Data storage method, data query method, server and storage medium

Through the data storage solution of the knowledge graph, the problem of low performance and poor scalability of relational databases during correlation queries is solved, and high-performance and fast correlation query and data storage scalability are achieved.

CN114461781BActive Publication Date: 2024-11-26ALIBABA CLOUD COMPUTING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111659502.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-30
Publication Date
2024-11-26
Estimated Expiration
2041-12-30

AI Technical Summary

Technical Problem

Relational databases have low performance and poor scalability in association queries, making it difficult to meet the needs of high-performance and fast association queries.

Method used

Using the data storage scheme of the knowledge graph, the relationship between conditional entities, conditional indicator entities and object indicator entities is constructed to achieve the data association between the target and the object through the method of nodes representing entities and edges representing relationships between entities.

Benefits of technology

It improves the performance and scalability of associated queries, provides a more flexible and efficient data storage solution, and is suitable for high-frequency associated queries scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114461781B_ABST
    Figure CN114461781B_ABST
Patent Text Reader

Abstract

The embodiment of the present application provides a data storage method, a data query method, a server and a storage medium, wherein the data storage method includes: obtaining a target text defining a target and object data of an object; identifying at least from the target text the conditions corresponding to the target and the condition indicators associated with the conditions; the condition indicators define associated object indicators associated with the object and the conditions, and the rules that the associated object indicators must satisfy; obtaining object indicators from the object data; constructing at least in the knowledge graph a condition entity corresponding to the condition, a condition indicator entity corresponding to the condition indicator, and an object indicator entity corresponding to the object indicator; constructing in the knowledge graph an association relationship between the condition entity and the condition indicator entity, and, based on the associated object indicators defined by the condition indicators, constructing an association relationship between the condition indicator entity and the object indicator entity. The embodiment of the present application can improve the scalability of data storage and provide support for high-performance association queries.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of data processing technology, and specifically to a data storage method, a data query method, a server, and a storage medium. Background Art

[0002] At present, data is mainly stored in relational databases, which are databases that store data in relational models. Although relational databases have many advantages, they manage data in rows and columns of tables, which makes them difficult to apply to scenarios of associative queries. Therefore, how to provide a new data storage solution to support associative queries has become a technical problem that technicians in this field need to solve urgently. Summary of the invention

[0003] In view of this, embodiments of the present application provide a data storage method, a data query method, a server, and a storage medium to provide support for associated queries.

[0004] To achieve the above objectives, the embodiments of the present application provide the following technical solutions.

[0005] In a first aspect, an embodiment of the present application provides a data storage method, including:

[0006] Get the target text defining the target and the object data of the object;

[0007] At least identifying the condition corresponding to the target and the condition index associated with the condition from the target text; the condition index defines an associated object index associated with the object and the condition, and a rule that the associated object index needs to satisfy;

[0008] Acquire an object index of the object from the object data;

[0009] At least constructing a condition entity corresponding to the condition, a condition indicator entity corresponding to the condition indicator, and an object indicator entity corresponding to the object indicator in the knowledge graph;

[0010] An association relationship between a conditional entity and a conditional indicator entity is constructed in a knowledge graph, and an association relationship between the conditional indicator entity and an object indicator entity is constructed based on an associated object indicator defined by the conditional indicator.

[0011] In a second aspect, an embodiment of the present application provides a data query method, including:

[0012] Get data query request;

[0013] Retrieving a knowledge graph, wherein the knowledge graph stores data based on the data storage method described in the first aspect above;

[0014] Based on the knowledge graph, a data query result is determined.

[0015] In a third aspect, an embodiment of the present application provides a data query method, including:

[0016] Displaying an object indicator submission page, wherein the object indicator submission page displays multiple object indicator items;

[0017] Determine an object index corresponding to the first object in the plurality of object index items, and send a data query request based on the object index corresponding to the plurality of object index items;

[0018] Obtain data query results determined based on the knowledge graph, and display the data query results on a data result page; wherein the knowledge graph stores data based on the data storage method described in the first aspect above.

[0019] In a fourth aspect, an embodiment of the present application provides a server comprising at least one memory and at least one processor, wherein the memory stores one or more computer executable instructions, and the processor calls the one or more computer executable instructions to execute the data storage method as described in the first aspect above, and / or the data query method as described in the second aspect above.

[0020] In a fifth aspect, an embodiment of the present application provides a storage medium, which stores one or more computer-executable instructions, and when the one or more computer-executable instructions are executed, they implement the data storage method as described in the first aspect above, or the data query method as described in the second aspect above, or the data query method as described in the third aspect above.

[0021] In a sixth aspect, an embodiment of the present application provides a computer program, which, when executed, implements the data storage method as described in the first aspect above, or the data query method as described in the second aspect above, or the data query method as described in the third aspect above.

[0022] It can be seen that in the embodiment of the present application, the knowledge graph is composed of at least entities on the target side, entities on the object side, and the association relationship between entities, wherein the entities on the target side include at least condition entities and condition indicator entities, and the entities on the object side include at least object indicator entities. The embodiment of the present application can use the target text of the target and the object data of the object as the data source for the knowledge graph to store the target and object data, so as to at least identify the conditions corresponding to the target and the condition indicators associated with the conditions from the target text, wherein the condition indicators define the associated object indicators that associate the object with the conditions, and the rules that the associated object indicators must satisfy; and the object indicators of the object are obtained from the object data. Based on the knowledge graph using nodes to represent entities, the embodiment of the present application can at least construct the condition entity corresponding to the condition, the condition indicator entity corresponding to the condition indicator, and the object indicator entity corresponding to the object indicator in the knowledge graph, so as to realize the storage of the information of the entity on the target side and the entity on the object side in the knowledge graph; based on the knowledge graph using edges to represent the relationship between entities, the embodiment of the present application can construct the association relationship between the condition entity and the condition indicator entity in the knowledge graph, and, based on the associated object indicator defined by the condition indicator, construct the association relationship between the condition indicator entity and the object indicator entity, so as to realize the association of the information of the entity on the target side and the object side stored in the knowledge graph at least through the association relationship between the condition indicator entity and the object indicator entity, so as to realize the association of the data of the target and the object. It can be seen that the embodiment of the present application stores and describes the data of the target and the object by using the knowledge graph to represent entities with nodes and the edges to represent the relationship between entities, and can use the flexible model definition capability of the knowledge graph to realize the high scalability of data storage, and establish relationships at the node level of the entity, rather than through the table-level relationship of the relational database, to provide support for realizing high-performance and fast association queries. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0024] Figure 1 This is a system architecture block diagram of an embodiment of the present application.

[0025] Figure 2 A flowchart of a data storage method provided in an embodiment of the present application.

[0026] Figure 3A An example diagram of the knowledge graph provided for an embodiment of the present application.

[0027] Figure 3B Another example diagram of the knowledge graph provided in an embodiment of the present application.

[0028] Figure 3C is an example graph of the target text.

[0029] Figure 4 Another example diagram of the knowledge graph provided in the embodiment of the present application.

[0030] Figure 5 Another flow chart of the data storage method provided in an embodiment of the present application.

[0031] Fig. 6A Another example diagram of the knowledge graph provided in the embodiment of the present application.

[0032] Figure 6B Yet another example diagram of the knowledge graph provided in an embodiment of the present application.

[0033] Figure 7 Another flow chart of the data storage method provided in the embodiment of the present application.

[0034] Fig. 8A Yet another example diagram of the knowledge graph provided in an embodiment of the present application.

[0035] Figure 8B Another example diagram of the knowledge graph provided in an embodiment of the present application.

[0036] Fig. 9 A flowchart of a data query method provided in an embodiment of the present application.

[0037] Fig. 10A Another flow chart of the data query method provided in an embodiment of the present application.

[0038] Fig. 10B This is an example diagram of the object indicator submission page provided in an embodiment of the present application.

[0039] Fig. 10C This is an example diagram of a data result page provided in an embodiment of the present application.

[0040] Fig.11 A block diagram of a data storage device provided in an embodiment of the present application.

[0041] Fig.12 A block diagram of the server. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] In the fields of data intelligent recommendation and data intelligent analysis, it is necessary to frequently perform related queries. Related queries refer to queries of related data. For example, in government scenarios, related policies can be queried through enterprises, related enterprises can be queried through enterprises, and related policies can be queried through policies. In relational databases, relational databases manage data in the form of rows and columns of tables. This means that when using relational databases for related queries, relational databases need to perform multiple table join operations, which undoubtedly seriously affects the performance of related queries and causes excessive latency in related queries. Therefore, it is urgent to provide a new data storage solution to provide support for high-performance and fast related queries. Furthermore, relational databases manage data in the form of rows and columns of tables, which also leads to poor scalability of data storage. Therefore, new data storage solutions must also meet the high scalability requirements of data storage.

[0044] In the embodiments of the present application, the association query mainly involves the association query between the target and the object; the target can be the action guidelines and requirements formulated to achieve a specific purpose, such as policies, selection rules, assessment rules, etc.; the object can be any type of subject, such as enterprises, individual users, institutions, schools, families, etc. Further, the association query of the embodiments of the present application can also involve the association query between objects (such as the association query between enterprises) and the association query between targets (such as the association query between policies).

[0045] To facilitate the association query between targets and objects, the embodiments of the present application can store the data of targets and objects through the knowledge graph, that is, the knowledge graph uses nodes to represent entities and edges to represent the association relationship between entities to describe the data of targets and objects, thereby utilizing the knowledge graph's flexible schema definition capabilities to achieve high scalability of data storage, and utilizes the association relationship established at the node level of the knowledge graph, rather than the table-level relationship of the relational database, to support high-performance and fast association queries.

