Knowledge graph-based description and practice method capable of reasoning accurate semantics

Abstract

The invention discloses a description and practice method capable of reasoning accurate semanteme based on a knowledge graph, and relates to the technical field of intelligent education. According to the description and practice method, a cell description language for the mathematics subject is defined to describe topic semanteme, a set of complete concept, individual and relation description framework is defined, and common binary relations are included; according to the method, a multi-element relation is obtained and expanded to description of the multi-element relation, meanwhile, for the multi-element relation, a corresponding Atom expansion rule is defined, all atomic information is obtained from simple relation description, a basis is provided for knowledge reasoning, diversity and ambiguity of natural languages are eliminated, definition with uniqueness and accuracy is given, and the method has the advantages of being high in practicability and easy to popularize. In addition, aiming at the problem of non-uniqueness of semantic information, standardization rules of a cell language are defined, ambiguity is eliminated, the problem that an existing method cannot accurately describe topic semantics in an unambiguous mode is solved, on the basis, intelligent solution of topics can be achieved through an inference engine, and the teaching cost and the tutoring cost of teachers are greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent education technology, and in particular to a description and practice method of inferable precise semantics based on knowledge graphs. Background Art

[0002] In the field of education, question bank search technology is one of the most commonly used tools, designed to help users quickly find questions related to learning, exams or homework grading. However, the limitation of this type of technology is that when there are no matching questions in the question bank, it can only provide references to similar questions, but cannot deeply understand the semantics of the questions or complete reasoning. To achieve truly intelligent problem-solving capabilities, the system needs to be able to understand the internal logic of the questions and make inferences like a teacher. This has become a key research direction for artificial intelligence in the field of education.

[0003] At present, reasoning technologies based on artificial intelligence and knowledge representation mainly include expert systems and knowledge graphs. Expert systems simulate the knowledge and decision-making process of domain experts and use pre-written rules to solve complex problems in specific fields. They are widely used in medical diagnosis, financial analysis, and information retrieval. Knowledge graphs are a graphical knowledge organization structure that can represent entities and their relationships, based on which reasoning can be performed. They play an important role in scenarios such as information retrieval, recommendation systems, and data analysis. However, these technologies still have certain limitations when faced with complex, cross-domain, or semantically ambiguous reasoning tasks.

[0004] Although expert systems and knowledge graphs have achieved many application results, they still face many problems in the reasoning process: expert systems are complex to construct, rely on the collection and organization of a large amount of domain knowledge, and are difficult to adapt to new problems or cross-domain reasoning needs. Although knowledge graphs have powerful knowledge representation capabilities, their update and maintenance costs are high, and they are often incomplete and difficult to support certain reasoning tasks. In addition, although the SWRL rule language can handle complex logic and conditional reasoning, it has a strong dependence on rules and performs poorly when faced with fuzzy or uncertain information, which can easily lead to inconsistent reasoning results. Therefore, there is an urgent need for a description and practical method of reasonable precise semantics based on knowledge graphs to solve such problems. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a description and practice method of reasonable precise semantics based on knowledge graph to solve the problems in existing solutions, such as complex expert system construction, reliance on the collection and organization of a large amount of domain knowledge, and difficulty in adapting to new problems or cross-domain reasoning needs. Although knowledge graphs have powerful knowledge representation capabilities, their update and maintenance costs are high, and they are often incomplete and difficult to support certain reasoning tasks.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] The present invention provides a description and practice method of inferable precise semantics based on knowledge graph, which includes:

[0009] Step S1, define a unified Cell language description. In order to achieve semantic reasoning, the present invention proposes a special description language - Cell language. Cell language is used to convert the natural language information in the question into a structured expression that can be processed by a computer. Cell language is defined to describe complex semantic relationships.

