CN110598003A - Knowledge graph construction system and construction method based on public data resource catalog - Google Patents
Knowledge graph construction system and construction method based on public data resource catalog Download PDFInfo
- Publication number
- CN110598003A CN110598003A CN201910754920.5A CN201910754920A CN110598003A CN 110598003 A CN110598003 A CN 110598003A CN 201910754920 A CN201910754920 A CN 201910754920A CN 110598003 A CN110598003 A CN 110598003A
- Authority
- CN
- China
- Prior art keywords
- entity
- knowledge
- data
- unit
- entities
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a knowledge graph construction system, which comprises an entity identification unit, a first knowledge graph generation unit and a knowledge graph construction unit, wherein the entity identification unit is used for identifying a corresponding first entity in a plurality of data sources; the relationship identification unit is used for identifying entity relationships among the first entities; the mapping relation construction unit is used for constructing the mapping relation between the second entity and each first entity in the body; the ambiguity eliminating unit is used for eliminating entity ambiguity of the same first entities in the same data source; a redundancy elimination unit for eliminating redundant entities; the relationship reasoning unit is used for completing and correcting the entity relationship among the first entities after the redundant entities are eliminated; the first visualization unit is used for substituting the entity relationships between the first entities into a first knowledge expression structure and then performing visualization processing on the first knowledge expression structure to obtain a knowledge graph.
Description
Technical Field
The invention relates to the technical field of knowledge graphs, in particular to a knowledge graph construction system and a construction method based on a public data resource directory.
Background
The Knowledge map (also called scientific Knowledge map) is a Knowledge domain visualization or Knowledge domain mapping map in the book intelligence world, and is a series of different graphs for displaying the relationship between the Knowledge development process and the structure, describing Knowledge resources and carriers thereof by using a visualization technology, and mining, analyzing, constructing, drawing and displaying Knowledge and the mutual relation among the Knowledge resources and the carriers.
The knowledge graph is essentially a knowledge base of a Semantic Network (Semantic Network), and as a novel knowledge representation mode generated under the background of big data, the knowledge graph provides a new management mode for the data, and the knowledge graph becomes a new hotspot in the field of knowledge services in recent years.
Currently, government affair data in China is usually stored in a relational database. However, the relational database adopts an E-R (entity-relationship) model, which records the association relationship between data by establishing database table relationship, but the model cannot adapt to data relationship management with large data volume and high complexity, the execution efficiency is not high, and the maintenance cost of the relational database is high. Meanwhile, in the relational database, a large number of table connections may be caused by data query, so that the response speed of the data query is influenced, and the user experience is influenced.
Disclosure of Invention
In view of the above technical problems, the present invention provides a knowledge graph construction system to solve the above technical problems.
The technical scheme adopted by the invention for solving the technical problem is to provide a knowledge graph construction system for constructing a knowledge graph based on a plurality of data sources, wherein the knowledge graph construction system comprises:
an entity identification unit, configured to identify a corresponding first entity in the plurality of data sources;
the relationship identification unit is connected with the entity identification unit and used for identifying the entity relationship among the first entities;
the mapping relation construction unit is respectively connected with the entity identification unit and the relation identification unit 2 and is used for constructing the mapping relation between a second entity in the ontology and each first entity based on a first ontology expression structure;
the ambiguity eliminating unit is connected with the entity identifying unit and is used for carrying out entity fusion on the same first entities in the same data source so as to eliminate entity ambiguity;
a redundancy elimination unit connected with the ambiguity elimination unit and used for carrying out entity analysis on each first entity after ambiguity elimination so as to eliminate redundant entities;
the relationship reasoning unit is connected with the redundancy elimination unit and used for completing and correcting the entity relationship among the first entities after redundant entities are eliminated;
and the first visualization unit is respectively connected with the redundancy elimination unit and the relationship reasoning unit and used for substituting the entity relationships between the first entities and the first entities into the first knowledge expression structure based on the mapping relationship, and then performing visualization processing on the first knowledge expression structure to obtain the knowledge graph.
