and representation module: extracts and represents knowledge from preprocessed data, including extracting entities, relations and attributes from domain text data, and establishing knowledge representation models; Knowledge fusion and integration module: assign weights to each piece of knowledge based on its source and quality, integrate knowledge from different sources, and build a knowledge base; Ontology modeling module: Ontology modeling is based on the integrated knowledge base and uses ontology modeling tools to build domain ontology, including: Importing knowledge base: First, import the integrated knowledge base into the ontology modeling tool as the basis for ontology construction; Define classes and attributes: In the ontology modeling tool, define the classes and attributes in the domain. For fuzzy attributes, define the fuzzy degree and the membership function of the fuzzy set to represent them, and use the fuzzy set to represent the fuzzy degree and membership function of the temperature. Define relationships: In the ontology, define the relationships between entities. For fuzzy relationships, define fuzzy sets to represent the uncertainty and ambiguity of the relationships, and use fuzzy sets to represent the uncertainty and membership of the similarity between entities. Embedding fuzzy set representation: In the ontology modeling process, the knowledge represented by fuzzy sets is embedded into the ontology, including embedding the definitions of fuzzy attributes and fuzzy relations into the ontology, and associating the membership function information of fuzzy sets with the corresponding attributes and relations; Define fuzzy reasoning rules: In the ontology, define fuzzy reasoning rules to use the knowledge represented by fuzzy sets for reasoning and inference; Reasoning and inference module: Use the ontology reasoning engine to perform reasoning and inference; during the reasoning process, use fuzzy logic reasoning to deal with uncertainty and ambiguity.

Images