CN116127098A - Knowledge graph construction method and device - Google Patents

Abstract

The present application discloses a method and a device for constructing a knowledge map. The method includes: obtaining ontology corresponding to the target knowledge map, extraction rules and mapping rules, the ontology includes the definition of multiple entities and the definition of related information of multiple entities, and the extraction rule is to extract data from the source data corresponding to the entity A rule, the mapping rule is a rule for mapping the fields in the source data corresponding to the entity to the ontology; receiving a query request, including the first entity in the query request, and determining whether the first entity is in the graph database corresponding to the target knowledge map according to the query request has been constructed, the query request is used to query the first relevant knowledge of the first entity in the graph database; if the first entity has not been constructed, according to the ontology, extraction rules and mapping rules, the source data corresponding to the first entity Perform extraction to determine the first relevant knowledge; import the first relevant knowledge into the graph database to construct the target knowledge map.

Description

Construction method and device of knowledge map

technical field

The present application belongs to the technical field of knowledge graph, and specifically relates to a method and device for constructing a knowledge graph.

Background technique

In recent years, Knowledge Graph has been widely used in industry and academia. The knowledge map uses the graph data structure to express and store knowledge, and uses entities, relationships, and attributes to describe the real world. There are generally two construction methods: top-down and bottom-up. The bottom-up construction method is a process of fully extracting data sources, converting and loading them into the knowledge map.

However, in the context of massive data, full extraction of data sources to construct or update knowledge graphs may result in a large amount of relevant knowledge that is hardly searchable, occupying a large amount of space resources, and full extraction takes a long time.

Contents of the invention

The embodiment of the present application provides a method and device for constructing a knowledge graph, which can reduce the time and storage space occupied by constructing a knowledge graph under massive data.

In the first aspect, the embodiment of the present application provides a method for constructing a knowledge graph, the method includes: obtaining an ontology, extraction rules, and mapping rules corresponding to the target knowledge graph; wherein, the ontology includes the definitions of multiple entities and the The definition of related information of multiple entities, the extraction rule is a rule for extracting data from the source data corresponding to the entity, and the mapping rule is a rule for mapping the fields in the source data corresponding to the entity to the ontology; receiving the query request , the query request includes the first entity, and according to the query request, it is determined whether the first entity has been constructed in the graph database corresponding to the target knowledge graph, and the query request is used in the graph database Query the first relevant knowledge of the first entity in ; if the first entity has not been constructed, according to the ontology, the extraction rules and the mapping rules, the source corresponding to the first entity Extracting data to determine the first relevant knowledge; importing the first relevant knowledge into the graph database to construct the target knowledge map.

In the second aspect, the embodiment of the present application provides a knowledge map construction device, including: an ontology preset module, used to obtain the ontology, extraction rules, and mapping rules corresponding to the target knowledge map; wherein, the ontology includes multiple The definition of the entity and the definition of the relevant information of the multiple entities, the extraction rule is a rule for extracting data from the source data corresponding to the entity, and the mapping rule is a rule for mapping the fields in the source data corresponding to the entity to the ontology Rules; a first determination module, configured to receive a query request, the query request includes a first entity, and according to the query request, determine whether the first entity has been constructed in the graph database corresponding to the target knowledge graph , the query request is used to query the first related knowledge of the first entity in the graph database; the second determination module is used to, if the first entity is not constructed, according to the ontology, The extraction rule and the mapping rule extract the source data corresponding to the first entity to determine the first related knowledge; a building module is used to import the first related knowledge into the graph database, Carry out the construction of the target knowledge map.

In a third aspect, an embodiment of the present application provides an electronic device, the electronic device includes a processor, a memory, and a program or instruction stored in the memory and operable on the processor, and the program or instruction is The processor implements the steps of the method described in the first aspect when executed.

In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or an instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented .

In the fifth aspect, the embodiment of the present application provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions, so as to implement the first aspect the method described.

In this embodiment of the application, by obtaining the ontology, extraction rules, and mapping rules corresponding to the target knowledge graph, the ontology includes definitions of multiple entities and definitions of related information of multiple entities, and the extraction rules are obtained from the source data corresponding to entities The rule for extracting data, the mapping rule is the rule for mapping the fields in the source data corresponding to the entity to the ontology; receive the query request, the query request includes the first entity, and determine the graph corresponding to the first entity in the target knowledge map according to the query request Whether the database has been constructed, the query request is used to query the first relevant knowledge of the first entity in the graph database; if the first entity has not been constructed, according to ontology, extraction rules and mapping rules, corresponding to the first entity Extract the source data to determine the first relevant knowledge; import the first relevant knowledge into the graph database to construct the target knowledge graph. It can be seen that, compared to directly extracting a large amount of knowledge from massive data and constructing a knowledge map, knowledge that is hardly searchable takes up a lot of space resources, and it is not easy to construct knowledge maps multiple times. The technical solution can extract the source data corresponding to the first entity according to the user's query request, and the amount of extracted data is smaller, which saves the time for knowledge extraction. Correspondingly, when storing the knowledge, it occupies relatively less space, and Due to the acquisition of ontology, extraction rules and mapping rules, according to each query request, the ontology, extraction rules and mapping rules can be used to extract the knowledge required for each query request, and then multiple constructions of knowledge graphs (dynamic construction) can be realized. , which reduces the time and storage space occupied by building knowledge graphs under massive data.