As a preferred aspect of the present invention, the knowledge-graph constructing system further includes:
the semantic extraction unit is used for extracting semantic information corresponding to each data in the relational database;
a schema mapping unit, configured to convert the relational schema in the relational database into an RDF (resource description framework) schema;
the data mapping unit is respectively connected with the semantic extraction unit and the mode mapping unit and used for mapping each data in the relational database into corresponding RDF data based on the semantic information corresponding to each data and the RDF mode corresponding to the relational database;
and the second visualization unit is connected with the data mapping unit and used for forming a second knowledge expression structure based on the RDF data and the RDF mode, substituting the RDF data and the entity relationship among the RDF data into the second knowledge expression structure, and then performing visualization processing on the second knowledge expression structure to obtain the knowledge graph.
As a preferred aspect of the present invention, the data source includes one or more of a public data resource directory, text information in the application scenario description, and an existing relational database.
As a preferred scheme of the invention, the method for identifying each first entity in each data source by the knowledge graph building system comprises an entity linking technology and a named entity identification technology.
As a preferable aspect of the present invention, the method for recognizing the entity relationship between the first entities by the knowledge graph constructing system is an entity relationship recognition method based on a bidirectional long-and-short term memory recurrent neural network.
The invention also provides a knowledge graph construction method, which is realized by applying the knowledge graph construction system and comprises the following steps:
step S1, the knowledge-graph building system identifies a corresponding first entity in the data source;
step S2, the knowledge graph construction system identifies the relation of each first entity to obtain the entity relation among the first entities;
step S3, the knowledge graph constructing system constructs the mapping relation between the second entity in the ontology and each first entity based on the first ontology expression structure;
step S4, the knowledge graph construction system carries out entity fusion on the same first entities in the same data source so as to eliminate entity ambiguity;
step S5, the knowledge graph construction system carries out entity analysis on each first entity after disambiguation so as to eliminate redundant entities;
step S6, the knowledge graph construction system completes and corrects the entity relationship among the first entities;
step S7, the knowledge graph construction system substitutes the first entities and the entity relationships among the first entities into the first knowledge expression structure based on the mapping relationships, and then performs visualization processing on the first knowledge expression structure to obtain the knowledge graph.
As a preferred aspect of the present invention, the method for identifying the first entity in the data source by the knowledge spectrogram constructing system in the step S1 includes an entity linking technology and a named entity identification technology.
As a preferable aspect of the present invention, the method for identifying the entity relationship in step S2 is an entity relationship identification method based on a bidirectional long-and-short term memory recurrent neural network.
The invention also provides another knowledge graph construction method, which is realized by applying the knowledge graph construction system and is characterized by comprising the following steps:
l1, extracting semantic information corresponding to each data in the relational database by the knowledge graph construction system;
step L2, the knowledge-graph building system converts the relational schema of the relational database into an RDF (resource description framework) schema;
step L3, the knowledge-graph construction system maps each data in the relational database into corresponding RDF data based on the semantic information corresponding to each data and the RDF mode corresponding to the relational database;
and L4, forming a second knowledge expression structure by the knowledge graph construction system based on the RDF data and the RDF mode, substituting the RDF data and the entity relationship among the RDF data into the second knowledge expression structure, and then carrying out visualization processing on the second knowledge expression structure to obtain the knowledge graph.
The invention constructs the knowledge graph through the RDF (resource description framework) model, can better adapt to the data relation management with large data volume and high complexity, is beneficial to improving the data query efficiency of users, improves the utilization value of data and further improves the work efficiency of enterprises or government departments.