[0046] It should be noted that the knowledge graph is a knowledge management storage structure based on a graph data structure. In the knowledge graph, nodes represent entities, and edges between nodes represent the relationships between entities. Entities can be regarded as formal representations of knowledge concepts and knowledge relationships.

[0047] As an optional implementation, Figure 1 The system architecture block diagram of the embodiment of the present application is shown as an example. Figure 1 As shown, the system architecture may include: a terminal device 11 and a server 12. The terminal device 11 may be a user device used by a user, such as a smartphone, a laptop computer, or other user device used by a user. The server 12 may be a server device arranged on the network side, and may be implemented by a single server or a server cluster; in an optional implementation, the server 12 may be a cloud server based on cloud computing. In an embodiment of the present application, the server may store at least the data of the target and the object in the form of a knowledge graph. When the terminal device 11 sends a data query request, the server may perform an association query on the data based on the knowledge graph, thereby feeding back the data query result to the terminal device 11.

[0048] In the embodiment of the present application, in order to realize the association query between the target and the object based on the knowledge graph, the server needs to store at least the data of the target and the target object in the form of the knowledge graph. Based on this, the embodiment of the present application provides a new data storage solution to store at least the data of the target and the object through the knowledge graph, to provide support for high-performance and fast association query, and to further improve the scalability of data storage.

[0049] As an optional implementation, Figure 2 An optional flow chart of the data storage method provided in the embodiment of the present application is shown as an example. The method flow can be applied to the server, refer to Figure 2 The method flow may include the following steps.

[0050] In step S210, a target text defining a target and object data of an object are acquired.

[0051] Based on the embodiment of the present application, the knowledge graph is used to store at least the data of the target and the object. The target text can be regarded as the data source of the data of the target stored in the knowledge graph, and the object data can be regarded as the data source of the data of the object stored in the knowledge graph. In some embodiments, the target text of the target can describe the full text content of the target, and the target text is, for example, a policy text. As an optional implementation, the embodiment of the present application can obtain the target text of the target from a data source such as a website, a database, etc. of an institution that publishes the target (e.g., a government agency that publishes the policy). In some embodiments, the object data of an object can be data describing the situation of the object, such as enterprise data.

[0052] In step S211, at least the condition corresponding to the target and the condition indicator associated with the condition are identified from the target text, wherein the condition indicator defines an associated object indicator that associates the object with the condition and a rule that the associated object indicator needs to satisfy.

[0053] In step S212, an object index of the object is acquired from the object data.

[0054] In step S213, at least a condition entity corresponding to the condition, a condition indicator entity corresponding to the condition indicator, and an object indicator entity corresponding to the object indicator are constructed in the knowledge graph.

[0055] Since knowledge graphs use nodes to represent entities, to store target and object data in knowledge graphs, it is necessary to construct target-side entities and object-side entities in the form of nodes in the knowledge graph. For example, to store policy and enterprise data in the knowledge graph, it is necessary to construct policy-side entities and enterprise-side entities in the form of nodes in the knowledge graph.

[0056] In an embodiment of the present application, the entity on the target side may include at least a condition entity and a condition indicator entity; therefore, the embodiment of the present application may identify the condition corresponding to the target and the condition indicator associated with the condition from the target text, thereby constructing the condition entity corresponding to the condition and the condition indicator entity corresponding to the condition indicator with nodes in the knowledge graph. As an optional implementation, the embodiment of the present application may perform intelligent recognition on the target text to identify the condition corresponding to the target and the condition indicator associated with the condition from the target text.

[0057] It should be noted that in an embodiment of the present application, the target text of the target can define the conditions for the object to meet the needs of the target (for example, the policy text can define the conditions for the enterprise to meet the needs of the policy), and the conditions are associated with conditional indicators, and the conditional indicators can define object indicators associated with the conditions in the object (called associated object indicators), and the rules that the associated object indicators must meet. Among them, the object indicator can be an indicator data indicating the specific situation of the object, such as the annual sales, annual tax amount, number of employees and other enterprise indicators of the enterprise. In an example, the policy text can define the conditions for the object to meet the needs of the policy, and the conditional indicators can be associated with the conditions, and the conditional indicators can define the associated enterprise indicators associated with the conditions in the enterprise, and the rules that the associated enterprise indicators must meet. As an optional implementation, the conditional indicators can define the rules that the associated object indicators must meet through regular expressions, for example, regular expressions can define formulas that the associated object indicators must meet, such as the annual sales of the enterprise> 1 million, the number of employees> 50 people, etc.

[0058] In an embodiment of the present application, the entity on the object side may include at least an object indicator entity; therefore, the embodiment of the present application needs to obtain the object indicator of the object from the object data of the object (for example, obtaining the enterprise indicator of the enterprise from the enterprise data of the enterprise), so as to construct the object indicator entity corresponding to the object indicator with nodes in the knowledge graph (for example, construct the enterprise indicator entity corresponding to the enterprise indicator with nodes in the knowledge graph).

[0059] In step S214, an association relationship between a condition entity and a condition indicator entity is constructed in the knowledge graph, and an association relationship between the condition indicator entity and an object indicator entity is constructed based on an associated object indicator defined by the condition indicator.

[0060] Based on the knowledge graph using edges to represent the relationship between entities, the embodiment of the present application needs to use edges to construct the association relationship between conditional entities and conditional indicator entities in the knowledge graph, and use edges to construct the association relationship between conditional indicator entities and object indicator entities, so as to achieve the association of the data of targets and objects stored in the knowledge graph through the association relationship between the conditional indicator entities on the target side and the object indicator entities on the object side.

[0061] In some embodiments, the embodiments of the present application may construct an association relationship between a condition entity and a condition indicator entity in a knowledge graph based on an associated condition indicator of a condition. Furthermore, based on the associated condition of the condition indicator entity, an association relationship between a condition indicator entity and a condition indicator entity may be constructed in a knowledge graph with an edge to realize the construction of an association relationship between a condition entity and a condition indicator entity in the knowledge graph. As an optional implementation, the associated condition indicator of a condition may be recorded in the attributes of the condition entity, and the associated condition of the condition indicator may be recorded in the attributes of the condition indicator entity.

[0062] In some embodiments, the embodiments of the present application may construct an association relationship between a condition indicator entity and an object indicator entity in a knowledge graph based on an associated object indicator of a condition indicator; for example, through an associated object indicator defined by a condition indicator, an object indicator entity corresponding to the associated object indicator is found in the object indicator entity, thereby constructing an association relationship between the condition indicator entity and the object indicator entity with an edge. Further, the embodiments of the present application may construct an association relationship between an object indicator entity and a condition indicator entity in a knowledge graph based on an associated condition indicator of an object indicator; for example, through an associated condition indicator defined by an object indicator, a condition indicator entity corresponding to the associated condition indicator is found in the condition indicator entity, thereby constructing an association relationship between an object indicator entity and the condition indicator entity with an edge. As an optional implementation, the associated object indicator of a condition indicator may be recorded in the attributes of a condition indicator entity, and the associated condition indicator of an object indicator may be recorded in the attributes of an object indicator entity.

[0063] In some further embodiments, the embodiments of the present application may save entity information through node attributes, for example, save entity attributes through node attributes.

[0064] In one example, Figure 3A An optional example diagram of the knowledge graph provided in the embodiment of the present application is shown as an example. Figure 3AAs shown, after identifying the conditions corresponding to the target and the condition indicators associated with the conditions from the target text, the embodiment of the present application can construct the condition entity and the condition indicator entity with nodes in the knowledge graph, and through the associated condition indicators of the condition entity and the associated conditions of the condition indicator entity, the association relationship between the condition entity and the condition indicator entity can be constructed with edges. At the same time, after obtaining the object indicator from the object data of the object, the embodiment of the present application can construct the object indicator entity with nodes in the knowledge graph, and through the associated object indicators of the condition indicator entity and the associated condition indicators of the object indicator entity, the association relationship between the condition indicator entity and the object indicator entity can be constructed with edges, thereby realizing the storage of the target and object data in the knowledge graph.

[0065] It can be seen that in the embodiment of the present application, the knowledge graph is composed of at least entities on the target side, entities on the object side, and the association relationship between entities, wherein the entities on the target side include at least condition entities and condition indicator entities, and the entities on the object side include at least object indicator entities. The embodiment of the present application can use the target text of the target and the object data of the object as the data source for the knowledge graph to store the target and object data, so as to at least identify the conditions corresponding to the target and the condition indicators associated with the conditions from the target text, wherein the condition indicators define the associated object indicators that associate the object with the conditions, and the rules that the associated object indicators must satisfy; and the object indicators of the object are obtained from the object data. Based on the knowledge graph using nodes to represent entities, the embodiment of the present application can at least construct the condition entity corresponding to the condition, the condition indicator entity corresponding to the condition indicator, and the object indicator entity corresponding to the object indicator in the knowledge graph, so as to realize the storage of the information of the entity on the target side and the entity on the object side in the knowledge graph; based on the knowledge graph using edges to represent the relationship between entities, the embodiment of the present application can construct the association relationship between the condition entity and the condition indicator entity in the knowledge graph, and, based on the associated object indicator defined by the condition indicator, construct the association relationship between the condition indicator entity and the object indicator entity, so as to realize the association of the information of the entity on the target side and the object side stored in the knowledge graph at least through the association relationship between the condition indicator entity and the object indicator entity, so as to realize the association of the data of the target and the object. It can be seen that the embodiment of the present application stores and describes the data of the target and the object by using the knowledge graph to represent entities with nodes and the edges to represent the relationship between entities, and can use the flexible model definition capability of the knowledge graph to realize the high scalability of data storage, and establish relationships at the node level of the entity, rather than through the table-level relationship of the relational database, to provide support for realizing high-performance and fast association queries.