Description of drawings

FIG. 1 is a schematic flow diagram of a method for constructing a knowledge graph provided in an embodiment of the present application;

FIG. 2 is a schematic diagram of a target knowledge map provided by an embodiment of the present application;

Figure 3(a) is a schematic diagram of using a Bloom filter provided by another embodiment of the present application;

Figure 3(b) is a schematic diagram of using a Bloom filter provided by another embodiment of the present application;

FIG. 4 is a schematic flowchart of another method for constructing a knowledge graph provided in the embodiment of the present application;

Fig. 5 is a schematic structural diagram of a knowledge map construction device provided according to an embodiment of the present application;

Fig. 6 is a schematic structural diagram of an electronic device provided according to an embodiment of the present application.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

The terms "first", "second" and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific sequence or sequence. It should be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application can be practiced in sequences other than those illustrated or described herein, and that references to "first," "second," etc. distinguish Objects are generally of one type, and the number of objects is not limited. For example, there may be one or more first objects. In addition, "and/or" in the specification and claims means at least one of the connected objects, and the character "/" generally means that the related objects are an "or" relationship.

The method and device for constructing the knowledge map provided by the embodiments of the present application will be described in detail below through specific embodiments and application scenarios with reference to the accompanying drawings.

FIG. 1 shows a method for constructing a knowledge map provided by an embodiment of the present invention. The method can be executed by an electronic device, and the electronic device can include: a server and/or a terminal device, where the terminal device can be a mobile terminal, for example. In other words, the method can be executed by software or hardware installed in the electronic device, and the method includes the following steps:

Step 102: Obtain ontology, extraction rules and mapping rules corresponding to the target knowledge graph.

Among them, the ontology includes the definition of multiple entities and the definition of related information of multiple entities, the extraction rule is the rule for extracting data from the source data corresponding to the entity, and the mapping rule is to map the fields in the source data corresponding to the entity to the ontology the rule of.

The source data corresponding to the entity comes from the database, which is structured data. Of course, the source data corresponding to the entity can also be converted from semi-structured data or unstructured data to obtain structured data. Structured source data, for example, can be stored in the form of one or more tables.

In practical applications, the ontology includes definitions of multiple entities and definitions of related information of multiple entities, and the related information of multiple entities may include at least one of attributes of entities and relationship information between entities. Since the same entity has different meanings in different fields, the definition of the entity and its related information are also different. Therefore, before actually implementing the construction of the knowledge graph, first obtain the definition of entities in the field corresponding to the target knowledge graph and the definition of entity-related information, that is, obtain the ontology corresponding to the target knowledge graph.

The definition of entities in ontology and the definition of entity-related information are based on general concepts defined by domain dictionaries. For example, when the entity dog is defined in ontology, general concepts such as "canine" can be used instead of When defining the entity dog in the ontology, a definition with individual characteristics such as "has spots" is used.

Exemplarily, the target knowledge graph includes the two entities of book and author, and the related information of the entity includes the attribute of the entity and the relationship information between the entities. The ontology includes the general concept of the book, author, etc. The general concept of , that is, the ontology includes the definitions of the two entities of book and author. The ontology also includes the attributes of the book, such as the literary genre of the book, etc.; the attributes of the author, such as the age of the author; in addition, the ontology also includes the relationship information between the book and the author, such as the author’s creation Books etc.

Continuing from the above example, the mapping rule can be a rule that maps fields such as "Journey to the West" and novels in the source data corresponding to the book entity to the book on the ontology, the literary genre of the book, etc., that is, the mapping rule is The rules for mapping fields in source data corresponding to entities to ontology.

Extraction rules are rules for extracting data from the source data corresponding to entities, and may include rules for converting the extracted source data, that is, conversion functions, which can replace fields in the source data with corresponding storage fields in the graph database. Following the above example, the data of the entity named "Journey to the West" is stored in the graph database. Since "Journey to the West" is aliased as "Journey to the West", the extraction rules can include when "Journey to the West" appears in the source data. In one field, replace "Journey to the West" with "Journey to the West" and store its data. The extraction rules can also include that when the literary genre of the book entity in the graph database is stored in the form of m (short story), n (long novel), etc., when the literary genre of the book entity in the source data is a novel, the corresponding , the rules for converting novels into the form of n for storage. And when the method of updating the knowledge map is not set, the extraction rules may also include the rules of the extraction period, and the electronic device can extract data according to the rules of the extraction period, thereby realizing the update of the knowledge map.

It should be noted that the above mentioned ontology, extraction rules and mapping rules are used to extract knowledge that can be stored in the graph database, but the ontology, extraction rules and mapping rules are not used in the above process to extract the source data, and then realize the knowledge Graph construction.

Compared with directly constructing the knowledge map, this embodiment first obtains the ontology, extraction rules and mapping rules, and then according to the requirements, can use the ontology, extraction rules and mapping rules to extract the source data multiple times, so as to realize the construction of dynamic knowledge Atlas.

Step 104: A query request is received, and the query request includes the first entity, and according to the query request, it is determined whether the first entity has been constructed in the graph database corresponding to the target knowledge graph.

Wherein, the query request is used to query the first relevant knowledge of the first entity in the graph database.

In practical applications, the electronic device may receive a query request input by a user. The query request may include a noun input by the user, or a sentence input by the user, which is not limited in this embodiment. However, when a certain sentence is included in the query request, a corresponding extracting device is required, and the extracting device extracts one or some nouns in a certain sentence. A certain noun mentioned above, a certain noun or some nouns extracted from a certain sentence are entities corresponding to the query, that is, the first entity.