Drawings
FIG. 1 is a schematic structural diagram of a knowledge graph building system according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a knowledge-graph building system according to a second embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a knowledge graph construction system provided by a third embodiment of the invention;
FIG. 4 is a diagram of the steps of a method for implementing knowledge graph construction by using the knowledge graph construction system according to the first embodiment of the present invention;
FIG. 5 is a diagram of the steps of a method for implementing the construction of a knowledge-graph by using the knowledge-graph construction system provided in the second embodiment or the third embodiment of the invention;
fig. 6 is a flowchart of a knowledge graph building system provided in the first embodiment or the second embodiment or the third embodiment of the present invention for implementing conversion of a relational database from a relational schema to an RDF schema.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
Example one
Referring to fig. 1, a knowledge graph constructing system provided in this embodiment is used for constructing a knowledge graph based on a plurality of data sources, and includes:
an entity identification unit 1, configured to identify a corresponding first entity in a plurality of data sources, where the data sources include text information in a public data resource directory and an application scenario description; the public data resource catalog comprises government affair information, and the content of the government affair information comprises data source information, data open range information, data level information, a data responsibility list and a data demand list;
the relationship identification unit 2 is connected with the entity identification unit 1 and used for identifying entity relationships among the first entities;
the mapping relation construction unit 3 is respectively connected with the entity identification unit 1 and the relation identification unit 2, and is used for constructing the mapping relation between a second entity and each first entity in the ontology based on a first ontology expression structure;
the ambiguity eliminating unit 4 is connected with the entity identifying unit 1 and is used for carrying out entity fusion on the same first entities in the same data source so as to eliminate entity ambiguity;
a redundancy elimination unit 5 connected to the ambiguity elimination unit 4, configured to perform entity resolution on each first entity after ambiguity elimination to eliminate redundant entities;
the relation reasoning unit 6 is connected with the redundancy eliminating unit 5 and used for completing and correcting the entity relation among the first entities after the redundant entities are eliminated;
and the first visualization unit 7 is respectively connected with the redundancy elimination unit 5 and the relationship inference unit 6, and is used for substituting the entity relationships between the first entities and the first entities into the first knowledge expression structure based on the mapping relationship, and then performing visualization processing on the first knowledge expression structure to obtain the knowledge graph.
In the above technical solution, the method for identifying the corresponding first entity from the text information of the public data resource directory by the knowledge graph construction system preferably adopts an entity linking technology, and the entity linking is to map the entity into a given knowledge base. The entity linking technology is an entity identification technology existing in the prior art, and a specific identification process of an entity is not the scope of the claimed invention and is not set forth herein.
In the above technical solution, the method for the knowledge graph construction system to identify the corresponding first entity from the administration information and the text information described in the application scenario is preferably a named entity identification method. The named entity recognition method is also an entity recognition technology existing in the prior art, and the specific recognition process of the entity is not the scope of the claimed invention and is not elaborated herein.
In the above technical solution, the mapping relationship between the knowledge graph construction system construction ontology and the first entity adopts a data mapping technology existing in the prior art. The process of the data mapping technique to construct the mapping relationship is not the scope of the claimed invention and is not set forth herein.
In the technical scheme, the method for eliminating entity ambiguity by the knowledge graph construction system comprises the steps of connecting the entity designation chain of the ambiguous first entity to a given knowledge base, and then performing entity fusion on the same first entities in the same data source to eliminate entity ambiguity.
In the above technical solution, the knowledge graph construction system preferably adopts a link prediction technology to complete and correct the entity relationship between the first entities after the redundancy is eliminated. The link prediction technology is a technology existing in the prior art, and the supplementing and correcting functions of the entity relationship are not explained here.
The embodiment also provides a method for constructing a knowledge graph, which is implemented by applying the knowledge graph construction system described above, with reference to fig. 4, and the method specifically includes the following steps:
step S1, the knowledge-graph constructing system identifies a corresponding first entity in the data source;
step S2, the knowledge graph construction system identifies the relationship of each identified first entity to obtain the entity relationship of each first entity;
step S3, the knowledge graph constructing system constructs the mapping relation between the second entity and each first entity in the ontology based on a first ontology expression structure;
step S4, the knowledge graph construction system carries out entity fusion on the same first entities in the same data source so as to eliminate entity ambiguity;
step S5, the knowledge graph construction system carries out entity analysis on each first entity after the ambiguity is eliminated so as to eliminate redundant entities;
step S6, the knowledge graph construction system completes and corrects the entity relationship among the first entities;
and step S7, the knowledge graph construction system substitutes the entity relations between the first entities into the first knowledge expression structure based on the mapping relation, and then visualizes the first knowledge expression structure to obtain a knowledge graph.