[0010] Step S2, expanding the ternary relationship into a multi-tuple relationship. Traditional knowledge representation methods usually adopt the form of triples (subject, predicate, object), which is difficult to express the complex semantics in mathematical problems. The present invention expands the triples, allows the relationship between multiple individuals, and defines a multi-tuple relationship structure;

[0011] Step S3, expanding the reasoning information, based on the multivariate relationship in step S2, expanding all relevant individual and attribute information, assuming that the relationship "line segments EF and AC intersect at point N" is expanded, expanding the relevant individual information of angles, points and lines;

[0012] Step S4, defining individual data attributes and binary relationships, and for each individual, further defining its attributes and relationships with other individuals;

[0013] Step S5, defining a unified naming rule. To ensure the accuracy of reasoning, the present invention introduces a standardized operation to eliminate ambiguity. By defining a unified naming rule, the expression of all geometric elements is unique and unambiguous:

[0014] Step S6, processing dominant angles and inferior angles. The angles in math problems are usually expressed in two forms: dominant angles and inferior angles. To ensure consistency in the reasoning process, the present invention normalizes the angles and uses the same form to describe them.

[0015] Step S7, correcting the intersection relationship of line segments. For the description of the intersection of line segments, the order of the starting edge and the terminal edge of the line segments in the description is corrected. For example, when describing "line segments AB and BC intersect at point B", it is necessary to ensure that AB is the starting edge and BC is the terminal edge. If it cannot be directly determined through the graph, the preset rules are used to correct such relationships to ensure the uniqueness of the expression in the reasoning process.

[0016] Step S8, reasoning by the inference engine. Steps S1 to S7 are semantic description steps, i.e., normalized cell language and relationships. After completing the semantic description, the inference engine performs reasoning (practice) based on the above-mentioned normalized cell language and relationships. The inference engine parses the above-mentioned relationships and uses SWRL rules to infer new conclusions. For example, for geometric problems, specific results such as angle size, line segment length, intersection point position, etc. can be calculated, and finally the automatic solution of math problems is realized.

[0017] Furthermore, in step S1, the constituent elements of the defined Cell language include:

[0018] concept, which refers to abstract concepts in mathematics, such as "triangle" and "angle",

[0019] ind, represents a specific individual, such as "point A", "line segment AB", etc.

[0020] data, which refers to data or attributes related to an individual, such as "length" or "size".

[0021] conceptind, which indicates the relationship between a concept and an individual, such as "triangle ABC contains point A",

[0022] indind, which indicates the relationship between two entities, such as "line segment AB is perpendicular to line segment BC",

[0023] inddata, represents the data attributes of an individual, such as "the length of line segment AB is 5".

[0024] Furthermore, in step S2, the multi-relation structure defined is:

[0025] indind#Relationship description[ind 1 ,ind 2 ,…,ind n ], where n represents the number of individuals.

[0026] Furthermore, in step S2, the defined multi-relation structure includes the following multi-relation interfaces, which are used to express common relations in geometry:

[0027] Two line segments intersect at a point:

[0028] indind#SeglineCrossSeglineAtPoint[ind 1 ,ind 2 ,ind 3 ],

[0029] Two line segments are perpendicular at a point:

[0030] indind#SeglineVerticalSeglineAtPoint[ind 1 ,ind 2 ,ind 3 ],

[0031] Altitude on a side of a triangle:

[0032] indind#TriangleHasEdgeHighline[ind 1 ,ind 2 ,ind 3 ],

[0033] The median of a side of a triangle:

[0034] indind#TriangleHasEdgeMiddleline[ind 1 ,ind 2 ,ind 3 ],

[0035] Angle bisector of an angle of a triangle:

[0036] indind#TriangleHasAngleMiddleline[ind 1 ,ind 2 ,ind 3 ].

[0037] Furthermore, in step S3, the individual information of the expanded corners, points and lines is:

[0038] Individual information of angle: conceptind#conceptHasInd[Angle,Angle@FNC],

[0039] Point information: conceptind#conceptHasInd[Point,N],

[0040] Line information: conceptind#conceptHasInd[Segline,EN].

[0041] Furthermore, in step S4, the data attributes and binary relationship of individuals are defined as follows:

[0042] The inddata relationship describes the data attributes of an individual, including the length of a line segment and the size of an angle:

[0043] inddata#hasDefaultLength(EN,EN@Length),

[0044] inddata#hasSize(Angle@FNC,FNC@Size).