It should be noted that the first entity, the entity, and the second entity mentioned below in this technical solution are different from the entity in the ontology. The entity in the ontology is a general concept, and the first entity, entity, and the following The second entity mentioned refers to a noun that has a specific meaning or a strong reference in the text, such as a person's name, place name, institution name, time, etc. Exemplarily, among the entities in the ontology, there may be the entity of book, and the first entity, the entity, and the second entity mentioned below may be the entity of "Journey to the West".

According to the query request, the electronic device judges the entities in the graph database to determine whether the first entity has been constructed in the graph database. In practical applications, the electronic device can judge the entities in the graph database through the Bloom Filter to determine whether the first entity has been constructed in the graph database.

Step 106: If the first entity has not been constructed, extract the source data corresponding to the first entity according to ontology, extraction rules and mapping rules, and determine the first related knowledge.

When the electronic device judges the entities in the graph database and determines that the first entity has not been constructed in the graph database, use the ontology, extraction rules and mapping rules obtained in step 102 to extract the source data corresponding to the first entity, Determine the first relevant knowledge.

For example, the source data of multiple entities is shown in Table 1 below, where the first entity is "Wu Chengen", and the relevant information includes the attributes of entities and the relationship information between entities. The author and the age of the author are defined in the ontology (the attribute of the author), the work (information about the relationship with the entity of the author), and the year of creation (the attribute of the work) as examples.

Table 1

work Creation period author Author's age journey to the west Ming Dynasty Wu Chengen Ming Dynasty Yu Ding Ji 1500-1582 Wu Chengen Ming Dynasty Yang Liuqing Ming Dynasty Wu Chengen Ming Dynasty boat trip Ming Dynasty Wu Chengen Ming Dynasty Taoyuan map Ming Dynasty Wu Chengen Ming Dynasty Mr. Sheyang draft Ming Dynasty Wu Chengen Ming Dynasty Difficult to Travel (Three Poems) Tang Dynasty Li Bai Tang Dynasty happy spring night Tang Dynasty Du Fu Tang Dynasty

According to the mapping rules and the ontology, the column of the "work" field in Table 1 (source data of multiple entities) can be mapped to the entity of the work in the ontology, and the column of the "creation year" field in the source data can be mapped to the ontology In terms of the creation year in the source data, map the column of the "author" field in the source data to the author entity in the ontology, and map the column of the "author's year" field in the source data to the author's year in the ontology .

According to the extraction rules, the source data corresponding to "Wu Chengen" (the first entity) can be extracted to determine the first related knowledge: "Wu Chengen" (author); Ming Dynasty (the author's year); "Wu Chengen" created the following works (Relationship information between entities), Journey to the West, Ming Dynasty, Yu Ding Ji, 1500-1582, Yangliuqing, Ming Dynasty, Zhou Xing, Ming Dynasty, Taoyuan Map, Ming Dynasty, Mr. Sheyang's manuscript, Ming Dynasty (works and creation time) . It should be understood that the first relevant knowledge may be expressed in a form such as a table, and the above representation is only for displaying the content included in the first relevant knowledge, and this embodiment does not specifically limit the representation of the first relevant knowledge. Extract the source data corresponding to the first entity to determine the first relevant knowledge. Compared with the full extraction of knowledge in related technologies, the source data corresponding to the first entity is extracted, and the amount of extracted data is smaller, saving knowledge extraction. time.

Step 108: Import the first relevant knowledge into the graph database to construct the target knowledge graph.

The constructed target knowledge graph follows the above example, as shown in FIG. 2 , the first relevant knowledge of the first entity 201 may include the attribute 202 of the first entity, the entity 203 related to the first entity, and the attribute of the entity related to the first entity 204, and the relationship information between entities: "Wu Chengen" created Journey to the West, Yu Ding Ji, Yang Liuqing, Zhou Xing, Taoyuan Map, and Mr. Sheyang's draft.

The knowledge map construction method provided by the embodiment of the present invention obtains the ontology, extraction rules, and mapping rules corresponding to the target knowledge map. The ontology includes the definition of multiple entities and the definition of related information of multiple entities. The rules for extracting data from the corresponding source data, the mapping rules are the rules for mapping the fields in the source data corresponding to the entities to the ontology; receive the query request, the query request includes the first entity, and according to the query request, determine the first entity in the target Whether the graph database corresponding to the knowledge graph has been constructed, the query request is used to query the first related knowledge of the first entity in the graph database; if the first entity has not been constructed, according to ontology, extraction rules and mapping rules, The source data corresponding to the first entity is extracted to determine the first relevant knowledge; the first relevant knowledge is imported into the graph database to construct the target knowledge map. It can be seen that, compared to directly extracting a large amount of knowledge from massive data and constructing a knowledge map, knowledge that is hardly searchable takes up a lot of space resources, and it is not easy to construct knowledge maps multiple times. The technical solution can extract the source data corresponding to the first entity according to the user's query request, and the amount of extracted data is smaller, which saves the time for knowledge extraction. Correspondingly, when storing the knowledge, it occupies relatively less space, and Due to the acquisition of ontology, extraction rules and mapping rules, according to each query request, the ontology, extraction rules and mapping rules can be used to extract the knowledge required for each query request, and then multiple constructions of knowledge graphs (dynamic construction) can be realized. , which reduces the time and storage space occupied by building knowledge graphs under massive data.