In the above method steps, specifically, in step S1, when the knowledge graph constructing system performs entity identification on the text information in the application scenario description, it is preferable to link candidate entities possibly existing in the text information to a preset knowledge base through an entity linking technology existing in the prior art, so as to identify a first entity existing in the text information described in the application scenario, and remove useless knowledge from the text information;
when the knowledge graph construction system identifies the entity of the text information in the public data resource directory, the named entity identification technology existing in the prior art is preferably adopted to identify the first entity existing in the text information, and useless knowledge is removed.
The entity linking technology and the named entity recognition technology are both entity recognition technologies existing in the prior art, and the detailed processes of the two for recognizing the entities are not described herein.
In step S2, the knowledge graph constructing system preferably identifies the entity relationship between the application scenario description and each first entity in the text information of the public data resource directory by constructing a bidirectional long-and-short term memory recurrent neural network model through a deep learning technique. The entity relationship identification method based on the bidirectional long-time memory recurrent neural network is an entity relationship identification method existing in the prior art, and the process of identifying the entity relationship is not explained here.
In step S3, the knowledge-graph constructing system preferably constructs a mapping relationship between the second entity and each first entity in the ontology by using a data mapping technique existing in the prior art, so as to perform data fusion on each first entity originating from different data sources. For example, a pattern matching technique in a data mapping technique may be used, and a plurality of data matching algorithms are structured, similarity calculation is performed on the basic terms of each first entity, calculation results are combined, then consistency check is performed on the combined calculation results, the corresponding relation causing inconsistency is removed, and the process is continuously circulated until a new entity relation cannot be found.
In step S4, the knowledge graph constructing system links the ambiguous entity designation chains to a given knowledge base by an entity linking technique, identifies a corresponding first entity, then performs pairwise combination on the identified first entities, then performs entity relationship classification, and finally performs semantic similarity calculation between the first entities having entity relationships and the corresponding entity relationships as triple data and the candidate entities, thereby implementing entity ambiguity elimination.
Since the same entity name may represent different things and different entity names may represent the same things, the knowledge graph building system needs to perform redundant entity elimination on the first entities after the entity ambiguity is eliminated in step S5. In one embodiment, the knowledge graph construction system preferably predicts whether different entities represent the same object through the semantic entity analysis of the context by using a knowledge inference link technology existing in the prior art, so as to eliminate redundant entities.
The entity resolution process is specifically as follows:
the knowledge graph construction system pushes the determined entity relationship and the grouped first entity as an entity analysis object into an event buffer queue to trigger an inference mechanism and start knowledge inference. When the knowledge graph construction system detects the data abnormal state, the possible reasons are deduced forward according to the current abnormal state, then the further evidence is obtained through backward reasoning, part of the reasons are eliminated, then forward reasoning is carried out, the self-adaptive learning is carried out on the reasoning result, the reasoning rule is continuously corrected, and finally the entity analysis process is completed.
In step S6, the knowledge-graph constructing system preferably uses a link prediction technique to complete and correct the entity relationship between the first entities after the redundancy is eliminated, that is, to perform link prediction on the entity relationship possibly existing in the knowledge graph, to complete the missing entity relationship between the entities, and to correct the erroneous entity relationship between the entities.