[0045] Furthermore, in step S4, defining the data attributes and binary relationship of individuals also includes:

[0046] The indind relationship is used to describe the relationship between two individuals, including the relationship between line segments and points and angles:

[0047] indind#seglineHasPoint([AC,N]),

[0048] indind#angleEqualAngle([Angle@CNE,Angle@ANF]).

[0049] Furthermore, in step S5, the naming rules include:

[0050] Line segment naming rules: Line segments are named in alphabetical order. For example, line segments AB and BA are uniformly named AB to avoid ambiguity caused by different writing orders.

[0051] Furthermore, in step S5, the naming rule also includes:

[0052] Angle naming rules: Angles are named based on the clockwise or counterclockwise order from the starting side to the terminal side. For example, Angle@ABC represents the angle formed by rotating counterclockwise from ray AB to ray BC. Regardless of whether the question involves superior or inferior angles, they are all expressed in counterclockwise form.

[0053] Furthermore, in step S6, the method of normalizing the angle includes:

[0054] A superior angle is defined as an angle less than 180 degrees, and a inferior angle is defined as an angle greater than 180 degrees but less than 360 degrees.

[0055] When angles within the same definition have different names (such as Angle@ABC and Angle@CBA), they are all described in the same form.

[0056] The beneficial effects of the present invention are:

[0057] The present invention eliminates the diversity and ambiguity of natural language, provides a unique and accurate definition, and performs precise semantic understanding and reasoning on this basis to achieve understanding and automatic solution of the question.

[0058] The present invention defines a cell description language for mathematics to describe the semantics of questions, and defines a complete set of concepts, individuals, and relationship description frameworks, which not only include common binary relationships, but also extend to the description of multi-variable relationships. At the same time, for multi-variable relationships, corresponding Atom extension rules are defined to obtain all atomic information from simple relationship descriptions, providing a basis for knowledge reasoning.

[0059] The present invention aims at the problem of non-uniqueness of semantic information, defines the normalization rules of cell language and eliminates ambiguity.

[0060] In summary, the present invention establishes semantic description rules for mathematics questions, which are more comprehensive, more rigorous in description logic, and more complex in extracted semantics compared to existing semantic extraction methods. At the same time, the present invention further standardizes the definition rules of basic individuals and relationships, and solves the problem that existing methods cannot accurately and unambiguously describe the semantics of questions. On this basis, the present invention can realize intelligent solving of questions through an inference engine, greatly reducing the teaching and tutoring costs of teachers. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0062] Figure 1 It is a flowchart of the description and practice method of the inferable precise semantics based on knowledge graph of the present invention. DETAILED DESCRIPTION

[0063] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0064] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0065] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0066] SWRL is a rule language used to define and execute rules. SWRL has rich logical expression capabilities and can represent various rules and logical conditions, including numerical comparison, string operations, etc. At the same time, SWRL rules can be executed by the reasoning engine to infer new facts, relationships or conclusions. The main purpose of SWRL is to support logical reasoning so as to infer new information or conclusions from knowledge. The performance and effectiveness of SWRL reasoning are highly dependent on the integrity of the rules. If there are missing or wrong conditions, wrong reasoning conclusions will appear. On the other hand, SWRL rules are usually built based on precise logical conditions and are not very suitable for processing uncertainty or fuzzy information. For ambiguous rules, inconsistent reasoning results will appear. For this reason, the following improvement scheme is proposed:

[0067] Example 1, reference Figure 1 This embodiment is aimed at mathematics and aims to design a method that can accurately describe the complex logical semantics in mathematics problems. Although SWRL supports complex logical reasoning, it needs to express all knowledge in the form of unambiguous Atom atoms.

[0068] The present invention mainly solves two problems: converting the text information in the title into a unified cell language that can be processed by a computer (the present invention defines the proposed semantic description as a cell); and converting the cell language into an unambiguous and precise Atom atomic expression.