In an implementation manner, the method also includes the following steps A1-A3:

Step A1: When the source data corresponding to the second entity is updated, determine whether the second entity has been constructed in the graph database according to the updated source data.

In practical applications, it may be determined whether the second entity has been constructed in the graph database through a Bloom filter.

It should be noted that, in order to more accurately determine whether the second entity has been constructed in the graph database, it is necessary to keep the entity name of the second entity consistent with the entity name of the entity in the graph database. If the entity name of the second entity is inconsistent with the entity name of the entity in the graph database, in order to make the entity name of the entity consistent, the entity extraction rule can be used to extract the source data corresponding to the second entity and convert the entity name of the second entity , get the entity name for storing in the graph database, and then judge whether the second entity has been constructed in the graph database. Of course, in order to make the entity name of the second entity consistent with the entity name of the entity in the graph database, normalization processing may also be performed, which is not limited in this embodiment.

Step A2: When the second entity has been constructed, extract the source data corresponding to the second entity according to ontology, extraction rules and mapping rules, and determine the second related knowledge of the second entity.

Step A3: import the second relevant knowledge into the graph database, and update the knowledge graph.

According to the updated source data, when it is determined that the second entity has been constructed in the graph database, the source data corresponding to the second entity is extracted, and the source data corresponding to the second entity and the source data corresponding to the first entity come from the same A database is structured data.

Extract the source data corresponding to the second entity to determine the second relevant knowledge. Compared with the full extraction of knowledge in related technologies, the source data corresponding to the second entity is extracted, and the amount of extracted data is smaller, saving knowledge extraction. time. The second related knowledge is imported into the graph database, and the target knowledge graph is updated, so that the second related knowledge covers the related knowledge of the second entity that has been constructed in the graph database before updating. Compared with the overall update of the knowledge map in the related technology, it is easy to be in the update period of the knowledge map when searching the knowledge map, and the search effect is poor. The time required for the update of this scheme is shorter. It is easy to search for new knowledge, and the search effect is better.

In an implementation manner, the source data corresponding to the first entity is extracted to determine the first relevant knowledge (that is, step 106), which may specifically include the following steps B1-B3:

Step B1: Obtain an index table constructed for source data corresponding to multiple entities, the index table includes entity name, entity unique code (Identity document, ID).

Step B2: According to the first entity, by searching the entity name of the entity in the index table, determine the entity ID corresponding to the entity name.

Following the example in step 106 above, when the first entity is "Journey to the West", for example, "work" (entity name of the entity) can be searched in the index table to determine the entity ID of "work".

Step B3: According to the entity ID, one or more of the following information corresponding to the first entity is determined: attributes of the entity, related entities, and relationship information between entities.

In practical applications, the electronic device acquires an index table constructed for source data corresponding to multiple entities, the index table includes an entity name field and an entity ID field, the electronic device receives the first entity, and according to the entity of the entity corresponding to the first entity The name field can be indexed by searching the entity name field and the entity ID field to obtain the source data corresponding to the first entity, and then extract one or more of the following information from the source data: attributes of the first entity, related The entity, the attribute of the entity related to the first entity, and the relationship information between the first entity and the entity related to the first entity, that is, the first related knowledge.

Similarly, extracting the source data corresponding to the second entity and determining the second related knowledge of the second entity (ie step A2) may also specifically be the processing of the first entity in the above steps B1-B3.

Since the source data in the database is stored in the form of a table, there may be a large amount of data in the table, that is, the number of columns in the table is large, or there are multiple tables. Compared with directly traversing the entire table, this The technical solution can increase the speed of extracting the first relevant knowledge or the second relevant knowledge, thereby reducing the time taken for knowledge extraction.

In an implementation manner, after determining whether the first entity has been constructed in the graph database according to the query request (that is, step 104), the following steps may also be performed:

The first related knowledge of the first entity is acquired, and the target knowledge graph corresponding to the first related knowledge is displayed.

According to the query request, it is judged whether the first entity has been constructed in the graph database, and two judgment results are obtained:

The first judgment result indicates that the first entity has been constructed in the graph database. Correspondingly, the electronic device obtains the first related knowledge of the first entity from the graph database, and displays the target knowledge graph corresponding to the first related knowledge.

The second judgment result is that the first entity has not been constructed in the graph database. Correspondingly, the electronic device first constructs the knowledge map corresponding to the first related knowledge of the first entity, and then obtains the first entity’s first relevant knowledge from the graph database. Relevant knowledge, displaying the target knowledge graph corresponding to the first relevant knowledge.

In an implementation manner, the method also includes the following step C3:

Through the Bloom filter, it is determined whether the target entity has been constructed in the graph database.

The target entity is the first entity or the second entity judged whether it has been constructed in the graph database.

Bloom filter, including a binary vector and hash Hash function. The Bloom filter can be used to determine whether the first entity or the second entity is in the constructed entity collection of the graph database, and does not need to additionally store the first entity, the second entity and the constructed entity, compared to using other The algorithm reduces memory usage.

In one implementation, as shown in Fig. 3 (a), (b), through the Bloom filter, it is determined whether the first entity or the second entity has been constructed in the graph database (ie, step C3), which can be executed For the following steps D1-D5:

Step D1: Obtain a bit array with a length of n, and the value of each bit in the bit array is zero.