Example two
Referring to fig. 2, the knowledge graph constructing system provided in the second embodiment includes:
a semantic extraction unit 8, configured to extract semantic information corresponding to each data in the relational database;
a schema mapping unit 9, configured to convert a relational schema of a relational database into an RDF (resource description framework) schema;
the data mapping unit 10 is respectively connected with the semantic extraction unit 8 and the mode mapping unit 9, and is used for mapping each data in the relational database into corresponding RDF data based on the semantic information and the RDF mode corresponding to each data;
and the second visualization unit 11 is connected to the data mapping unit 10, and is configured to form a second knowledge expression structure based on each RDF data and the RDF mode, substitute each RDF data and the entity relationship between each RDF data into the second knowledge expression structure, and perform visualization processing on the second knowledge expression structure to obtain a knowledge graph.
The second embodiment further provides a method for constructing a knowledge graph, which is implemented by applying the knowledge graph construction system provided in the second embodiment, with reference to fig. 5, and specifically includes the following steps:
l1, extracting semantic information corresponding to each data in the relational database by the knowledge graph construction system;
l2, converting the relational schema of the relational database into an RDF (resource description framework) schema by the knowledge graph construction system;
l3, the knowledge graph construction system maps each data in the relational database into corresponding RDF data based on the semantic information corresponding to each data and the RDF mode corresponding to the relational database;
and L4, forming a second knowledge expression structure by the knowledge graph construction system based on the RDF data and the RDF mode, substituting the RDF data and the entity relationship among the RDF data into the second knowledge expression structure, and then carrying out visualization processing on the second knowledge expression structure to obtain a knowledge graph.
In the above technical solution, the knowledge graph construction system converts the relational schema of the relational database into the RDF schema in accordance with the following rules:
a first rule: each table in the relational database is mapped to a resource class in the RDF mode. The table name is referred to as the resource class name.
The second rule is as follows: and mapping each column in the table to be an attribute of the resource class corresponding to the table, wherein the column name is used as a resource attribute name. For non-foreign key columns, their values will map to words; for foreign key columns, their values will map to resources.
In the above technical solution, the knowledge graph construction system converts each data in the relational database into corresponding RDF data according to the following rules:
a third rule: each entity (each tuple) in the relational data table in the relational database is mapped to a RDF Resource, and a unique URL (uniform Resource Identifier) is determined for the Resource. The resource URL may be generated in conjunction with the primary key value. If the primary key in the original relational data table contains a single attribute, the table name and the primary key value of the tuple are combined to serve as the URL of the resource. If the relational data table contains a plurality of attributes, the URL of the resource is defined as the value obtained by adding the attribute values of the primary key to the table name.
The fourth rule is that: and mapping the non-foreign key column value in the relational database to the attribute value of the RDF resource corresponding to the entity, wherein the attribute value is a character, and appointing a corresponding RDF data type for the character according to the data type of the column in the table.
A fifth rule: and mapping the foreign key column value in the table in the relational database to the attribute value of the RDF resource corresponding to the entity, wherein the attribute value is one resource, namely the resource corresponding to the entity referred by the foreign key.
Referring to fig. 6, the process of the knowledge graph building system implementing RDF conversion of the relational database according to the above rules is as follows:
the first step is as follows: the knowledge graph construction system reads a table from the relational database, and then declares a resource class for the table according to the RDF mode requirement, wherein the name of the resource class is consistent with the name of the table;
the second step is that: the knowledge-graph building system reads a column in the table and then declares a corresponding attribute of a resource class corresponding to the table for the column based on the RDF schema requirements. Then checking the type of the column, and declaring that the attribute value type of the non-foreign key column is a character; and for the foreign key column, declaring that the attribute value type is a resource, and the resource type is a class corresponding to the referenced entity.