[0066] In some further embodiments, Figure 3B Another optional example diagram of the knowledge graph provided in the embodiment of the present application is shown exemplarily. Figure 3A and Figure 3BAs shown, in Figure 3B In the knowledge graph shown, entities on the target side may also include: target entity, sub-target entity and target label entity. In an embodiment of the present application, the target text of the target may describe one or more sub-targets, and a sub-target may have one or more target labels; for example, the realization of a target contains multiple sub-targets, and the content under the sub-target may be refined and summarized into one or more content refinement points, and the label information of each content refinement point may be used as the target label under the sub-target; and, one or more conditions may be associated with the content refinement point corresponding to each target label, and one or more conditional indicators may be associated with each condition. As an example, Figure 3C An example diagram of the target text is shown as follows: Figure 3C As shown, the target text can record the content of at least one sub-target contained in the target, each sub-target content is associated with at least one target tag, each target tag is associated with at least one condition, and each condition is associated with at least one condition indicator.

[0067] In some embodiments, the embodiments of the present application can identify sub-goals, target tags associated with sub-goals, conditions associated with target tags, and condition indicators associated with conditions from the target text of the target, thereby constructing the target entity corresponding to the target, the sub-goal entity corresponding to the sub-goal, the target tag entity corresponding to the target tag, the condition entity corresponding to the condition, and the condition indicator entity corresponding to the condition indicator in the knowledge graph with nodes. Furthermore, the embodiments of the present application can construct the association relationship between the target entity and the sub-goal entity, the association relationship between the sub-goal entity and the target tag entity, the association relationship between the target tag entity and the condition entity, and the association relationship between the condition entity and the condition indicator entity in the knowledge graph with edges.

[0068] As an optional implementation, combine Figure 3B As shown, the embodiment of the present application can construct an association relationship between a target entity and a subtarget entity based on the subtargets contained in the target and the target to which the subtarget belongs; wherein, the subtargets contained in the target can be recorded in the attributes of the target entity, and the target to which the subtarget belongs can be recorded in the attributes of the subtarget entity. Further, based on the associated target tag of the subtarget and the associated subtarget of the target tag, an association relationship between the subtarget entity and the target tag entity is constructed; wherein, the associated target tag of the subtarget can be recorded in the attributes of the subtarget entity, and the associated subtarget of the target tag can be recorded in the attributes of the target tag entity. Further, based on the conditions determined by the target tag and the target tag to which the conditions belong, an association relationship between the target tag entity and the condition entity is constructed; wherein, the conditions determined by the target tag can be recorded in the attributes of the target tag entity, and the target tag to which the conditions belong can be recorded in the attributes of the condition entity. The association relationship between the condition entity and the condition indicator entity can be constructed with reference to the description of the corresponding part of the previous text.

[0069] In some embodiments, the embodiments of the present application can identify sub-goals, target tags, conditions, condition indicators and other information from the target text by intelligently identifying the target text. As an optional implementation, the present application can encode the target text to obtain the text features of the target text; extract the text features corresponding to the sub-goals from the text features; decode the text features corresponding to the sub-goals, identify the text phrases of the tag category from the decoding results, standardize the text phrases and refine them into target tags; and identify data phrases from the decoding results, and extract the conditions under the target tags and the condition indicators under the conditions from the data phrases.

[0070] As an optional implementation, the above-mentioned process of intelligent recognition of the target text can be implemented based on an intelligent recognition model, such as an intelligent recognition model based on BERT (Bidirectional Encoder Representation from Transformers, bidirectional encoding representation based on transformers)-LSTM (Long Short-Term Memory, long short-term memory network)-CRF (Conditional Random Field, conditional random field). In a possible implementation, the embodiment of the present application can use a large-scale text corpus (such as a large-scale target text) to pre-train a BERT model, so as to encode the target text using the BERT model to obtain the text features of the target text; then use a recurrent neural network in the form of LSTM to extract text features, and extract text features corresponding to the sub-targets. Then, the text features corresponding to the sub-targets are decoded (for example, the text features are decoded based on CRF), so as to identify text phrases of at least one label category from the decoding results, and the embodiment of the present application can standardize the identified text phrases and refine them into target labels. After identifying a text phrase of at least one tag category, the embodiment of the present application does not directly refine the text phrase into a target tag, but further standardizes the text phrase and extracts the target tag based on the standardized result. This is mainly to solve the problem of data redundancy. For example, "Computer Science and Technology R&D Center" and "Computer Science and Technology Research Center" can be combined into the standardized "Computer Science and Technology R&D Center". In the optional implementation of standardizing text phrases, considering that the length of the text phrases is short, the embodiment of the present application can use a string similarity algorithm to cluster the text phrases.

[0071] Furthermore, after decoding the text features corresponding to the sub-goals, the embodiments of the present application can identify data phrases from the decoding results, thereby extracting the conditions under the target label and the conditional indicators under the conditions from the data phrases. As an optional implementation, the data phrases may include numerical phrases and requirement language phrases; numerical phrases are generally attribute values ​​of annual report attributes, such as "annual R&D expenses reach 5 million", so the characteristic pattern of the numerical phrase is "numerical value" and "unit". Using the characteristic pattern of the numerical phrase, the embodiments of the present application can extract features of the numerical phrase through the feature template to obtain the conditions and conditional indicators of the numerical phrase; for example, for the numerical phrase "annual R&D expenses reach 5 million" in the above example, the embodiments of the present application can extract the conditions "annual R&D expenses reach 5 million" and the conditional indicator "annual R&D expenses", and define the rule expression "annual R&D expenses ≥ 5 million" in the conditional indicator. As an optional implementation, the requirement language phrase can be a phrase that defines the requirement, such as "the company needs to go public", etc. The embodiment of the present application can extract the requirement language phrase from the decoding results of the text features through the requirement words (such as "should", "need", "satisfy", etc.), and intelligently identify the conditions and conditional indicators under the conditions from the requirement language phrase.

[0072] In some further embodiments, Figure 4 Another optional example diagram of the knowledge graph provided by the embodiment of the present application is shown as an example. Figure 3B and Figure 4 As shown, Figure 4 In the knowledge graph shown, the entity on the object side may also include an object entity. The embodiment of the present application may construct an object entity corresponding to the object with a node in the knowledge graph for an object (e.g., an enterprise), and save the entity information with node attributes based on the knowledge graph. The embodiment of the present application may save the attributes of the object entity and the object indicator entity through the node attributes, and build the association relationship between the object entity, the object indicator entity and the node attributes.

[0073] It should be noted that object indicators can be a special attribute, and object indicators can inherit from the built-in attribute types (type, property, etc.) of the knowledge graph. In possible implementations, object indicators can describe the attribute content of the object itself, but there are also differences between object indicators and object attributes: object indicators can be used as elements of conditional indicators. For example, the annual sales of an enterprise is an enterprise indicator. This enterprise indicator can correspond to a specific conditional indicator in the condition that the annual sales reach a certain value. For example, the enterprise indicator of annual sales can correspond to the conditional indicator of "annual sales > 1 million" in the condition. Therefore, object indicators can be used as elements of conditional indicators.

[0074] Further, combined with Figure 4As shown, the embodiment of the present application can also construct an association relationship between an object entity and a target tag entity in the knowledge graph. For example, the embodiment of the present application can construct an association relationship between an object entity and a target tag entity based on the associated target tag of the object entity.

[0075] The knowledge graph constructed in the above-mentioned embodiment of the present application is implemented based on existing data, and the existing data may not be able to fully and completely represent the relationship between the entities on the target side and the entities on the object side. This may lead to missing relationships between the entities on the target side and the entities on the object side. Therefore, the embodiment of the present application needs to supplement the missing relationships between the entities on the target side and the entities on the object side. The embodiment of the present application can use missing paths to supplement the missing relationships between the entities on the target side and the entities on the object side. In some embodiments, the embodiment of the present application can determine the missing paths between entities in the knowledge graph, combined with Figure 4 As shown, the missing path includes the missing path of the object entity and the conditional entity. Furthermore, the missing path may also include the missing path of the object entity and the target label entity, and / or the target entity, or the missing path of the missing object indicator entity and the conditional indicator entity. The missing path may be in the form of an entity in the knowledge graph. The embodiment of the present application may define the association relationship between the missing path and the entity on the object side, and the association relationship between the missing path and the entity on the target side, so as to supplement the association relationship between the entity on the object side and the entity on the target side.

[0076] In some further embodiments, in combination Figure 4 As shown, the embodiment of the present application may use the missing path and the condition indicator entity as a connection path (for example, the connection path includes the missing path and the condition indicator entity), and determine the confidence of the connection path. The confidence can represent the degree of credibility of the association relationship between the entities connected by the connection path.

[0077] In some further embodiments, in combination Figure 4 As shown, the condition indicator entity and the missing path can be CVT (Compound Value Type, composite object type) entities. CVT entities allow multiple entities to be bound together. Furthermore, in addition to ordinary attributes, object entities can also have multiple composite attributes, which can be represented by multiple CVT entities, such as combining Figure 4 As shown, the object entity may have composite attributes F1 and F2 in addition to common attributes P1 and P2. The composite attributes F1 and F2 may be represented by CVT entities respectively.