The bit array is a binary vector with two values, 0 and 1.

Step D2: Obtain k hash functions and constructed entities, where k is less than n.

Step D3: According to k hash functions, for each constructed entity, k first bit positions are determined and assigned a value of one.

A hash function maps a constructed entity to a first position with a value of 1, and obtaining k hash functions means that k hash functions map each constructed entity to k first positions and The value is 1.

Step D4: Determine k second bit positions of each target entity according to k hash functions.

The target entity is the first entity or the second entity judged whether it has been constructed in the graph database.

When k is 1, a hash function can map a target entity to a bit position (a second position) in the bit array. When using a hash function for mapping, it is only necessary to judge whether the value of a second position is 0, when the constructed entity and the target entity are mapped to the same bit position, that is, the second position of the target entity appears When the value is 1, it is easier to conclude that the target entity has been constructed in the graph database. In order to reduce the possibility of making mistakes through the judgment of the Bloom filter, multiple hash functions can be obtained, and k is greater than 1 and less than n.

Step D5: Determine whether the target entity has been constructed according to the values of the k second bit positions.

According to the values of the k second bit positions, it is judged whether there is a second bit position with a value of zero. If there is a second bit position with a value of zero, it can be determined that the target entity has not been constructed in the graph database. Exemplary Take k as 3 and n as 10 as an example, one of the cases is shown in Figure 3(a), the constructed entity 310 is mapped to three first bit positions 311, and the target entity 320 is mapped to three second bit positions Bit position 321, there are two second bit positions with a value of zero, it is determined that the target entity is not constructed in the graph database. If the values of the second bit positions are all 1, there are the following two situations.

In the first case, the target entity has already been constructed in the graph database. Since the target entity is actually a constructed entity, that is, the second bit position corresponding to the target entity is the first bit position of the certain constructed entity, the values corresponding to the second bit position are all 1.

In the second case, the target entity is not constructed in the graph database. The target entity should be mapped to the second bit position with a value of zero, but because a certain second bit position corresponding to the target entity is repeatedly mapped to a certain first bit position of the constructed entity, and the k second bit positions corresponding to the target entity The position is completely mapped to the first bit position of the constructed entity, resulting in the value of the second bit position being all 1, but in fact the target entity has not been constructed in the graph database. For example, k is 3, n is 10 as an example, one of the cases is shown in Figure 3(b), since the three second bit positions 321 corresponding to the target entity 320 are repeatedly mapped to the three first bit positions 311 of the constructed entity, resulting in The value of the second bit position is all 1, but actually the target entity is not constructed in the graph database.

The occurrence of the above-mentioned second situation is related to k and n. In order to reduce the occurrence of the above-mentioned second situation, k and n need to be set in advance.

It should be noted that if the values of the k second bit positions of the target entity are all 1, correspondingly, the electronic device obtains the first related knowledge of the target entity from the graph database, and when the first related knowledge of the target entity cannot be obtained from the graph database When there is relevant knowledge, execute step 106 to step 108 .

Since the Bloom filter judges whether the first entity or the second entity is in the constructed entity set of the graph database by calculating the value of the Hash function mapping position, the required judgment time is shorter than other algorithms.

Fig. 4 is a schematic flowchart of a method for constructing a knowledge map according to another embodiment of the present application. As shown in Fig. 4, the method for constructing a knowledge map may include the following steps:

Step 401, obtain ontology, extraction rules and mapping rules corresponding to the target knowledge graph.

Among them, the ontology includes the definition of multiple entities and the definition of related information of multiple entities, the extraction rule is the rule for extracting data from the source data corresponding to the entity, and the mapping rule is to map the fields in the source data corresponding to the entity to the ontology the rule of.

In step 402, a query request is received, and the query request includes the first entity.

Step 403, according to the query request, determine whether the first entity has been constructed in the graph database corresponding to the target knowledge graph through the Bloom filter.

The query request is used to query the first relevant knowledge of the first entity in the graph database.

If yes, then execute step 408, if not, then execute 404-step 408.

Step 404, if the first entity has not been constructed, according to ontology, extraction rules and mapping rules, obtain an index table constructed for source data corresponding to multiple entities, the index table includes entity name and entity ID.

Step 405, according to the first entity, by searching the entity name of the entity in the index table, determine the entity ID corresponding to the entity name.

Step 406, according to the entity ID, determine the attributes of the entity corresponding to the first entity, related entities, and relationship information between entities.

Step 407, importing the attribute of the entity corresponding to the first entity, related entities, and relationship information between entities into the graph database to construct the target knowledge graph.

Step 408, if the first entity has been constructed, acquire the first relevant knowledge of the first entity, and display the target knowledge graph corresponding to the first relevant knowledge.

The specific process from step 401 to step 408 has been described in detail in the above embodiment, and will not be repeated here.

