CN1466046A - Ontology-Based Knowledge Graph System and Method - Google Patents

Abstract

A knowledge graphic system and method based on ontology, the system includes a data storage device and a processor; the data storage device comprises an ontology database with a plurality of concepts and associations among the concepts; the processor is coupled to the data storage device, so that the inclusion relations between the objects, the corresponding object names and the objects are extracted from a graphic file; for each object, inquiring a corresponding concept from the ontology database according to the object name of the object, if the corresponding concept does not exist, newly adding the concept of the corresponding object into the ontology database, designating the association between the concept and other concepts, and establishing an ontology concept association diagram corresponding to the graph file, corresponding content knowledge and the graph file in a knowledge graph library according to the inclusion relationship between the concept of the corresponding object and the object; the invention automatically updates the concepts and the association thereof in the ontology database and establishes an ontology concept association diagram of the corresponding graph file by identifying the graph file; in addition, the traditional graph is used for displaying an ontology-based knowledge body, so that a user can understand the ontology-based knowledge body conveniently.

Description

Ontology-Based Knowledge Graph System and Method

technical field

The present invention relates to computer software, especially a knowledge graph system and method based on ontology (Ontology), and especially relates to a concept (Concept) in the ontology database that can be automatically updated by identifying graphic files and its association and An ontology-based knowledge graph system and method for establishing an ontology concept correlation graph of a corresponding graph file.

Background technique

With the shaping of the trend of the electronic age, the continuous transmission of information between networks also forms the continuous accumulation of knowledge. For an organization or a specific field, the collection and application of knowledge is usually carried out by establishing a knowledge base system for knowledge management and experience sharing. How to effectively and truly manage knowledge is one of the most popular and important research topics at present. Since ontology can directly and structurally describe human knowledge, by establishing the relationship between different concepts, it can faithfully and clearly represent the knowledge structure of a specific domain.

FIG. 1 shows a schematic diagram of an ontology conceptual framework of the network communication industry. As shown in the figure, the network communication industry 100 may include two main concepts: the Internet 110 and the traditional telecommunication industry 120 . Internet 110 can include concepts such as dial-up 111; traditional telecommunication industry 120 can be divided into two concepts of fixed network 121 and wireless 122, and also have telecommunication services 123, equipment 124, microwave/satellite 125, and mobile phones 126 concept, and so on. Therefore, ontology can be suitable for efficiently and completely architecting knowledge base systems.

Traditional knowledge ontology, whether it is hierarchical (Hierarchical) or graphical (Graphical) representation, often has too many concepts or relationships, making it difficult for users to understand a body of knowledge or the relationship between knowledge and concepts. The ontology representation method based on context will select appropriate knowledge attributes and relationships according to the context, but it lacks an easy-to-understand representation method, and it is not easy for users to understand a body of knowledge or Relationships between knowledge and concepts. However, the traditional methods of graphically representing the body of knowledge, such as architecture diagrams or flow charts, cannot understand the similarities and differences between two graphs in the body of knowledge from two similar graphics.

In addition, the maintenance work of the ontology-based knowledge base system is extremely complex and cumbersome. Since new knowledge or information must be captured and corresponded to concepts, experts who are very familiar with this field are required to be competent in updating and maintaining concepts in the knowledge base.

Contents of the invention

In view of this, the main purpose of the present invention is to provide an ontology-based knowledge graph system and method that can automatically update the concepts and their associations in the ontology database by identifying the graph files and establish the ontology-concept relationship graph of the corresponding graph files .

In addition, another object of the present invention is to enable users to better understand the body of knowledge or the relationship between knowledge and concepts through the combination of graphics and background information of the corresponding graphics.

In order to achieve the above object of the present invention, it can be achieved by the ontology-based knowledge graph system and method provided by the present invention.

An ontology-based knowledge graph system according to an embodiment of the present invention includes a data storage device and a processor. The data storage device includes an ontology database with multiple concepts and associations between the concepts and a knowledge graph library.

The processor is coupled to the data storage device, so that a plurality of objects, corresponding object names and inclusion relationships among the objects are retrieved from a graphics file. Then, for each object, query the corresponding concept from the ontology database according to its object name, if the concept of the corresponding object does not exist, then add the concept of the corresponding object in the ontology database, and specify the relationship between the concept and other concepts in the ontology database association. Afterwards, according to the concept of the corresponding object and the inclusion relationship between the objects, the ontology-concept association graph corresponding to the graphic file, the corresponding content knowledge, and the graphic file are established in the knowledge graph database.

Wherein, the processor queries the corresponding concept from the ontology database according to the object name of the object, and performs query through concept keyword comparison or concept similarity comparison. In addition, the inclusion relationship between objects can be known according to the coordinates and area of each object in the graphics file.

In addition, according to the ontology-based knowledge graph method of the embodiment of the present invention, firstly, a plurality of objects, corresponding object names, and inclusion relationships among the objects are extracted from the graph file. Then, for each object, the corresponding concept is queried from the ontology database according to its object name. If the concept corresponding to the object does not exist, then add the concept corresponding to the object in the ontology database, and specify the association between the concept and other concepts in the ontology database.