[0078] It should be noted that Figure 3B and Figure 4 The knowledge graph shown is only Figure 3AAn optional extension implementation based on the knowledge graph shown, based on Figure 2 The principle of the method flow shown and Figure 3A The basic structure of the knowledge graph shown in the embodiment of the present application can fully adapt and expand the knowledge graph structure according to the specific goals and objects, which should also be regarded as within the scope of the embodiment of the present application.

[0079] The data storage method provided in the embodiment of the present application stores and describes the data of targets and objects by using nodes to represent entities and edges to represent the relationships between entities through a knowledge graph. The flexible schema definition capability of the knowledge graph can be used to achieve high scalability of data storage, and relationships can be established at the node level of entities, providing support for high-performance and fast associative queries.

[0080] In an optional application example, the data storage method provided in the embodiment of the present application can be applied to store enterprise and policy data to support related queries between enterprises and policies. It should be noted that in the government scenario, government agencies have issued many policies that benefit enterprises, and these policies and the company's own information need to be digitized to facilitate the query of policy and enterprise data. In this context, if a relational database is used to store policy and enterprise data, since the relational database (such as MySQL) manages enterprise and policy data by creating tables, this may cause the following problems:

[0081] Poor flexibility. A relational database is a strong-mode database. When the mode is changed, columns need to be added or deleted. During this process, the table will be locked, resulting in service interruption.

[0082] The relationship between any two types of data, such as enterprise basic information, enterprise extended information, policy highlights, conditions, and policy tags, requires the maintenance of relationship tables, which will lead to an increase in the number of relationship tables.

[0083] Due to the requirements for performance and associated queries, each type of enterprise extended information (such as annual sales, industry information, key personnel information, and associated tags) requires a specific table to be created, which has high maintenance costs;

[0084] The performance of associated queries is poor. The overhead of each joint table query is the Cartesian product of the data in the two tables. This is not suitable for real-time recommendation scenarios that require low latency.

[0085] At the same time, due to the differences in policy content and requirements for enterprises in various regions, if a strong-mode relational database is used for modeling, changes in the model will bring greater overhead to the relational database for storing data.

[0086] In order to solve the problem of storing enterprise and policy data in relational databases, the data storage method provided in the embodiment of the present application can be used to describe and store enterprise and policy data by using the knowledge graph nodes to represent entities and edges to represent relationships between entities, thereby using the knowledge graph's flexible model definition capabilities to achieve high scalability of enterprise and policy data storage, and by establishing relationships at the node level, support for high-performance and fast associated queries. In other words, the embodiment of the present application can provide a lighter and more flexible storage solution for enterprise and policy data storage.

[0087] Based on the data storage solution provided in the embodiments of the present application, in the data storage scenario of enterprises and policies, for the target side, the target described above includes policies, the target text includes policy text, the sub-target includes policy points under the policy, the target label includes policy labels under policy points, the policy labels are associated with conditions, and the conditions are associated with conditional indicators; for the object side, the objects include enterprises, and the object indicators include enterprise indicators.

[0088] As an optional implementation, Figure 5 Another optional flow chart of the data storage method provided by the embodiment of the present application is shown as an example. The method flow can be implemented by the server, such as Figure 5 As shown, the method flow may include the following steps.

[0089] In step S510, the policy text of the policy is obtained.

[0090] Policies are the rules of action and requirements formulated by policy agencies to achieve certain goals. Policies can be issued by government agencies. The full content of the policy can be described in the policy text. In some embodiments, the embodiments of the present application can obtain the policy text of the policy from data sources such as websites and databases of government agencies.

[0091] In step S511, the policy text is analyzed to determine policy highlights from the policy text.

[0092] To achieve the policy goal of a policy, it may be necessary to achieve multiple sub-goals. A sub-goal can be regarded as a policy point, and a policy can contain multiple policy points. For example, for a policy that aims to accelerate the training of corporate talents, it may have sub-goals such as "accelerating the promotion of intelligent manufacturing", "stabilizing the corporate talent team", and "giving recognition rewards". These sub-goals can be regarded as policy points.

[0093] In some embodiments, the embodiments of the present application may encode the policy text to obtain the text features of the policy text, and then extract the text features corresponding to the policy points to obtain the policy points. As an optional implementation, the embodiments of the present application may perform intelligent recognition of the policy text of the policy through an intelligent recognition model, such as an intelligent recognition model based on BERT-LSTM-CRF. As an optional implementation, the embodiments of the present application may use a large-scale text corpus (such as a large-scale policy text) to pre-train a BERT model, and then use the BERT model to encode the policy text to obtain the text features of the policy text; then use a recurrent neural network in the form of LSTM to extract text features, thereby extracting text features corresponding to the policy points to obtain the policy points.

[0094] In step S512, a policy label is determined according to the policy key points.

[0095] A policy tag can be associated with a policy point, and can be tag information after the content of the policy point is refined; for example, a policy tag is tag information of a content refinement point of a policy point. In an embodiment of the present application, multiple conditions can abstract a policy tag, and a policy point can be associated with multiple different policy tags. Through the policy tags of policies, the embodiment of the present application can query similar policies.

[0096] In some embodiments, after extracting the text features corresponding to the policy points, the embodiments of the present application can decode the text features corresponding to the policy points (for example, decoding the text features based on CRF), thereby identifying at least one text phrase of the policy label category from the decoding results, and then the embodiments of the present application can standardize the identified text phrases and refine them into policy labels.

[0097] In step S513, the conditions corresponding to the policy tags and the condition indicators of the conditions are determined.

[0098] After determining the policy tag, the embodiment of the present application can determine the conditions associated with the policy tag. In some embodiments, the conditions can be the conditional content that needs to be met in order to meet the requirements of the policy tag. For example, a policy tag has specific requirements for annual sales and annual tax amounts for whether an enterprise meets the requirements, and these conditional contents can be used as conditions describing the policy tag, that is, the conditions associated with the policy tag.

[0099] After determining the conditions, the embodiments of the present application can determine the content of the rules describing the conditions, thereby obtaining the conditional indicators associated with the conditions. For example, if a condition requires that the annual sales of an enterprise reach a certain specific value, then this condition is associated with the conditional indicator corresponding to the annual sales, and the conditional indicator defines the rules that the annual sales must meet (for example, the regular expression of the specific value that the annual sales must reach). The embodiments of the present application can make a specific description of the conditions and rules that the enterprise needs to meet under the policy label through the conditions of the policy label and the conditional indicators of the conditions. For example, the policy text of a policy describes that the annual sales of an excellent enterprise should reach 500 million yuan, then the policy label of the excellent enterprise has the condition of annual sales reaching 500 million yuan, and the condition is associated with the conditional indicator corresponding to the annual sales, and the conditional indicator defines the regular expression of the annual sales.

[0100] In some embodiments, after decoding the text features corresponding to the policy points, the embodiments of the present application can identify data phrases from the decoding results, thereby extracting the conditions under the policy label and the conditional indicators under the conditions from the data phrases.

[0101] In step S514, policies, policy highlights, policy labels, conditions, and condition indicators are respectively used as policy-side entities in the knowledge graph, and nodes are used to represent policy-side entities in the knowledge graph, and edges are used to represent the relationships between policy-side entities.

[0102] A knowledge graph is a graph-based data structure consisting of nodes and edges. Nodes represent entities, and edges represent the relationships between entities. Entities can refer to things in the real world, such as people, places, companies, animals, etc., and relationships are used to represent the connections between different entities. Therefore, a knowledge graph is essentially a semantic network. When using a knowledge graph to store policy-side information, the embodiment of the present application needs to determine the entities corresponding to the policy-side information in the knowledge graph (referred to as policy-side entities), and to represent the relationships between policy-side entities through edges.

[0103] In some embodiments, the policy-side entities constructed in the knowledge graph by the embodiments of the present application may include: policy entities, policy key points entities, policy label entities, condition entities, and condition indicator entities. Thus, after obtaining information such as policies, policy key points, policy labels, conditions, and condition indicators from the policy text, the embodiments of the present application may construct policy entities, policy key points entities, policy label entities, condition entities, and condition indicator entities as nodes in the knowledge graph, and use edges to represent the association relationship between the above-mentioned policy-side entities.

[0104] In one example, Fig. 6A Another example diagram of the knowledge graph provided by the embodiment of the present application is shown as follows: Fig. 6AAs shown, in the knowledge graph, the embodiment of the present application can construct policy entities, policy key points entities, policy label entities, condition entities, and condition indicator entities with nodes in the knowledge graph. It should be noted that the policy entity, policy key points entity, policy label entity, and condition entity can be common type entities in the knowledge graph, and the condition indicator entity can be a CVT (Compound Value Type, composite object type) entity. The CVT entity allows multiple entities to be bound together. For example, a row in the "cast list" includes information such as actors and roles played, which can be understood as a CVT.

[0105] At the same time, the policy-side entities that have an association relationship with each other represent the association relationship between the entities with edges (directed edges). In some embodiments, the attributes of a policy-side entity can define the policy-side entities with an association relationship, so that the embodiments of the present application can represent the association relationship between the policy-side entities in the form of edges based on the attributes of the policy-side entities.