By adopting the technical solution of the embodiment of the present application, by obtaining the ontology, extraction rules and mapping rules corresponding to the target knowledge map, the ontology includes definitions of multiple entities and definitions of related information of multiple entities, and the extraction rules are derived from the source corresponding to the entity The rules for extracting data from the data, the mapping rules are the rules for mapping the fields in the source data corresponding to the entities to the ontology; receive the query request, the query request includes the first entity, and according to the query request, determine that the first entity corresponds to the target knowledge map Whether the graph database has been constructed, the query request is used to query the first related knowledge of the first entity in the graph database; if the first entity has not been constructed, according to the ontology, extraction rules and mapping rules, the first The source data corresponding to the entity is extracted to determine the first relevant knowledge; the first relevant knowledge is imported into the graph database to construct the target knowledge map. It can be seen that, compared to directly extracting a large amount of knowledge from massive data and constructing a knowledge map, knowledge that is hardly searchable takes up a lot of space resources, and it is not easy to construct knowledge maps multiple times. The technical solution can extract the source data corresponding to the first entity according to the user's query request, and the amount of extracted data is smaller, which saves the time for knowledge extraction. Correspondingly, when storing the knowledge, it occupies relatively less space, and Due to the acquisition of ontology, extraction rules and mapping rules, according to each query request, the ontology, extraction rules and mapping rules can be used to extract the knowledge required for each query request, and then multiple constructions of knowledge graphs (dynamic construction) can be realized. , which reduces the time and storage space occupied by building knowledge graphs under massive data.

It should be noted that, for the method for constructing a knowledge graph provided in the embodiment of the present application, the execution body may be an electronic device, or a control module in the electronic device for executing the method for constructing a knowledge graph. In the embodiment of the present application, the method for constructing the knowledge graph executed by an electronic device is taken as an example to illustrate the device for constructing the knowledge graph provided in the embodiment of the present application.

Fig. 5 is a schematic structural diagram of an apparatus for constructing a knowledge graph according to an embodiment of the present invention. As shown in FIG. 5 , the device for constructing a knowledge graph includes: an ontology preset module 510 , a first determination module 520 , a second determination module 530 , and a construction module 540 .

The ontology preset module 510 is used to obtain the ontology, extraction rules and mapping rules corresponding to the target knowledge graph. The ontology includes the definition of multiple entities and the definition of related information of multiple entities. The extraction rule is from the source data corresponding to the entity The rules for extracting data, the mapping rules are the rules for mapping the fields in the source data corresponding to entities to ontology;

The first determining module 520 is configured to receive a query request, the query request includes the first entity, and according to the query request, determine whether the first entity has been constructed in the graph database corresponding to the target knowledge graph, and the query request is used to Query the first related knowledge of the first entity;

The second determination module 530 is configured to extract the source data corresponding to the first entity according to ontology, extraction rules and mapping rules when the first entity has not been constructed, and determine the first relevant knowledge;

A construction module 540, configured to import the first relevant knowledge into the graph database to construct the target knowledge graph.

In an implementation manner, the device for constructing a knowledge graph further includes: a third determination module 550 , a fourth determination module 560 , and an update module 570 .

The third determination module 550 is configured to determine whether the second entity has been constructed in the graph database according to the updated source data when the source data corresponding to the second entity is updated;

The fourth determination module 560 is configured to extract the source data corresponding to the second entity according to ontology, extraction rules and mapping rules when the second entity has been constructed, and determine the second relevant knowledge of the second entity;

The update module 570 is configured to import the second relevant knowledge into the graph database to update the target knowledge graph.

In one implementation, the second determination module 540 is configured to obtain an index table constructed from source data corresponding to multiple entities, the index table includes entity names and entity IDs; according to the first entity, by retrieving the entity in the index table Determine the entity ID corresponding to the entity name; according to the entity ID, determine one or more of the following information corresponding to the first entity: attributes of entities, related entities, and relationship information between entities.

In one implementation manner, the first determining module 520 is further configured to acquire first relevant knowledge of the first entity, and display a target knowledge graph corresponding to the first relevant knowledge.

In an implementation manner, the judging module 580 is further configured to determine whether the target entity has been constructed in the graph database through a Bloom filter.

The target entity is the first entity or the second entity judged whether it has been constructed in the graph database.

In one implementation, the judging module 580 is configured to acquire a bit array with a length of n, where the value of each bit in the bit array is zero; acquire k hash functions and created entities, where k is less than n; According to k hash functions, for each created entity, determine k first bit positions and assign a value to one; according to k hash functions, determine k second bit positions of each target entity; according to k The value of the second bit position, which determines whether the target entity has been constructed.

Using the knowledge map construction device provided by the embodiment of the present invention, by obtaining the ontology, extraction rules and mapping rules corresponding to the target knowledge map, the ontology includes definitions of multiple entities and definitions of related information of multiple entities, and the extraction rules are from The rule for extracting data from the source data corresponding to the entity, the mapping rule is the rule for mapping the fields in the source data corresponding to the entity to the ontology; receive the query request, the query request includes the first entity, according to the query request, determine the first entity in Whether the graph database corresponding to the target knowledge graph has been constructed, the query request is used to query the first related knowledge of the first entity in the graph database; if the first entity has not been constructed, according to ontology, extraction rules and mapping rules , extract the source data corresponding to the first entity, and determine the first relevant knowledge; import the first relevant knowledge into the graph database, and construct the target knowledge graph. It can be seen that, compared to directly extracting a large amount of knowledge from massive data and constructing a knowledge map, knowledge that is hardly searchable takes up a lot of space resources, and it is not easy to construct knowledge maps multiple times. The technical solution can extract the source data corresponding to the first entity according to the user's query request, and the amount of extracted data is smaller, which saves the time for knowledge extraction. Correspondingly, when storing the knowledge, it occupies relatively less space, and Due to the acquisition of ontology, extraction rules and mapping rules, according to each query request, the ontology, extraction rules and mapping rules can be used to extract the knowledge required for each query request, and then multiple constructions of knowledge graphs (dynamic construction) can be realized. , which reduces the time and storage space occupied by building knowledge graphs under massive data.