[0069] The present invention designs a series of cell definitions for describing complex semantic relationships. The present invention expands the representation of triples so that cells can describe multi-dimensional relationships. Concept represents concept, ind represents individual, data represents data, conceptind represents the relationship between concept and individual, indind represents the relationship between individuals, and inddata represents the data attributes of individuals. In conventional representation methods, the relationship between individuals and the data attributes of individuals are usually represented as binary relationships. Based on the relevant knowledge of mathematics, the present invention defines multi-dimensional relationships based on the original binary relationships of SWRL. The general form of describing the relationship between n individuals can be expressed as:

[0070] indind#relation description [ind1, ind2, ..., indn]

[0071] Since binary relations are difficult to describe all common topic semantics, although polynomial relations can be split into binary relations, the definition of polynomial relations is crucial from the perspective of simplifying engineering processes and maintainability. The present invention defines a series of corresponding cell interfaces based on polynomial relations between mathematical geometric individuals, for example:

[0072] indind#SeglineCrossSeglineAtPoint:[ind1,ind2,ind3]

[0073] indind#SeglineVerticalSeglineAtPoint:[ind1,ind2,ind3]

[0074] indind#TriangleHasEdgeHighline:[ind1,ind2,ind3]

[0075] indind#TriangleHasEdgeMiddleline:[ind1,ind2,ind3]

[0076] indind#TriangleHasAngleMiddleline:[ind1,ind2,ind3]

[0077] They respectively represent the meanings of "two line segments intersect at a certain point", "two line segments are perpendicular to a certain point", "a triangle has an altitude on a certain side", "a triangle has a median on a certain side", and "a triangle has an angle bisector of a certain angle".

[0078] For ternary relations, we have implemented the expansion of relations based on specific definitions, that is, expanding all indind or inddata contained in the multi-relation to provide the most complete reasoning information input. Taking SeglineCrossSeglineAtPoint[EF,AC,N] as an example, according to the intersection of two line segments EF and AC at point N, the present invention can expand the information of all lines, points, and angles. For example, the specific individual information of all concepts such as angle Angle, point Point, and line Segline can be expanded, for example:

[0079] conceptind#conceptHasInd[Angle,Angle@FNC],

[0080] conceptind#conceptHasInd[Point,N],

[0081] conceptind#conceptHasInd[Segline,EN]

[0082] Data attributes of all individuals, such as:

[0083] inddata#hasDefaultLength(EN,EN@Length),

[0084] inddata#hasSize(Angle@FNC,FNC@Size),

[0085] All individual binary relations, such as:

[0086] indind#seglineHasPoint([AC,N]),

[0087] indind#angleEqualAngle([Angle@CNE,Angle@ANF]),

[0088] In general, in the design of the cell language, compared with SWRL, the present invention defines new relationships in some concepts, individuals, attributes, etc. based on the characteristics of mathematics, defines multi-dimensional relationships based on complex semantics, and performs Atom-level information expansion for multi-dimensional relationships.

[0089] In order to conduct deeper reasoning and achieve the goal of intelligent problem solving, after describing natural language as cell language, it is necessary to further eliminate ambiguity and conflict in the cell description so that the correct conclusion can be inferred through the reasoning engine. For some questions, it is necessary to combine the text information of the question stem and the reference information of the graphic to determine the unique semantic information. For example, triangle ABC and triangle BAC represent the same triangle, and angle ABC and angle CBA represent the same angle. However, for computers, it is impossible to think and judge like humans, and existing reasoning engines cannot perceive image information. Therefore, making the description of the cell unique is a prerequisite for accurate computer reasoning.

[0090] The present invention designs semantic naming rules to perform normalization operations for semantic descriptions. Specifically, the present invention makes constraints in the most basic individual definitions, such as the naming of line segments and angles. For line segments, the line segment names are uniformly processed in alphabetical order, for example, AB and BA are uniformly named AB.