[0106] In some embodiments, a policy may include one or more policy points, and the attributes of a policy entity may include at least one of the following: policy name, promulgation time, promulgation department, effective time, expiration time, key node content, etc. In some embodiments, the attributes of a policy point entity may include at least one of the following: associated policy label, declaration start time, declaration end time, whether it can be redeemed, whether it can be declared, district, county, city, province, affiliated policy, full text of policy points, policy point number, etc. In some embodiments, the attributes of a policy label entity may include at least one of the following: label name, label id, associated policy points, judgment conditions, etc. In some embodiments, the attributes of a condition entity may include at least one of the following: associated condition indicator, affiliated policy label, condition code, etc. In some embodiments, the attributes of a condition indicator entity may include at least one of the following: associated enterprise indicator, associated condition, confidence, regular expression, etc.; wherein the regular expression of the condition indicator defines the requirements that the enterprise indicator should meet when the condition is met.

[0107] Combination Fig. 6AAs shown, the policy points included in the policy can be defined in the attributes of the policy entity, and the policy points entity can define the policy to which the policy points belong. In this embodiment of the present application, in the knowledge graph, the policy includes the policy points, and the policy points belong to the policy through edges. The policy points entity's attributes can define the associated policy tags of the policy points, and the policy tags entity's attributes can define the associated policy points of the policy tags. In this embodiment of the present application, in the knowledge graph, the policy points are associated with the policy tags (multi-valued), and the policy tags are associated with the policy points (multi-valued). The multi-valued and single-valued can represent the corresponding relationship between multiple and single. The policy tag entity's attributes can define the conditions for the policy tag determination, and the condition entity's attributes can define the policy tags to which the conditions belong. In this embodiment of the present application, in the knowledge graph, the policy tags are associated with the policy tags (multi-valued), and the policy tags are associated with the policy points (multi-valued). The associated condition indicators of the conditions can be defined in the attributes of the condition entity, and the attributes of the condition indicator entity can define the associated conditions of the condition indicators. In this case, the embodiment of the present application can represent the associated condition indicators (multi-valued) of the conditions and the associated relationship between the associated conditions (single-valued) of the condition indicators through edges in the knowledge graph.

[0108] Furthermore, the embodiments of the present application can also extract industry, region and other labels contained in policy labels from existing enterprise and policy data through NLP (Natural Language Processing) and manual review.

[0109] In step S515, in the knowledge graph, an association relationship is established from the condition indicator entity to the entity on the enterprise side.

[0110] After constructing the policy-side entity in the knowledge graph, in order to associate the policy-side entity with the enterprise-side entity, the embodiment of the present application may at least use the conditional indicator entity as a connecting entity connecting the enterprise and the policy in the knowledge graph. For example, based on the associated enterprise indicator defined by the attribute of the conditional indicator entity, the embodiment of the present application may use the conditional indicator entity as a bridge connecting enterprise information and policy information, and realize the connection between the policy-side and enterprise-side entities through the conditional indicator entity. In the case of using the conditional indicator entity as a connecting entity, the embodiment of the present application may construct an edge of the conditional indicator entity pointing to the enterprise-side entity based on the attribute of the conditional indicator entity (for example, the edge of the conditional indicator entity pointing to the enterprise indicator entity corresponding to the associated enterprise indicator), thereby establishing an association between the policy-side entity and the enterprise-side entity.

[0111] In one example, Figure 6B Another example diagram of the knowledge graph provided by the embodiment of the present application is shown in FIG. Fig. 6A and Figure 6B As shown, the embodiment of the present application can construct an edge from the conditional indicator entity to the enterprise indicator entity in the knowledge graph through the associated enterprise indicator (single value) indicated in the attribute of the conditional indicator entity.

[0112] Figure 7 Another optional flow chart of the data storage method provided by the embodiment of the present application is shown as an example. The method flow can be implemented by the server, such as Figure 7 As shown, the method flow may include the following steps.

[0113] In step S710, enterprise indicators are obtained from enterprise data of the enterprise.

[0114] The enterprise data of an enterprise may be provided by a government agency or an enterprise. The enterprise data of an enterprise carries the enterprise indicators of the enterprise, and the embodiment of the present application may obtain the enterprise indicators from the enterprise data.

[0115] Enterprise indicators such as business scope, address, tax amount, sales, etc. are indicator data used to characterize the enterprise situation. As an optional implementation, in the knowledge graph, enterprise indicators can be a special attribute, and enterprise indicators can inherit the attribute types (type, property, etc.) built into the knowledge graph. In a possible implementation, enterprise indicators can describe the attribute content of the enterprise itself, but there are also differences between enterprise indicators and enterprise attributes: enterprise indicators can be used as elements by conditional indicators. For example, the annual sales of an enterprise is an enterprise indicator. This enterprise indicator can correspond to a conditional indicator that requires the enterprise to reach a certain sales value in the condition.

[0116] In step S711, the enterprise and enterprise indicators are respectively used as entities on the enterprise side in the knowledge graph, and the entities on the enterprise side are represented by nodes in the knowledge graph.

[0117] In step S712, the attributes of the enterprise entity and the enterprise indicator entity are saved as node attributes in the knowledge graph, and the association relationship between the enterprise entity and the enterprise indicator entity and the node attributes is constructed.

[0118] In some embodiments, the enterprise-side entities constructed in the knowledge graph by the embodiments of the present application may include enterprise entities and enterprise indicator entities. Thus, the embodiments of the present application may respectively use enterprises and enterprise indicators as the enterprise-side entities in the knowledge graph, and use nodes in the knowledge graph to represent the enterprise-side entities, and use edges to represent the associations between the entity-side entities.

[0119] In one example, Fig. 8A Another example diagram of the knowledge graph provided by the embodiment of the present application is shown in FIG. Figure 6B and Fig. 8AAs shown, in the knowledge graph, the embodiment of the present application regards enterprises and enterprise indicators as entities on the enterprise side and represents them with nodes respectively, so as to construct enterprise entities and enterprise indicator entities in the knowledge graph. In some further embodiments, the embodiment of the present application can save the information of the entity through node attributes, for example, save the attributes of the entity through node attributes, and the attributes of the entity include two types, one is relational and the other is basic type; the attributes of the basic type can be represented by the basic type (basic types include strings, Booleans, numbers, etc.); relational attributes refer to the attribute pointing to another entity. Thus, the embodiment of the present application can further construct node attributes in the knowledge graph to save the attributes of enterprise entities and enterprise indicator entities, and construct the association relationship between enterprise entities and node attributes (enterprise entities have node attributes), and the association relationship between enterprise indicator entities and node attributes (enterprise indicator entities contain node attributes, and enterprise indicator entities inherit the attribute types built into the knowledge graph).

[0120] In some embodiments, the attributes of an enterprise entity may include at least one of the following: enterprise unique credit code, industry, enterprise qualification, associated policy label, business scope, detailed address, regional information, net assets, output value, net profit, revenue, total tax, maximum investment amount for a single project, equipment investment, R&D expenses, regional information, whether there is foreign trade, etc. The attributes of an enterprise indicator entity may include at least one of the following: enterprise indicator identification, associated condition indicator, candidate value, etc.

[0121] In step S713, in the knowledge graph, an association relationship from the enterprise indicator entity to the condition indicator entity is established in the form of an edge.

[0122] When the embodiment of the present application establishes an association relationship between the conditional indicator entity and the entity on the enterprise side in step S515, an association relationship between the conditional indicator entity and the enterprise indicator entity can be established; to complete the association relationship between the enterprise indicator entity and the conditional indicator entity, in step S713, the embodiment of the present application can establish an association relationship between the enterprise indicator entity and the associated conditional indicator in the form of an edge in the knowledge graph based on the associated conditional indicator indicated in the attributes of the enterprise indicator entity.

[0123] In one example, after extracting data phrases from a policy text, the embodiment of the present application can extract data from the data phrases through feature templates, so that the subject in the data phrase is regarded as an enterprise indicator that needs to be associated, and the numerical condition in the data phrase is used as the condition indicator of the condition. Thus, based on the data phrases in the policy text, the associated condition indicators of the enterprise indicator (which can be defined in the attributes of the enterprise indicator) and the associated enterprise indicators of the condition indicator (which can be defined in the attributes of the condition indicator) are defined, so as to achieve the association of the enterprise indicator to the specific condition through the specific condition indicator. For example, if the data phrase in the policy text is "annual R&D expenses reach 5 million", the association relationship between the enterprise indicator of "annual R&D expenses" and the condition indicator of "greater than or equal to 5 million" can be extracted.

[0124] In step S714, in the knowledge graph, an association relationship from the enterprise entity to the policy label entity is established in the form of an edge.

[0125] Combination Fig. 8A As shown, the embodiment of the present application can further realize the association between the entity on the enterprise side and the entity on the policy side through the association between the enterprise entity and the policy tag entity. In some embodiments, the embodiment of the present application can establish an association relationship from the enterprise entity to the policy tag entity in the form of an edge in the knowledge graph based on the associated policy tag (multi-value) indicated in the attribute of the enterprise entity.

[0126] In step S715 , the missing paths of the entities on the enterprise side and the entities on the policy side and the confidence of the missing paths are determined.

[0127] After the embodiment of the present application constructs the association relationship between the condition indicator entity and the enterprise indicator entity, and the association relationship between the enterprise entity and the policy label entity in the form of an edge, the association relationship between the constructed entities is realized based on existing data (such as existing enterprise indicators), and the existing data may not be able to fully and completely represent the association relationship between the entity on the enterprise side and the entity on the policy side, which may cause the association relationship between the entity on the enterprise side and the entity on the policy side to be missing, so the embodiment of the present application needs to supplement the missing relationship between the entity on the enterprise side and the entity on the policy side. In an embodiment of the present application, the embodiment of the present application can achieve the missing relationship between the entity on the enterprise side and the entity on the policy side by constructing the missing path between the entity on the enterprise side and the entity on the policy side. In some embodiments, the missing path can be in the form of an entity in the knowledge graph.