The device for constructing the knowledge graph in the embodiment of the present application may be a device, or a component, an integrated circuit, or a chip in a terminal. The device may be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device may be a mobile phone, a tablet computer, a notebook computer, a handheld computer, a vehicle electronic device, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant). assistant, PDA), etc., non-mobile electronic devices can be servers, network attached storage (Network Attached Storage, NAS), personal computer (personal computer, PC), television (television, TV), teller machine or self-service machine, etc., this application Examples are not specifically limited.

The device for constructing the knowledge graph in the embodiment of the present application may be a device with an operating system. The operating system may be an Android (Android) operating system, an ios operating system, or other possible operating systems, which are not specifically limited in this embodiment of the present application.

The knowledge map construction device provided in the embodiment of the present application can implement various processes implemented in the method embodiments in FIG. 1 to FIG. 3 , and will not be repeated here to avoid repetition.

Based on the same technical idea, the embodiment of the present application also provides an electronic device, which is used to implement the above-mentioned method for constructing the knowledge graph. FIG. 5 is a schematic structural diagram of an electronic device implementing various embodiments of the present application. Electronic equipment may have relatively large differences due to different configurations or performances, and may include a processor (processor) 610, a communication interface (Communications Interface) 620, a memory (memory) 630, and a communication bus 640, wherein the processor 610, the communication interface At 620 , the memories 630 communicate with each other through the communication bus 640 . The processor 610 can invoke a computer program stored on the memory 630 and executable on the processor 610 to perform the following steps:

Obtain the ontology, extraction rules and mapping rules corresponding to the target knowledge graph. The ontology includes the definition of multiple entities and the definition of related information of multiple entities. The extraction rule is the rule for extracting data from the source data corresponding to the entity. The mapping rule is Map the fields in the source data corresponding to the entity to the rules of the ontology; receive the query request, which includes the first entity, and determine whether the first entity has been constructed in the graph database corresponding to the target knowledge graph according to the query request, and query The request is used to query the first relevant knowledge of the first entity in the graph database; if the first entity has not been constructed, extract the source data corresponding to the first entity according to ontology, extraction rules and mapping rules, and determine the first 1. Relevant knowledge; import the first relevant knowledge into the graph database to construct the target knowledge graph.

Using the electronic device provided by the embodiment of the present invention, by acquiring the ontology, extraction rules, and mapping rules corresponding to the target knowledge map, the ontology includes definitions of multiple entities and definitions of related information of multiple entities, and the extraction rules are from entities corresponding to The rules for extracting data in the source data, the mapping rules are the rules for mapping the fields in the source data corresponding to the entities to the ontology; receive the query request, the query request includes the first entity, and determine the first entity in the target knowledge map according to the query request Whether the corresponding graph database has been constructed, the query request is used to query the first related knowledge of the first entity in the graph database; if the first entity has not been constructed, according to the ontology, extraction rules and mapping rules, the second The source data corresponding to an entity is extracted to determine the first relevant knowledge; the first relevant knowledge is imported into the graph database to construct the target knowledge graph. It can be seen that, compared to directly extracting a large amount of knowledge from massive data and constructing a knowledge map, knowledge that is hardly searchable takes up a lot of space resources, and it is not easy to construct knowledge maps multiple times. The technical solution can extract the source data corresponding to the first entity according to the user's query request, and the amount of extracted data is smaller, which saves the time for knowledge extraction. Correspondingly, when storing the knowledge, it occupies relatively less space, and Due to the acquisition of ontology, extraction rules and mapping rules, according to each query request, the ontology, extraction rules and mapping rules can be used to extract the knowledge required for each query request, thereby realizing multiple constructions of knowledge graphs (dynamic construction) , which reduces the time and storage space occupied by building knowledge graphs under massive data.

For specific execution steps, please refer to the various steps in the above embodiment of the method for constructing the knowledge map, and can achieve the same technical effect. To avoid repetition, details will not be repeated here.

It should be noted that the electronic device in the embodiment of the present application includes: a server, a terminal or other devices except the terminal.

The above electronic device structure does not constitute a limitation to the electronic device, and the electronic device may include more or fewer components than shown in the illustration, or combine certain components, or arrange different components, for example, an input unit may include a graphics processor (Graphics Processing Unit, GPU) and a microphone, the display unit can be configured with a display panel in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit includes at least one of a touch panel and other input devices. A touch panel is also called a touch screen. Other input devices may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, and joysticks, which will not be repeated here.

The memory can be used to store software programs as well as various data. The memory may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, application programs or instructions required by at least one function (such as sound playback function, image playback function, etc.) etc. Also, memory may include volatile memory or nonvolatile memory, or memory may include both volatile and nonvolatile memory. Among them, the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electronically programmable Erase Programmable Read-Only Memory (Electrically EPROM, EEPROM) or Flash. Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (Synchlink DRAM, SLDRAM) and Direct Memory Bus Random Access Memory (Direct Rambus RAM, DRRAM).

The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly handles operations related to the operating system, user interface, and application programs, and the modem The tone processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the foregoing modem processor may not be integrated into the processor.

The embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, each process of the above-mentioned knowledge map construction method embodiment is implemented, and can To achieve the same technical effect, in order to avoid repetition, no more details are given here.

Wherein, the processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer readable storage medium, such as computer read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.