Afterwards, according to the concepts corresponding to these objects and the inclusion relationship between objects, an ontology-concept association graph corresponding to this graphic file is established. And store the graphic file, the content knowledge of the corresponding graphic file, and the ontology concept association graph of the corresponding graphic file in the knowledge graphic library.

Similarly, the processor queries the corresponding concept from the ontology database according to the object name of the object through concept keyword comparison or concept similarity comparison. In addition, the inclusion relationship between objects can be known according to the coordinates and area of each object in the graphics file.

The beneficial effect of the present invention is that, through the ontology-based knowledge graph system and method proposed by the present invention, the concepts and their associations in the ontology database can be automatically updated by identifying the graph files, and the ontology concept association graph of the corresponding graph files can be established . In addition, in the present invention, a body of knowledge based on ontology is displayed in traditional graphics, which is convenient for users to understand. Each graph represents the content in the knowledge ontology, which makes it easier for users to understand the differences between different contents.

Description of drawings

In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and understandable, the following specific examples, together with the accompanying drawings, are described in detail as follows:

FIG. 1 is a schematic diagram showing the ontology conceptual framework of the network communication industry;

FIG. 2 is a schematic diagram showing the system architecture of an ontology-based knowledge graph system according to an embodiment of the present invention;

Fig. 3 is a flowchart showing the operation process of the ontology-based knowledge graph method according to an embodiment of the present invention;

Figure 4 is an example of a graphics file;

FIG. 5 is an ontology concept correlation diagram obtained after the graphic file in FIG. 4 is processed by the present invention.

Symbol Description:

100, 110, ..., 126 concepts;

200 data storage devices;

201 ontology database;

202 knowledge database;

210 processors;

220 graphic files;

S300, S310, ..., S350 operation steps;

400 graphic files;

500, 510, ..., 520 concepts.

Detailed ways

FIG. 2 shows the system architecture of an ontology-based knowledge graph system according to an embodiment of the present invention.

The ontology-based knowledge graph system according to an embodiment of the present invention includes a data storage device 200 and a processor 210 coupled to the data storage device 200 . The data storage device 200 includes an ontology database 201 and a knowledge graph library 202 . The ontology database 201 records a plurality of concepts and the associations between the concepts. The knowledge graph library 202 records the ontology-concept association graphs and corresponding content knowledge for different graph archives.

The processor 210 retrieves a plurality of objects from the graphic file 220 , the object names corresponding to the objects, and the inclusion relationship between the objects. Wherein, the graphics file 220 may be a file with multiple objects, such as a Microsoft PowerPoint file or a Visio file. The processor 210 can obtain the objects in the graphics file 220 and their corresponding object names as well as the coordinates and areas of each object in the graphics file 220 by scanning and analyzing the files. Through the coordinates and area of each object in the graphics file 220, the inclusion relationship between these objects can be known.

Next, the processor 210 searches the ontology database 201 for a corresponding concept according to the object name of each retrieved object. Wherein, the processor queries the corresponding concept from the ontology database according to the object name of the object, and performs query through concept keyword comparison or concept similarity comparison. If the concept of the corresponding object does not exist in the ontology database 201, then add a concept corresponding to the object in the ontology database 201, and specify the association between the concept and other concepts in the ontology database.

Afterwards, the processor 210 establishes an ontology concept association graph corresponding to the graphic file according to the concept of the corresponding object and the inclusion relationship between the objects, and combines the graphic file 220 and the corresponding content knowledge, such as the background information of the graphic file, with the corresponding graphic file. The ontology-concept graph of the file 220 is stored in the knowledge graph library 202 .

It should be noted that when automatically extracting the objects in the graphic file 220 and their corresponding inclusion relationships, the system can also provide an interface for the user to perform interactive settings and corrections, as well as input background information related to the graphic file 220, so as to further strengthen the system's automatic Correctness when converting concepts.

FIG. 3 shows the operation flow of the ontology-based knowledge graph method according to an embodiment of the present invention.

First, as in step S300, a plurality of objects, corresponding object names, and inclusion relationships among the objects are extracted from the graphics file. Similarly, by scanning and analyzing the file, the objects in the graphic file and their corresponding object names, as well as the coordinates and area of each object in the graphic file can be known. Through the coordinates and area of each object in the graphics file, the inclusion relationship between these objects can be known.

Next, as in step S310 , for each object, the corresponding concept is queried from the ontology database using concept keyword comparison or concept similarity comparison according to the object name. If the concept corresponding to this object does not exist in the ontology database (yes in step S320), then as in step S330, add a concept corresponding to this object in the ontology database, and specify the relationship between this concept and other concepts in the ontology database associated.

Afterwards, as in step S340, an ontology-concept association graph corresponding to the graphic file is established according to the concepts corresponding to these objects and the inclusion relationship between the objects. Finally, as in step S350, the graph file, the content knowledge of the corresponding graph file, and the ontology concept association graph of the corresponding graph file are stored in the knowledge graph database.