[0128] As an optional implementation, the missing paths between the entities on the enterprise side and the entities on the policy side may include: missing paths between the entities on the enterprise side and the conditional entities, and / or policy label entities, and / or policy entities and other policy-side entities. In other possible implementations, the missing paths between the entities on the enterprise side and the entities on the policy side may include: missing paths between missing enterprise indicator entities and conditional indicator entities; for example, there are missing enterprise indicators determined based on existing data, and the user is required to supplement the missing enterprise indicators, so that after the enterprise indicator entities corresponding to the missing enterprise indicators are constructed in the knowledge graph, it is necessary to supplement the missing paths with the conditional indicator entities based on the missing enterprise indicator entities. In one example, combined with Fig. 8A As shown, the embodiment of the present application can supplement the missing path from the enterprise entity to the conditional entity in the knowledge graph, and the missing path can be an entity in the form of CVT. Further, the embodiment of the present application can determine the association relationship between the missing path and the enterprise entity (for example Fig. 8A The multi-valued missing path CVT between the enterprise entity and the missing path shown, the association relationship between the missing path and the conditional entity (e.g. Fig. 8A The missing path shown in the figure is a single-valued association condition between the conditional entity and the conditional entity) to achieve the connection between the enterprise entity and the conditional entity.

[0129] In some embodiments, the condition indicator entity and the missing path are used as the connection path connecting the entity on the enterprise side and the entity on the policy side. The embodiment of the present application can determine the confidence level for the connection path (for example, determining the confidence level of the condition indicator entity and determining the confidence level of the missing path). The confidence level can indicate the degree of credibility of the relationship between the entity on the enterprise side and the entity on the policy side connected by the connection path. Fig. 8A As shown, the embodiment of the present application can determine the confidence for the missing path and the condition indicator entity respectively, and record it in the attributes of the missing path and the condition indicator entity.

[0130] It should be noted that determining the missing path in step S715 is only an optional means to improve the comprehensiveness of the data. The embodiment of the present application can also be set not to execute step S715.

[0131] In some further embodiments, in addition to common attributes, the enterprise entity may also have multiple composite attributes, which may be represented by multiple CVT-formatted entities. Fig. 8A As shown, in addition to the common attributes P1 and P2, the enterprise entity can also represent the information of the enterprise in multiple years through the entity of the composite attribute F1 and the entity of the composite attribute F2 in the form of CVT.

[0132] As an example, taking a specific enterprise scenario as an example, Figure 8B Another example diagram of the knowledge graph provided by the embodiment of the present application is shown as an example, which can be used for reference. Figure 8BThe construction of the knowledge graph shown can be similarly referred to the description of the corresponding part of the previous text, and will not be repeated here.

[0133] The embodiments of the present application will split the concepts related to enterprises and policies into entities represented by nodes, such as enterprise entities, indicator entities, policy entities, policy key points entities, conditional entities, etc., save the entity information with node attributes, and represent various relationships between entities (such as related enterprise indicators, related conditional indicators, etc.) with edges. The type of relationship is described by the attributes of the edges, so that the data on enterprises and policies can be stored and described in the knowledge graph in a network structure instead of a table row and column structure, which improves the scalability of enterprise and policy data storage and provides support for high-performance and fast association queries between enterprises and policies.

[0134] The present application also provides a data query method, which can be implemented based on the data storage method provided in the present application. For example, after storing the data of the target and the object in the knowledge graph based on the data storage method provided in the present application, the present application can query the target associated with the object, the object associated with the object, the target associated with the target, etc. in the knowledge graph. In some embodiments, the server can obtain the data query request sent by the terminal device; retrieve the knowledge graph (the construction process of the knowledge graph can be implemented based on the data storage method provided in the present application, and the specific content can refer to the description of the corresponding part above); based on the knowledge graph, determine the data query result.

[0135] As an optional implementation, take the target associated with the query object as an example, Fig. 9 The optional flow chart of the data query method provided by the embodiment of the present application is shown as an example. The method flow can be implemented by the server, refer to Fig. 9 The method flow may include the following steps.

[0136] In step S910, a data query request is obtained, where the data query request is used to query a target associated with a first object.

[0137] The data query request can be sent by the terminal device to the server. For example, when the user of the terminal device needs to query the target that the first object complies with (such as querying the policies that the enterprise complies with), the data query request can be sent to the server through the terminal device to request to query the target associated with the first object. The first object can be an object specified by the user, such as the enterprise where the user is located or a specified enterprise, etc.

[0138] In step S911, the knowledge graph is retrieved.

[0139] The server can store the data of the target and the object through the knowledge graph. The method of storing data through the knowledge graph can refer to the description of the corresponding part above, which will not be repeated here. When the server obtains the data query request sent by the terminal device, it can retrieve the knowledge graph storing the data of the target and the object.

[0140] In step S912, conditional entities and conditional indicator entities associated with the first object are determined in the knowledge graph.

[0141] In the case where the knowledge graph stores data on targets and objects and records the association between targets and objects, based on the first object to be queried, the embodiment of the present application can determine the conditional entity and conditional indicator entity associated with the first object from the knowledge graph. In some further embodiments, the knowledge graph has a target label entity, an object entity, and an association between a target label entity and an object entity. The embodiment of the present application can determine the associated target label entity in the knowledge graph through the object entity of the first object, and then determine the conditional entity associated with the target label entity, and the conditional indicator entity associated with the conditional entity, thereby obtaining the conditional entity and conditional indicator entity associated with the first object.

[0142] In an implementation example, an embodiment of the present application can query the policies that an enterprise complies with. At this time, the first object may be the first enterprise that needs to be queried. After the server obtains the data query request, it can determine the associated policy label entity in the knowledge graph from the enterprise entity corresponding to the first enterprise, and then determine the condition entity associated with the policy label entity and the condition indicator entity associated with the condition entity, thereby determining the condition entity and condition indicator entity associated with the first enterprise that needs to be queried in the knowledge graph.

[0143] In step S913, the matching degree between the first object and the associated condition indicator entity is determined according to the object indicator of the first object and the rule defined by the condition indicator entity associated with the first object.

[0144] After determining the condition entity and condition indicator entity associated with the first object, the embodiment of the present application can determine the matching degree between the first object and the associated condition indicator entity based on the object indicator of the first object and the rule defined by the associated condition indicator entity. In some embodiments, the condition indicator entity associated with the first object can define a regular expression, and the embodiment of the present application can bring the object indicator of the first object into the regular expression to determine the matching degree between the object indicator of the first object and the regular expression, thereby obtaining the matching degree between the first object and the associated condition indicator entity.

[0145] As an optional implementation, the rule expression defined by the conditional indicator entity may include operators of multiple operators, which take operands as parameters and perform different operator calculations through codes; operators can be understood as functions, such as "contains / greater than / less than / equal to / aggregate / average", etc. Based on this, the embodiment of the present application brings the object indicator of the first object into the rule expression, and the object indicator can be calculated as a parameter to obtain the matching degree between the object indicator and the rule expression.

[0146] In step S914, a target matched by the first object is determined from multiple targets according to the matching degree, and the first object has at least one condition indicator entity associated with any target.

[0147] In step S915, a data query result is obtained according to the determined target.

[0148] After obtaining the matching degree of the first object and the associated condition indicator entity, the data of multiple targets can be stored based on the knowledge graph, and the first object can have at least one associated condition indicator entity under any target; thus, for any target, the embodiment of the present application can determine the matching degree of the condition indicator entity associated with the first object in the target, thereby combining the matching degree of the condition indicator entity associated with the first object in the target to obtain the matching degree of the first object in the target; and then the embodiment of the present application can obtain the matching degree of the first object in multiple targets respectively, according to the matching degree of the first object in multiple targets, the embodiment of the present application can determine the target matched by the first object from multiple targets, and obtain the data query result based on the determined target and feedback to the terminal device. In some embodiments, the embodiment of the present application can sort the multiple targets according to the matching degree of the first object in multiple targets, so as to determine the set number of targets with the highest ranking as the target matched by the first object (for example, the target with the highest matching degree in the ranking is determined as the target matched by the first object), so as to carry the determined target as the target that the first object meets and is associated with, and is carried in the data query result and fed back to the terminal device.

[0149] In one example, an embodiment of the present application can query the policies that the first enterprise complies with. After determining the conditional entities and conditional indicator entities associated with the first enterprise to be queried from the knowledge graph, the embodiment of the present application can bring the enterprise indicators of the first enterprise into the rule expression of the associated conditional indicator entity, thereby determining the matching degree between the first enterprise and the associated conditional indicator entity; based on the fact that the first enterprise has at least one associated conditional indicator entity under multiple policies, the embodiment of the present application can determine the matching degree of the first enterprise under each policy based on the matching degree of the conditional indicator entities associated with the first enterprise under each policy, thereby taking one or more policies with the highest matching degree as the policies that the first enterprise complies with and is associated with, and carrying the policies that the first enterprise complies with and is associated with in the data query results and feeding back to the terminal device.