The embodiment of the present application further provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to realize the implementation of the above knowledge map construction method Each process of the example, and can achieve the same technical effect, in order to avoid repetition, will not repeat them here.

It should be understood that the chips mentioned in the embodiments of the present application may also be called system-on-chip, system-on-chip, system-on-a-chip, or system-on-a-chip.

It should be noted that, in this document, the term "comprising", "comprising" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, It also includes other elements not expressly listed, or elements inherent in the process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions are performed, for example, the described methods may be performed in an order different from that described, and various steps may also be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product in essence or the part that contributes to the prior art, and the computer software product is stored in a storage medium (such as ROM/RAM, disk, CD) contains several instructions to enable a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in various embodiments of the present application.

The embodiments of the present application have been described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Under the inspiration of this application, without departing from the purpose of this application and the scope of protection of the claims, many forms can also be made, all of which belong to the protection of this application.

Claims (10)

1. The method for constructing the knowledge graph is characterized by comprising the following steps of:

acquiring an ontology, an extraction rule and a mapping rule corresponding to a target knowledge graph; the entity comprises definitions of a plurality of entities and definitions of related information of the entities, the extraction rule is a rule for extracting data from source data corresponding to the entities, and the mapping rule is a rule for mapping fields in the source data corresponding to the entities to the entity;

receiving a query request, wherein the query request comprises a first entity, and determining whether the first entity is constructed in a graph database corresponding to the target knowledge graph according to the query request, wherein the query request is used for querying first related knowledge of the first entity in the graph database;

under the condition that the first entity is not constructed, extracting source data corresponding to the first entity according to the ontology, the extraction rule and the mapping rule, and determining the first related knowledge;

and importing the first related knowledge into the graph database, and constructing the target knowledge graph.

2. The knowledge-graph construction method according to claim 1, characterized in that the method further comprises:

Under the condition that the source data corresponding to the second entity generates the update, determining whether the second entity is constructed in the graph database according to the updated source data;

under the condition that the second entity is constructed, extracting source data corresponding to the second entity according to the ontology, the extraction rule and the mapping rule, and determining second related knowledge of the second entity;

and importing the second related knowledge into the graph database, and updating the target knowledge graph.

3. The method for constructing a knowledge graph according to claim 1 or 2, wherein the extracting the source data corresponding to the first entity to determine the first related knowledge includes:

obtaining an index table constructed for source data corresponding to the entities, wherein the index table comprises entity names and entity unique code IDs;

according to the first entity, determining an entity ID corresponding to the entity name by searching the entity name of the entity in the index table;

according to the entity ID, one or more of the following information corresponding to the first entity is determined: attributes of entities, related entities, relationship information between entities.

4. The knowledge-graph construction method according to claim 1, further comprising, after said determining whether the first entity has been constructed in the graph database according to the query request:

and acquiring the first related knowledge of the first entity, and displaying a target knowledge graph corresponding to the first related knowledge.

5. The knowledge-graph construction method according to claim 2, characterized in that the method further comprises: determining, by a Bloom Filter, whether a target entity has been built in the graph database, the target entity being the first entity or the second entity that is determined whether or not it has been built in the graph database.

6. The knowledge-graph construction method according to claim 5, wherein the determining, by a bloom filter, whether the first entity or the second entity has been constructed in the graph database, comprises:

acquiring a bit array with the length of n, wherein the value of each bit in the bit array is zero;

obtaining k hash functions and constructed entities, wherein k is smaller than n;

determining k first bit positions for each constructed entity according to the k hash functions and assigning the k first bit positions as one;

Determining k second bit positions of each target entity according to the k hash functions;

and determining whether the target entity is constructed according to the values of the k second bit positions.

7. The knowledge graph construction device is characterized by comprising:

the ontology preset module is used for acquiring an ontology, an extraction rule and a mapping rule corresponding to the target knowledge graph; the entity comprises definitions of a plurality of entities and definitions of related information of the entities, the extraction rule is a rule for extracting data from source data corresponding to the entities, and the mapping rule is a rule for mapping fields in the source data corresponding to the entities to the entity;

the first determining module is used for receiving a query request, wherein the query request comprises a first entity, determining whether the first entity is constructed in a graph database corresponding to the target knowledge graph according to the query request, and the query request is used for querying first related knowledge of the first entity in the graph database;

the second determining module is configured to, when the first entity is not constructed, extract source data corresponding to the first entity according to the ontology, the extraction rule and the mapping rule, and determine the first related knowledge;

And the construction module is used for importing the first related knowledge into the graph database and constructing the target knowledge graph.

8. The apparatus as recited in claim 7, further comprising:

a third determining module, configured to determine, according to updated source data, whether a second entity has been built in the graph database, in a case where the source data corresponding to the second entity generates an update;

a fourth determining module, configured to, when the second entity has been constructed, extract, according to the ontology, the extraction rule, and the mapping rule, source data corresponding to the second entity, and determine second related knowledge of the second entity;

and the updating module is used for importing the second related knowledge into the graph database and updating the target knowledge graph.

9. An electronic device comprising a processor, a memory and a program or instruction stored on the memory and executable on the processor, which when executed by the processor, implements the steps of the knowledge-graph construction method of any one of claims 1-6.

10. A readable storage medium, wherein a program or instructions are stored on the readable storage medium, which when executed by a processor, implement the steps of the knowledge-graph construction method according to any one of claims 1-6.

Images