[0091] Angles are different from line segments and have their own particularity. The particularity of angles lies mainly in the fact that angles themselves can be divided into dominant angles and minor angles, that is, Angle@ABC may represent the dominant angle formed by the line segment ray AB and the ray BC, or it may refer to the minor angle formed by the two rays. In the actual process, whether Angle@ABC described in a question represents a dominant angle or a minor angle is usually judged by people in combination with the actual situation. In addition, Angle@ABC and Angle@CBA usually represent the same angle in actual situations, which also means that the description of the same angle is not unique. The computer's judgment on dominant angles and minor angles or the judgment of different names for an angle can only be based on clear rules. Therefore, the clear and unambiguous description of angles is more complicated. The present invention makes the following definition: In Angle@ABC, AB is the starting side and BC is the ending side. At this time, Angle@ABC represents the angle formed by rotating ray AB counterclockwise to BC. And Angle@CBA represents the angle formed by rotating ray BC counterclockwise to AB. Under this definition, if Angle@ABC is a minor angle, then Angle@CBA is its corresponding superior angle. After normalization, Angle@ABC and Angle@CBA are uniformly named as the minor angle represented by rays AB and BC, that is, uniformly corrected to Angle@ABC, in order to conform to the general habit of the question. This completely solves the problem of standardized description of angles and ensures the uniqueness of angle naming.

[0092] For the intersection relationship of line segments, such as indind#SeglineCrossSeglineAtPoint:[AB,BC,B], it is necessary to ensure that AB is the starting edge and BC is the ending edge according to the spatial position relationship of AB and BC in the figure, otherwise the parameter order of AB and BC needs to be corrected. If there is no figure to refer to, all line, angle, and intersection information will be corrected according to the default order rule.

[0093] In the present invention, Cell language is used as a way of knowledge representation, which is similar to the "entity-relationship-entity" structure, i.e., triples, in the knowledge graph. The main purpose of Cell language is to convert the semantics in natural language into unambiguous and structured expressions.

[0094] The design of multi-factor relationships is also an extension of the knowledge graph. Traditional knowledge graphs usually use triples to describe binary relationships between entities (such as entity A is associated with entity B). The present invention expands the knowledge representation capability in the form of multi-factor relationships to express complex logical relationships in mathematical problems. The extension of multi-factor relationships is regarded as a further development based on the knowledge graph for specific application scenarios.

[0095] At the same time, for IndInd and IndData in the text:

[0096] IndInd represents the relationship between two entities (individuals), such as "line segment AB is perpendicular to line segment BC", which is similar to the description of the relationship between entities in the knowledge graph, that is, the expression form of "predicate" in the knowledge graph.

[0097] IndData is used to describe the relationship between an individual and its attributes, such as "the length of line segment AB is 5". The association between individuals and data is the same as the typical structure in the knowledge graph, that is, the entity information in the knowledge graph is enriched through data attributes (such as length, size, etc.).

[0098] In the knowledge graph, reasoning is usually based on existing triple relationships. For example, new entity relationships can be discovered through deduction. The Atom expression of the present invention breaks down complex relationships into more basic units to support deeper reasoning, which can be understood as a mechanism for the knowledge graph reasoning engine to perform logical inference through triples.

[0099] The reasoning engine of the present invention performs reasoning on complex problems by using multi-dimensional relationships and attribute information represented in Cell language, and the process is similar to the working principle of the knowledge graph reasoning engine.

[0100] The Cell language, IndInd / IndData relationship, and reasoning process of the reasoning engine used in the present invention are similar to the core idea in the knowledge graph, that is, supporting semantic reasoning and knowledge expansion through the graphical representation of entities and relationships.