[0150] In some further embodiments, the embodiments of the present application can also realize the association query between objects (such as the association query between enterprises) and the association query between goals (such as the association query between policies) through the knowledge graph. The specific implementation principle is similar to the above description and can be exhibited. As an optional implementation, the embodiments of the present application can start from the enterprise entity of the enterprise, determine the policy-side entity corresponding to the specific type of industry label, regional label, and policy label in the knowledge graph, and then reversely find the enterprise entity of other related enterprises from the policy-side entity, so as to realize the search for related enterprises through specific industries, specific regions, and specific policy labels.

[0151] The embodiment of the present application forms a network association between elements such as objects (such as enterprises), goals (such as policies), conditions, and target labels (such as policy labels) by using the nodes of the knowledge graph to represent entities and the edges to represent the relationships between entities, replacing the relational database's storage method of rows and columns of tables, thereby reducing the query cost of high-frequency association queries in intelligent scenarios from multiple cross-products of the entire table to queries of specific edges and nodes of the knowledge graph, thereby improving the performance and speed of association queries. At the same time, when data storage is expanded, it can be achieved by expanding the nodes, edges, and attributes in the knowledge graph, rather than by adding rows and columns of new tables, which can simply and efficiently achieve data storage expansion and improve the scalability of data storage.

[0152] In some further embodiments, the embodiments of the present application may support the provision of object indicators by objects, and after the object provides the object indicators, the embodiments of the present application may automatically query the object for associated targets or associated objects based on the above-mentioned knowledge graph scheme, and feed back the query results to the object, so that the object can obtain recommended associated targets or associated objects. For example, after an enterprise provides enterprise indicators, the embodiments of the present application may automatically query the enterprise for associated policies based on the association relationship between the enterprise and the policy stored in the knowledge graph, and feed back the query results to the enterprise.

[0153] As an optional implementation, Fig. 10A Another optional flow chart of the data query method provided by the embodiment of the present application is exemplarily shown. The method flow can be executed by a terminal device, for example, when a first object queries an associated target, the method flow can be executed by the terminal device of the first object. Fig. 10A As shown, the method flow may include the following steps.

[0154] In step S101, an object indicator submission page is displayed, where a plurality of object indicator items are displayed.

[0155] When the first object queries the associated target or associated object, its terminal device can display the object indicator submission page so that the first object provides the object indicator of the first object. In some embodiments, the object indicator submission page can display multiple object indicator items, and one object indicator item can correspond to one object indicator. As an optional implementation, the object indicators corresponding to some object indicator items in the object indicator submission page can be automatically determined based on the server. For example, the server can automatically determine the object indicators corresponding to some object indicator items based on the object data related to the first object in the database; there may be some object indicator items in the multiple object indicator items that cannot automatically determine the object indicators, and the first object needs to manually input them. In other possible implementations, the object indicators corresponding to the multiple object indicator items on the object indicator submission page can all be input by the first object (for example, the object indicators of multiple object indicator items cannot be automatically determined by the server). In another possible implementation, the multiple object indicator items on the object indicator submission page can also be the corresponding object indicators that have been automatically filled in by the server, but the first object needs to confirm whether the object indicators filled in for each object indicator item are correct.

[0156] In one example, Fig. 10B An optional example diagram of an object indicator submission page is shown as an example. Fig. 10B As shown, the object indicator submission page can be an enterprise indicator submission page, which displays multiple enterprise indicator items that need to be filled in (enterprise indicator items can be regarded as optional examples of object indicator items), wherein the enterprise indicators corresponding to some enterprise indicator items have been automatically determined by the server based on the enterprise data entered into the enterprise database and government database, while the enterprise indicators corresponding to other enterprise indicator items have no data basis and need to be filled in and input by the enterprises themselves.

[0157] In step S102, the object index corresponding to the first object in the plurality of object index items is determined, and a data query request is sent based on the object index corresponding to the plurality of object index items.

[0158] In some embodiments, if there are some or all of the multiple object indicator items whose object indicators cannot be automatically determined, the embodiment of the present application can manually fill in the object indicators of the part or all of the object indicator items by the first object, and after the first object confirms the submission of the object indicators of each object indicator item, the object indicators corresponding to the first object in the multiple object indicator items can be determined. In other embodiments, if the object indicators of multiple object indicator items have been automatically determined by the server, the embodiment of the present application can determine the object indicators corresponding to the first object in the multiple object indicator items after the first object confirms the submission of the object indicators of each object indicator item.

[0159] As an optional implementation, based on the object indicators corresponding to the multiple object indicator items confirmed and submitted by the first object, the terminal device may send a data query request to the server to request the server to query and recommend targets or associated objects associated with the first object.

[0160] In one example, combining Fig. 10B As shown, after an enterprise fills in enterprise indicators for enterprise indicator items that have not been filled in, if the terminal device detects that the confirmation button has been clicked, the terminal device can determine that the enterprise confirms submission of the enterprise indicators corresponding to each enterprise indicator item, so that the terminal device can send a query request to the server to query the enterprise's associated policies or associated enterprises.

[0161] In step S103, the data query results determined based on the knowledge graph are obtained, and the data query results are displayed on the data result page.

[0162] After obtaining the data query request sent by the terminal device of the first object, the server can construct a knowledge graph based on the data storage method described above, and use the constructed knowledge graph to store the associated targets or associated objects of the first object, so that the server can determine the associated targets or associated objects of the first object based on the constructed knowledge graph, and then obtain the data query results.

[0163] The server may feed back the data query result to the terminal device, so that the terminal device obtains the data query result determined based on the knowledge graph. The terminal device may display a data result page, and display the data query result on the data result page so that the first object can understand the associated target or associated object.

[0164] In one example, Fig. 10C The following is an example of a data result page, which can be used for reference. Fig. 10B The enterprise indicator submission page shown in the figure shows that after submitting the enterprise indicators corresponding to multiple enterprise indicator items, the server can query the policies associated with the enterprise based on the knowledge graph and feed back the policies associated with the enterprise to the enterprise's terminal device, so that the enterprise's terminal device can display Fig. 10C The data result page shown displays the policies associated with the enterprise, so as to recommend the associated policies to the enterprise.

[0165] Furthermore, based on different query requirements of the first object, such as querying associated targets or associated objects, the terminal device of the first object can send a data query request that adapts to the query requirements to the server when the first object submits object indicators of multiple object indicator items, so that the server can feedback the corresponding data query results and display them on the data result page to meet the different requirements of the first object to query associated targets or associated objects.

[0166] When storing the data of targets and objects, the embodiments of the present application fully utilize the flexibility and scalability of knowledge graph modeling, which is conducive to the unification of multi-dimensional heterogeneous data and the reduction of software development costs. In addition, the complex associations between elements such as enterprises, policies, conditions, and policy labels are described through a mesh structure, which solves the problem of the amount of joint table operations in relational databases and is conducive to providing low-latency association query services. Furthermore, the mesh structure of the knowledge graph can also make full use of the theoretical knowledge of graph algorithms (such as the shortest path algorithm) to more effectively extract the value of data. For example, by precipitating enterprise and policy data into the knowledge graph, the built-in multi-dimensional knowledge ecology of the knowledge graph can be effectively utilized to work together with other data to generate greater value.

[0167] The data storage device provided in the embodiment of the present application is introduced below. The data storage device described below can be considered as a functional module required to be set up by the server to implement the data storage method provided in the embodiment of the present application. The content of the device described below can be referenced in correspondence with the content described above.

[0168] As an optional implementation, Fig.11 The block diagram of the data storage device provided by the embodiment of the present application is exemplarily shown, and the device can be applied to a server, such as Fig.11 As shown, the device may include:

[0169] A data acquisition module 011 is used to acquire a target text defining a target and object data of an object;

[0170] The condition and condition indicator acquisition module 012 is used to identify at least from the target text the condition corresponding to the target and the condition indicator associated with the condition; the condition indicator defines an associated object indicator that associates the object with the condition and a rule that the associated object indicator needs to satisfy;

[0171] An object index obtaining module 013 is used to obtain the object index of the object from the object data;

[0172] An entity construction module 014, used to construct at least in the knowledge graph a condition entity corresponding to the condition, a condition indicator entity corresponding to the condition indicator, and an object indicator entity corresponding to the object indicator;

[0173] The relationship construction module 015 is used to construct an association relationship between a condition entity and a condition indicator entity in the knowledge graph, and to construct an association relationship between the condition indicator entity and an object indicator entity based on the associated object indicator defined by the condition indicator.

[0174] In some embodiments, the goal includes at least one sub-goal; one sub-goal is associated with at least one goal tag, and one goal tag is tag information of a content extraction point of the sub-goal; one goal tag is associated with at least one condition, and one condition is associated with at least one condition indicator.

[0175] In some embodiments, the condition and condition indicator acquisition module 012 is used to identify the condition corresponding to the target from at least the target text, and the condition indicator associated with the condition includes:

[0176] Encoding the target text to obtain text features of the target text;

[0177] Extracting text features corresponding to the sub-goal from the text features;

[0178] Decoding the text features corresponding to the sub-goals, identifying text phrases of the label category from the decoding results, and standardizing and refining the text phrases into target labels;

[0179] And, a data phrase is identified from the decoding result, and a condition under the target tag and a condition indicator under the condition are extracted from the data phrase.

[0180] In some embodiments, the condition and condition indicator acquisition module 012 is used to identify a data phrase from the decoding result, and extract the condition under the target tag and the condition indicator under the condition from the data phrase, including:

[0181] A numerical phrase is identified from the decoding result, and features are extracted from the numerical phrase using a feature template to obtain conditions and condition indicators of the numerical phrase.