The third step: repeating the second step until all columns in the table are processed;
the fourth step: the knowledge graph construction system reads a tuple (row) in the table from the relational database, creates a resource for the tuple, and the URL creation of the resource follows the third rule;
the fifth step: taking a column of the tuple, if the tuple is a non-foreign key column, adding an attribute to the resource, wherein the attribute value is the value of the tuple on the attribute column, and the attribute value type is an XML (extensible markup language) data type corresponding to the column type; if the resource is the foreign key column, the attribute value type is the resource corresponding to the value of the tuple on the foreign key column;
and a sixth step: repeating the fifth step until all the columns in the tuple are processed;
the seventh step: returning to the fourth step until each primitive ancestor in the table is processed;
eighth step: and returning to the first step until each table in the relational database is processed.
EXAMPLE III
In the third embodiment, please refer to fig. 3. The third embodiment comprises all technical features of the knowledge graph construction system provided by the first embodiment and the second embodiment respectively. For example, the knowledge graph construction system provided by the first embodiment can construct a knowledge graph based on the public data resource directory and the text information in the application scene description, but cannot construct a knowledge graph based on the existing relational database; the knowledge graph construction system provided by the second embodiment can construct the knowledge graph based on the existing relational database, but cannot construct the knowledge graph base based on the public data resource catalog and the text information in the application scene description. And in the third embodiment, the knowledge graph can be constructed based on the public data resource catalog, the text information in the application scene description and the existing relational database, so that the third embodiment has stronger functions and stronger practicability.
In conclusion, the knowledge graph is constructed through the RDF (resource description framework) model, so that the method and the system can better adapt to the data relation management with large data volume and high complexity, are favorable for improving the data query efficiency of the user, improve the utilization value of the data and further improve the work efficiency of enterprises or government departments.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (9)
1. A knowledge graph construction system for constructing a knowledge graph based on a plurality of data sources, comprising:
an entity identification unit, configured to identify a corresponding first entity in the plurality of data sources;
the relationship identification unit is connected with the entity identification unit and used for identifying the entity relationship among the first entities;
the mapping relation construction unit is respectively connected with the entity identification unit and the relation identification unit and is used for constructing the mapping relation between a second entity in the ontology and each first entity based on a first ontology expression structure;
the ambiguity eliminating unit is connected with the entity identifying unit and is used for carrying out entity fusion on the same first entities in the same data source so as to eliminate entity ambiguity;
a redundancy elimination unit connected with the ambiguity elimination unit and used for carrying out entity analysis on each first entity after ambiguity elimination so as to eliminate redundant entities;
the relationship reasoning unit is connected with the redundancy elimination unit and used for completing and correcting the entity relationship among the first entities after redundant entities are eliminated;
and the first visualization unit is respectively connected with the redundancy elimination unit and the relationship reasoning unit and used for substituting the entity relationships between the first entities and the first entities into the first knowledge expression structure based on the mapping relationship, and then performing visualization processing on the first knowledge expression structure to obtain the knowledge graph.
2. The knowledge-graph building system of claim 1, further comprising:
the semantic extraction unit is used for extracting semantic information corresponding to each data in the relational database;
a schema mapping unit for converting the relational database relational schema into an RDF (resource description framework) schema;
the data mapping unit is respectively connected with the semantic extraction unit and the mode mapping unit and used for mapping each data in the relational database into corresponding RDF data based on the semantic information corresponding to each data and the RDF mode corresponding to the relational database;
and the second visualization unit is connected with the data mapping unit and used for forming a second knowledge expression structure based on the RDF data and the RDF mode, substituting the RDF data and the entity relationship among the RDF data into the second knowledge expression structure, and then performing visualization processing on the second knowledge expression structure to obtain the knowledge graph.
3. The knowledgegraph building system of claim 1, wherein the data source comprises one or more of a public data resource catalog, textual information in an application scenario description, and an existing relational database.
4. The knowledge-graph building system of claim 1 wherein the method of the knowledge-graph building system identifying each of the first entities in each of the data sources comprises an entity linking technique and a named-entity identification technique.
5. The knowledge-graph building system of claim 1 wherein the method of the knowledge-graph building system identifying the entity relationships between the first entities is an entity relationship identification method based on a two-way long-and-short-term memory recurrent neural network.