[0101] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A description and practice method of reasonable precise semantics based on knowledge graph, characterized by: include, Step S1, define a unified Cell language description, define the Cell language to describe complex semantic relationships, Step S2, expanding the ternary relationship into a multi-tuple relationship, expanding the triple, allowing the relationship between multiple individuals, and defining a multi-tuple relationship structure; Step S3, expanding the reasoning information, based on the multivariate relationship in step S2, expanding all relevant individual and attribute information, assuming that the relationship "line segments EF and AC intersect at point N" is expanded, and the relevant individual information of angles, points and lines is expanded; Step S4, defining individual data attributes and binary relationships, and for each individual, further defining its attributes and relationships with other individuals; Step S5, defining a unified naming rule; Step S6, processing dominant angles and inferior angles, normalizing the angles, and describing them in the same form; Step S7, correcting the intersection relationship of line segments. For the description of the intersection of line segments, the order of the starting edge and the ending edge of the line segment in the description is corrected; Step S8, use the reasoning engine to perform reasoning. Steps S1 to S7 are semantic description steps, that is, standardizing cell language and relationships. After completing the semantic description, the reasoning engine performs reasoning based on the above-mentioned standardized cell language and relationships. The reasoning engine parses the above-mentioned relationships and uses SWRL rules to infer new conclusions, and finally realizes the automatic solution of mathematical problems.

2. According to claim 1, a method for describing and practicing inference-based precise semantics based on knowledge graph, characterized in that: In step S1, the constituent elements of the defined Cell language include: concept, which refers to an abstract concept in mathematics. ind, which indicates a specific individual, data, which refers to data or attributes related to an individual, conceptind, which indicates the relationship between a concept and an individual, indind, which indicates the relationship between two individuals, inddata, represents the data attributes of an individual.

3. According to claim 2, a method for describing and practicing inference-based precise semantics based on knowledge graphs, characterized in that: In step S2, the multi-relation structure is defined as: indind#Relation description [ind1, ind2, ..., ind n ], where n represents the number of individuals.

4. According to claim 3, a method for describing and practicing inference-based precise semantics based on knowledge graphs, characterized in that: In step S2, the defined multi-relation structure includes the following multi-relation interfaces, which are used to express common relations in geometry: Two line segments intersect at a point: indind#SeglineCrossSeglineAtPoint[ind1,ind2,ind3], Two line segments are perpendicular at a point: indind#SeglineVerticalSeglineAtPoint[ind1,ind2,ind3], Altitude on a side of a triangle: indind#TriangleHasEdgeHighline[ind1,ind2,ind3], The median of a side of a triangle: indind#TriangleHasEdgeMiddleline[ind1,ind2,ind3], Angle bisector of an angle of a triangle: indind#TriangleHasAngleMiddleline[ind1,ind2,ind3].

5. According to claim 4, a method for describing and practicing inference-based precise semantics based on knowledge graph, characterized in that: In step S3, the individual information of the expanded corners, points and lines is: Individual information of the angle: conceptind#conceptHasInd[Angle,Angle@FNC], Point information: conceptind#conceptHasInd[Point,N], Line information: conceptind#conceptHasInd[Segline,EN].

6. The method for describing and practicing inference-based precise semantics based on knowledge graph according to claim 5, characterized in that: Step S4, define the individual data attributes and binary relationship as follows: The inddata relationship describes the data attributes of an individual, including the length of a line segment and the size of an angle: inddata#hasDefaultLength(EN,EN@Length), inddata#hasSize(Angle@FNC,FNC@Size).

7. The method for describing and practicing inference-based precise semantics based on knowledge graph according to claim 6, characterized in that: Step S4, defining individual data attributes and binary relationship methods also includes: The indind relationship is used to describe the relationship between two individuals, including the relationship between line segments and points and angles: indind#seglineHasPoint([AC,N]), indind#angleEqualAngle([Angle@CNE,Angle@ANF]).

8. The method for describing and practicing inference-based precise semantics based on knowledge graph according to claim 7, characterized in that: In step S5, the naming rules include: Line segment naming rules: Line segments are named uniformly in alphabetical order.

9. The method for describing and practicing inference-based precise semantics based on knowledge graph according to claim 8, characterized in that: In step S5, the naming rule also includes: Angle Naming Rules: Angles are named based on the clockwise or counterclockwise order from the starting edge to the ending edge.

10. The method for describing and practicing inference-based precise semantics based on knowledge graph according to claim 9, characterized in that: In step S6, the method of normalizing the angle includes: A superior angle is defined as an angle less than 180 degrees, and a inferior angle is defined as an angle greater than 180 degrees but less than 360 degrees. When angles within the same definition have different names, they are all described in the same form.

Images