[0182] In some further embodiments, the device may also be used to: construct a target entity corresponding to the target, a sub-target entity corresponding to the sub-target, and a target tag entity corresponding to the target tag in the knowledge graph, and construct an association relationship between the target entity and the sub-target entity, an association relationship between the sub-target entity and the target tag entity, and an association relationship between the target tag entity and a condition entity;

[0183] And, constructing an object entity corresponding to the object in the knowledge graph, saving the attributes of the object entity and the object indicator entity through node attributes, and constructing an association relationship between the object entity, the object indicator entity and the node attributes;

[0184] And, constructing an association relationship between the object entity and the target label entity in the knowledge graph.

[0185] In some further embodiments, the device may also be used to: determine missing paths between entities in the knowledge graph, wherein the missing paths include missing paths of object entities and conditional entities, and / or target label entities, and / or target entities, or missing paths of missing object indicator entities and conditional indicator entities;

[0186] A confidence level of a connection path is determined, the connection path including the missing path and a condition indicator entity.

[0187] In some embodiments, the relationship building module 015 is used to build the association relationship between the condition indicator entity and the object indicator entity based on the associated object indicator defined by the condition indicator, including:

[0188] According to the associated object indicator indicated in the attribute of the condition indicator, constructing an association relationship from the condition indicator entity to the object indicator entity in the knowledge graph;

[0189] And, according to the associated condition indicator indicated in the attribute of the object indicator entity, an associated relationship from the object indicator entity to the condition indicator entity is constructed in the knowledge graph.

[0190] In some embodiments, in a government scenario, the goal includes a policy, the goal text includes a policy text, the sub-goals include policy points under the policy, the goal tags include policy tags under the policy points, the objects include enterprises, and the object indicators include enterprise indicators.

[0191] An embodiment of the present application also provides a data query device, which is used to: obtain a data query request; retrieve a knowledge graph, wherein the knowledge graph stores data based on the data storage method provided in an embodiment of the present application; and determine a data query result based on the knowledge graph.

[0192] In some embodiments, the data query request is used to query a target associated with the first object; the data query device is used to determine the data query result based on the knowledge graph, including:

[0193] Determining a condition entity and a condition indicator entity associated with the first object in the knowledge graph;

[0194] Determine the matching degree between the first object and the associated condition indicator entity according to the object indicator of the first object and the rule that the associated object indicator defined by the condition indicator entity associated with the first object needs to satisfy;

[0195] According to the matching degree, determining a target matched by the first object from a plurality of targets, the first object having at least one condition indicator entity associated with any target;

[0196] The data query results are obtained according to the determined goals.

[0197] The embodiment of the present application also provides a server, which can implement the data storage method provided by the embodiment of the present application through the data storage device described above, and / or implement the data query method provided by the embodiment of the present application through the data query device described above.

[0198] As an optional implementation, Fig.12 An optional block diagram of a server is shown. Fig.12 As shown, the server may include: at least one processor 1 , at least one communication interface 2 , at least one memory 3 and at least one communication bus 4 .

[0199] In the embodiment of the present application, the number of the processor 1 , the communication interface 2 , the memory 3 , and the communication bus 4 is at least one, and the processor 1 , the communication interface 2 , and the memory 3 communicate with each other through the communication bus 4 .

[0200] Optionally, the communication interface 2 may be an interface of a communication module for performing network communication.

[0201] Optionally, processor 1 may be a CPU, GPU (Graphics Processing Unit), NPU (embedded neural network processor), FPGA (Field Programmable Gate Array), TPU (Tensor Processing Unit), AI chip, ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application. Memory 3 may include high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory. Among them, memory 3 stores one or more computer executable instructions, and processor 1 calls the one or more computer executable instructions to execute the data storage method provided in the embodiments of the present application, and / or the data query method provided in the embodiments of the present application.

[0202] An embodiment of the present application also provides a storage medium, which stores one or more computer-executable instructions. When the one or more computer-executable instructions are executed, the data storage method provided by the embodiment of the present application and / or the data query method provided by the embodiment of the present application are implemented.

[0203] The embodiments of the present application also provide a computer program, which, when executed, implements the data storage method provided by the embodiments of the present application and / or the data query method provided by the embodiments of the present application.

[0204] The above describes the multi-value implementation scheme provided by the embodiments of the present application. The various optional methods introduced in each implementation scheme can be combined and cross-referenced with each other without conflict, thereby extending a variety of possible implementation schemes, which can all be considered as implementation schemes disclosed and open in the embodiments of the present application.

[0205] Although the embodiments of the present application are disclosed above, the present application is not limited thereto. Any person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be subject to the scope defined by the claims.

Claims

1. A data storage method, wherein: include: Get the target text defining the target and the object data of the object; At least identifying the condition corresponding to the target and the condition index associated with the condition from the target text; the condition index defines an associated object index associated with the object and the condition, and a rule that the associated object index needs to satisfy; Acquire an object index of the object from the object data; At least constructing a condition entity corresponding to the condition, a condition indicator entity corresponding to the condition indicator, and an object indicator entity corresponding to the object indicator in the knowledge graph; An association relationship between a conditional entity and a conditional indicator entity is constructed in a knowledge graph, and an association relationship between the conditional indicator entity and an object indicator entity is constructed based on an associated object indicator defined by the conditional indicator.

2. The method according to claim 1, wherein: The target includes at least one sub-target; one sub-target is associated with at least one target tag, and one target tag is tag information of a content extraction point of the sub-target; A target tag is associated with at least one condition, and a condition is associated with at least one condition indicator.

3. The method according to claim 2, wherein: The step of identifying at least from the target text the condition corresponding to the target, and the condition indicator associated with the condition includes: Encoding the target text to obtain text features of the target text; Extracting text features corresponding to the sub-goal from the text features; Decoding the text features corresponding to the sub-goals, identifying text phrases of the label category from the decoding results, and standardizing and refining the text phrases into target labels; And, a data phrase is identified from the decoding result, and a condition under the target tag and a condition indicator under the condition are extracted from the data phrase.

4. The method according to claim 3, wherein: The step of identifying a data phrase from the decoding result and extracting a condition under a target tag and a condition indicator under the condition from the data phrase includes: A numerical phrase is identified from the decoding result, and features are extracted from the numerical phrase using a feature template to obtain conditions and condition indicators of the numerical phrase.

5. The method according to claim 2, wherein: Also includes: Constructing a target entity corresponding to the target, a subtarget entity corresponding to the subtarget, and a target tag entity corresponding to the target tag in the knowledge graph, and constructing an association relationship between the target entity and the subtarget entity, an association relationship between the subtarget entity and the target tag entity, and an association relationship between the target tag entity and the condition entity; And, constructing an object entity corresponding to the object in the knowledge graph, saving the attributes of the object entity and the object indicator entity through node attributes, and constructing an association relationship between the object entity, the object indicator entity and the node attributes; And, constructing an association relationship between the object entity and the target label entity in the knowledge graph.

6. The method according to claim 5, wherein: It also includes: determining missing paths between entities in the knowledge graph, wherein the missing paths include missing paths of object entities and conditional entities, and / or target label entities, and / or target entities, or missing paths of missing object indicator entities and conditional indicator entities; A confidence level of a connection path is determined, the connection path including the missing path and a condition indicator entity.

7. The method according to claim 1, wherein: The establishing of the association relationship between the condition indicator entity and the object indicator entity based on the associated object indicator defined by the condition indicator includes: According to the associated object indicator indicated in the attribute of the condition indicator, constructing an association relationship from the condition indicator entity to the object indicator entity in the knowledge graph; And, according to the associated condition indicator indicated in the attribute of the object indicator entity, an associated relationship from the object indicator entity to the condition indicator entity is constructed in the knowledge graph.

8. The method according to any one of claims 2 to 6, wherein: The goal includes policy, the goal text includes policy text, the sub-goal includes policy points under the policy, the goal label includes policy labels under the policy points, the object includes an enterprise, and the object indicator includes an enterprise indicator.

9. A data query method, wherein: include: Get data query request; Retrieving a knowledge graph, wherein the knowledge graph stores data based on the data storage method according to any one of claims 1 to 8; Based on the knowledge graph, a data query result is determined.

10. The method according to claim 9, wherein: The data query request is used to query a target associated with the first object; and determining the data query result based on the knowledge graph includes: Determining a condition entity and a condition indicator entity associated with the first object in the knowledge graph; Determine the matching degree between the first object and the associated condition indicator entity according to the object indicator of the first object and the rule that the associated object indicator defined by the condition indicator entity associated with the first object needs to satisfy; According to the matching degree, determining a target matched by the first object from a plurality of targets, the first object having at least one condition indicator entity associated with any target; The data query results are obtained according to the determined goals.

11. A data query method, wherein: include: Displaying an object indicator submission page, wherein the object indicator submission page displays multiple object indicator items; Determine an object index corresponding to the first object in the plurality of object index items, and send a data query request based on the object index corresponding to the plurality of object index items; Obtain data query results determined based on a knowledge graph, and display the data query results on a data result page; wherein the knowledge graph stores data based on the data storage method described in any one of claims 1-8.

12. A server, wherein: It includes at least one memory and at least one processor, the memory stores one or more computer-executable instructions, and the processor calls the one or more computer-executable instructions to execute the data storage method as described in any one of claims 1-8, and / or the data query method as described in any one of claims 9-10.

13. A storage medium, wherein: The storage medium stores one or more computer executable instructions, and when the one or more computer executable instructions are executed, they implement the data storage method according to any one of claims 1 to 8, or the data query method according to any one of claims 9 to 10, or the data query method according to claim 11.