Next, an example is given for illustration. A graphics file 400 is provided, as shown in FIG. 4 . The graphic file 400 is an introduction to the standard architecture of Web Service (network service), which is used to describe the existing standard content and related standard content.

Through the embodiment of the present invention, firstly, the objects in the graphics file 400 are deconstructed one by one, and the object name of each object is found. Therefore, the graphic file 400 can deconstruct 11 objects, which are 11 block objects including Web Service Standard, Infrastructure, Description, Search, XML, SOAP, Process Flows, WSDL, XML Schema, Inspection, and UDDI. At the same time, the inclusion relationship between objects can be known through the coordinates and areas of the objects, that is, the Web Service standard includes all other block objects; the infrastructure includes XML and SOAP two block objects; the description includes Process Flows, WSDL and XML Schema three block objects Object; Search contains two objects, Inspection and UDDI.

Afterwards, according to the found object name, use ontology's concept comparison method to find similar concepts from the ontology database, and return the found concepts. Taking the graphic file 400 as an example, the similar concepts found for each object are Web Service Standard 510, Web Service Infrastructure Standard 511, Web Service Description Standard 512, Web Service Search Standard 513, XML 514, SOAP 515, Process Flows 516 , WSDL 517, XML Schema 518, Inspection 519, and UDDI 520 and other similar concepts.

Therefore, according to the relationship between the concepts of the objects in the corresponding graphic file 400 in the original ontology database and the inclusion relationship of the objects, an ontology-concept association graph of the corresponding graphic file 400 is established, as shown in FIG. 5 . Afterwards, the graph file 400 , the ontology-concept relationship graph of the corresponding graph file 400 , and the content knowledge (which can be input by the user) of the corresponding graph file 400 are stored in the knowledge graph database.

Therefore, through the ontology-based knowledge graph system and method proposed by the present invention, the concepts and their associations in the ontology database can be automatically updated by identifying the graph files, and the ontology-concept relationship graph of the corresponding graph files can be established.

In addition, in the present invention, a body of knowledge based on ontology is displayed in traditional graphics, which is convenient for users to understand. Each graph represents the content in the knowledge ontology, which makes it easier for users to understand the differences between different contents. For example, the technical architecture diagram of Web Service and the main product analysis diagram of Web Service belong to two knowledge bodies with different content. Users can easily compare the differences and complementarities of the two knowledge bodies by seeing the graphics plus the background data of the graphics. place.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Anyone familiar with the art can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the present invention The scope of protection shall prevail as defined by the scope of the claims.

Claims (10)

1. A knowledge graph system based on ontology, characterized by: comprising: a data storage device including an ontology database recording plural concepts and associations between the concepts; and A processor, coupled to the data storage device, causes a plurality of objects and corresponding object names to be retrieved from a graphics file, as well as the inclusion relationship between these objects, and for each of these objects according to its object name from the ontology Query the corresponding concept in the database. If the concept of the corresponding object does not exist, add the concept of the corresponding object in the ontology database, and specify the relationship between the concept and other concepts in the ontology database, and according to the corresponding The concepts of such objects and the inclusion relationship between these objects are used to establish an ontology-concept association graph corresponding to the graphic file. 2. The knowledge graph system based on ontology as claimed in claim 1, further comprising a knowledge graph database for storing the graph file, a content knowledge corresponding to the graph file, and the corresponding graph The ontology concept association graph of the archive. 3. The knowledge graph system based on ontology as claimed in claim 1, characterized in that: the processor queries the corresponding concept from the ontology database according to the object name of the object, which is carried out by comparing concept keywords query. 4. The knowledge graph system based on ontology as claimed in claim 1, characterized in that: the processor queries the corresponding concept from the ontology database according to the object name of the object, which is performed by comparing concept similarity query. 5. The knowledge graph system based on ontology as claimed in claim 1, characterized in that: the containment relationship between the objects is obtained according to the coordinates and area of each of the objects in the graph file . 6. A knowledge graph method based on ontology, characterized in that: comprising the following steps: Extracting multiple objects, corresponding object names and inclusion relationships among these objects from a graphics file; For each of these objects, query the corresponding concept from an ontology database according to its object name; If the concept corresponding to the object does not exist, adding a concept corresponding to the object in the ontology database, and specifying the relationship between the concept and other concepts in the ontology database; and An ontology-concept association graph corresponding to the graphic file is established according to the concepts of the corresponding objects and the inclusion relationship among the objects. 7. The knowledge graphic method based on ontology as claimed in claim 6, further comprising the graphic file, a content knowledge corresponding to the graphic file, and the ontology concept correlation graph corresponding to the graphic file Stored in a knowledge graph library. 8. The knowledge graph method based on ontology as claimed in claim 6, characterized in that: according to the object name of the object, the corresponding concept is queried from the ontology database, and the query is performed by comparing concept keywords. 9. The knowledge graph method based on ontology as claimed in claim 6, characterized in that: according to the object name of the object, the corresponding concept is queried from the ontology database, and the query is performed by comparison of concept similarity. 10. The knowledge graphic method based on ontology as claimed in claim 6, wherein the inclusion relationship between the objects is obtained according to the coordinates and area of each of the objects in the graphic file .

Images