6. A knowledge-graph construction method implemented by applying the knowledge-graph construction system according to any one of claims 1 to 5, comprising the steps of:
step S1, the knowledge-graph building system identifies a corresponding first entity in the data source;
step S2, the knowledge graph construction system identifies the relation of each first entity to obtain the entity relation among the first entities;
step S3, the knowledge graph constructing system constructs the mapping relation between the second entity in the ontology and each first entity based on the first ontology expression structure;
step S4, the knowledge graph construction system carries out entity fusion on the same first entities in the same data source so as to eliminate entity ambiguity;
step S5, the knowledge graph construction system carries out entity analysis on each first entity after disambiguation so as to eliminate redundant entities;
step S6, the knowledge graph construction system completes and corrects the entity relationship among the first entities;
step S7, the knowledge graph construction system substitutes the first entities and the entity relationships among the first entities into the first knowledge expression structure based on the mapping relationships, and then performs visualization processing on the first knowledge expression structure to obtain the knowledge graph.
7. The method of knowledge-graph construction according to claim 6, wherein the method of knowledge-graph construction system identifying the first entity in the data source in step S1 includes an entity linking technique and a named entity identification technique.
8. The method of constructing a knowledge graph according to claim 6, wherein the method of identifying the entity relationship in step S2 is an entity relationship identification method based on a two-way long-and-short-term memory recurrent neural network.
9. A method for constructing a knowledge graph by applying the knowledge graph construction system according to claim 2, comprising the steps of:
l1, extracting semantic information corresponding to each data in the relational database by the knowledge graph construction system;
step L2, the knowledge-graph building system converts the relational schema of the relational database into an RDF (resource description framework) schema;
step L3, the knowledge-graph construction system maps each data in the relational database into corresponding RDF data based on the semantic information corresponding to each data and the RDF mode corresponding to the relational database;
and L4, forming a second knowledge expression structure by the knowledge graph construction system based on the RDF data and the RDF mode, substituting the RDF data and the entity relationship among the RDF data into the second knowledge expression structure, and then carrying out visualization processing on the second knowledge expression structure to obtain the knowledge graph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910754920.5A CN110598003A (en) | 2019-08-15 | 2019-08-15 | Knowledge graph construction system and construction method based on public data resource catalog |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910754920.5A CN110598003A (en) | 2019-08-15 | 2019-08-15 | Knowledge graph construction system and construction method based on public data resource catalog |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110598003A true CN110598003A (en) | 2019-12-20 |
Family
ID=68854444
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910754920.5A Pending CN110598003A (en) | 2019-08-15 | 2019-08-15 | Knowledge graph construction system and construction method based on public data resource catalog |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110598003A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125361A (en) * | 2019-12-20 | 2020-05-08 | 天津幸福生命科技有限公司 | Data enhancement method and device, storage medium and electronic equipment |
| CN111797243A (en) * | 2020-07-03 | 2020-10-20 | 中国烟草总公司湖南省公司 | Knowledge graph data system construction method, system, terminal and readable storage medium |
| CN112182236A (en) * | 2020-09-18 | 2021-01-05 | 成都数联铭品科技有限公司 | Knowledge graph construction method and system and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160275180A1 (en) * | 2015-03-19 | 2016-09-22 | Abbyy Infopoisk Llc | System and method for storing and searching data extracted from text documents |
| CN108446368A (en) * | 2018-03-15 | 2018-08-24 | 湖南工业大学 | A kind of construction method and equipment of Packaging Industry big data knowledge mapping |
| CN108932340A (en) * | 2018-07-13 | 2018-12-04 | 华融融通（北京）科技有限公司 | The construction method of financial knowledge mapping under a kind of non-performing asset operation field |
| CN109147954A (en) * | 2018-07-26 | 2019-01-04 | 南京邮电大学 | The patient information processing unit of knowledge based map |
| US20190171945A1 (en) * | 2017-12-05 | 2019-06-06 | International Business Machines Corporation | Context-aware knowledge base system |
| CN110083709A (en) * | 2019-04-28 | 2019-08-02 | 宁波深擎信息科技有限公司 | A kind of knowledge mapping method for auto constructing and system based on description definition |
-
2019
- 2019-08-15 CN CN201910754920.5A patent/CN110598003A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160275180A1 (en) * | 2015-03-19 | 2016-09-22 | Abbyy Infopoisk Llc | System and method for storing and searching data extracted from text documents |
| US20190171945A1 (en) * | 2017-12-05 | 2019-06-06 | International Business Machines Corporation | Context-aware knowledge base system |
| CN108446368A (en) * | 2018-03-15 | 2018-08-24 | 湖南工业大学 | A kind of construction method and equipment of Packaging Industry big data knowledge mapping |
| CN108932340A (en) * | 2018-07-13 | 2018-12-04 | 华融融通（北京）科技有限公司 | The construction method of financial knowledge mapping under a kind of non-performing asset operation field |
| CN109147954A (en) * | 2018-07-26 | 2019-01-04 | 南京邮电大学 | The patient information processing unit of knowledge based map |
| CN110083709A (en) * | 2019-04-28 | 2019-08-02 | 宁波深擎信息科技有限公司 | A kind of knowledge mapping method for auto constructing and system based on description definition |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125361A (en) * | 2019-12-20 | 2020-05-08 | 天津幸福生命科技有限公司 | Data enhancement method and device, storage medium and electronic equipment |
| CN111797243A (en) * | 2020-07-03 | 2020-10-20 | 中国烟草总公司湖南省公司 | Knowledge graph data system construction method, system, terminal and readable storage medium |
| CN112182236A (en) * | 2020-09-18 | 2021-01-05 | 成都数联铭品科技有限公司 | Knowledge graph construction method and system and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111625659B (en) | Knowledge graph processing method, device, server and storage medium | |
| CN111339313A (en) | Knowledge base construction method based on multi-mode fusion | |
| WO2021083239A1 (en) | Graph data query method and apparatus, and device and storage medium | |
| CN103473224B (en) | Problem semantization method based on problem solving process | |
| CN111428054A (en) | Construction and storage method of knowledge graph in network space security field | |
| CN110598003A (en) | Knowledge graph construction system and construction method based on public data resource catalog | |
| CN110189831B (en) | Medical record knowledge graph construction method and system based on dynamic graph sequence | |
| CN111538848A (en) | Knowledge representation learning method fusing multi-source information | |
| CN105786980A (en) | Method and apparatus for combining different examples for describing same entity and equipment | |
| WO2023045417A1 (en) | Fault knowledge graph construction method and apparatus | |
| CN107391677A (en) | Carry the generation method and device of the Universal Chinese character knowledge mapping of entity-relationship-attribute | |
| CN103116574B (en) | From the method for natural language text excavation applications process body | |
| CN110097278B (en) | Intelligent sharing and fusion training system and application system for scientific and technological resources | |
| CN112115232A (en) | Data error correction method and device and server | |
| CN113094449B (en) | Large-scale knowledge map storage method based on distributed key value library | |
| CN111651447A (en) | Intelligent construction life-cycle data processing, analyzing and controlling system | |
| CN110276080B (en) | Semantic processing method and system | |
| CN114218472A (en) | Intelligent search system based on knowledge graph | |
| CN111190873B (en) | Log mode extraction method and system for log training of cloud native system | |
| CN108959366B (en) | Open question-answering method | |
| CN114117000A (en) | Response method, device, equipment and storage medium | |
| CN111930963B (en) | Knowledge graph generation method and device, electronic equipment and storage medium | |
| CN116680377B (en) | Chinese medical term self-adaptive alignment method based on log feedback | |
| CN106933844B (en) | Construction method of reachability query index facing large-scale RDF data | |
| EP4365754A1 (en) | Common question answering method and system, device and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191220 |