Intelligent knowledge graph automatic updating and optimizing method and system

Abstract

The invention discloses an intelligent knowledge graph automatic updating and optimizing method and system, and the method comprises the steps: carrying out the topology of a current knowledge graph, generating a structure of the current knowledge graph, extracting an edge set, a node combination set and a relation type set of the structure of the current knowledge graph, obtaining an embedding matrix of the to-be-updated data and an embedding matrix of the current knowledge graph; setting a similarity model of the to-be-updated data and the knowledge graph data, calculating the similarity between the embedding matrix of the to-be-updated data and the embedding matrix of the current knowledge graph, and if the similarity is lower than a preset similarity threshold, updating the corresponding node of the current knowledge graph into the to-be-updated data; and setting a knowledge graph optimization model, optimizing the updated current knowledge graph, fitting the knowledge graph optimization model through a least square method, and minimizing the knowledge graph optimization model.

Description

Technical Field

[0001] The present invention belongs to the technical field of knowledge graph optimization of satellite intelligence data, and more specifically, relates to an intelligent knowledge graph automatic updating and optimization method and system. Background Art

[0002] Knowledge Graph is a graphical structure used to represent and organize knowledge. It displays knowledge entities and their relationships in the form of nodes and edges, similar to a network graph. Nodes usually represent specific entities (such as people, places, events, etc.), while edges represent the relationships between these entities (such as "is the father of...", "is located in", etc.).

[0003] The key features of knowledge graphs are as follows: Structured data: Knowledge graphs organize information into structured data so that complex information can be obtained through simple queries.

[0004] Semantic relationship: The knowledge graph contains not only the data itself, but also the semantic relationship between the data, which helps to understand the actual meaning of the data.

[0005] Scalability: The knowledge graph can be continuously expanded and dynamically updated as new knowledge is discovered and added.

[0006] Cross-domain integration: Data from different fields can be integrated to form a comprehensive knowledge network.

[0007] Currently, knowledge graph technology has been used in the field of satellite intelligence data display technology, but there is currently no technical solution that can automatically and intelligently update the knowledge graph. Summary of the invention

[0008] In order to solve the above technical problems, the present invention proposes an intelligent knowledge graph automatic updating and optimization method, wherein the knowledge graph is a knowledge graph of satellite intelligence data, including: Perform topology analysis on the current knowledge graph, generate the structure of the current knowledge graph, extract the edge set, node combination and relationship type set of the structure of the current knowledge graph, and obtain the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph; Set a similarity model between the data to be updated and the knowledge graph data, calculate the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than the preset similarity threshold, update the corresponding node of the current knowledge graph to the data to be updated; A knowledge graph optimization model is set to optimize the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method to minimize the knowledge graph optimization model.

[0009] Furthermore, the similarity model between the data to be updated and the knowledge graph data includes: , in, For the current knowledge graph at time The embedding matrix The data to be updated at time The embedding matrix The similarity of is the edge set, For relationship type The first weight of For time The weight of For Node The first importance weight of For relationship type The second weight, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. is a node set, For Node The normalized weight of For Node The normalized weight of For Node The first importance weight of It is the ReLU function or the Sigmoid function.

[0010] Furthermore, the knowledge graph optimization model include: , in, For in time The edge set when For relationship type The third weight, For Node The second importance weight of For Node The second importance weight of For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, is the time consistency function, is the regularization function, For Node To Node distance.

[0011] Furthermore, the temporal consistency function include: , in, is the adjustment factor of the temporal consistency function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, For in time The node collection at that time.

[0012] Furthermore, the regularization function include: , in, is the adjustment factor of the regularization term, For in time is the regularization term of the structure of the current knowledge graph, For in time The structure of the current knowledge graph, For Node The weight of the local subgraph embedding vector of For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For Node The local subgraph embedding vector in the data to be updated, used to represent the node The context information in the data to be updated, For relationship type The weight of the local subgraph embedding vector of For in time The set of relationship types when For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. For relationship type The local subgraph embedding vector in the data to be updated is used to represent the context information of the relationship in the data to be updated. is the weight of sparsity regularization, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time The node collection at that time.

[0013] The present invention also proposes an intelligent knowledge graph automatic updating and optimization system, wherein the knowledge graph is a knowledge graph of satellite intelligence data, including: The topology module is used to perform topology analysis on the current knowledge graph, generate the structure of the current knowledge graph, extract the edge set, node combination and relationship type set of the structure of the current knowledge graph, and obtain the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph; A similarity calculation module is used to set a similarity model between the data to be updated and the knowledge graph data, calculate the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than a preset similarity threshold, update the corresponding node of the current knowledge graph to the data to be updated; The optimization module is used to set the knowledge graph optimization model and optimize the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method and the knowledge graph optimization model is minimized.

[0014] Furthermore, the similarity model between the data to be updated and the knowledge graph data includes: , in, For the current knowledge graph at time The embedding matrix The data to be updated at time The embedding matrix The similarity of is the edge set, For relationship type The first weight of For time The weight of For Node The first importance weight of For relationship type The second weight, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. is a node set, For Node The normalized weight of For Node The normalized weight of For Node The first importance weight of It is the ReLU function or the Sigmoid function.

[0015] Furthermore, the knowledge graph optimization model include: , in, For in time The edge set when For relationship type The third weight, For Node The second importance weight of For Node The second importance weight of For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, is the time consistency function, is the regularization function, For Node To Node distance.

[0016] Furthermore, the temporal consistency function include: , in, is the adjustment factor of the temporal consistency function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, For in time The node collection at that time.

[0017] Furthermore, the regularization function include: , in, is the adjustment factor of the regularization term, For in time is the regularization term of the structure of the current knowledge graph, For in time The structure of the current knowledge graph, For Node The weight of the local subgraph embedding vector of For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For Node The local subgraph embedding vector in the data to be updated, used to represent the node The context information in the data to be updated, For relationship type The weight of the local subgraph embedding vector of For in time The set of relationship types when For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. For relationship type The local subgraph embedding vector in the data to be updated is used to represent the context information of the relationship in the data to be updated. is the weight of sparsity regularization, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time The node collection at that time.

[0018] Compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects: The present invention sets a similarity model between the data to be updated and the knowledge graph data, calculates the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than a preset similarity threshold, updates the corresponding node of the current knowledge graph to the data to be updated; sets a knowledge graph optimization model, and optimizes the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method and minimized. Through the above technical scheme, the present invention can first calculate the similarity, determine whether it needs to be updated, and then optimize the knowledge graph. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of the method of embodiment 1 of the present invention; Figure 2 It is a system structure diagram of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0020] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0021] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.

[0022] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts in the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0023] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets or instructions.

[0024] The display is used to show the interactive cross-section of each application.

[0025] All subscripts in the formulas of the present invention are only used to distinguish parameters and have no actual meaning.

[0026] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.

[0027] Example 1 like Figure 1 As shown, the present invention proposes an intelligent knowledge graph automatic updating and optimization method, wherein the knowledge graph is a knowledge graph of satellite intelligence data, and the satellite intelligence data includes map information, meteorological information, building information, etc., including: Step 101, topologically analyze the current knowledge graph, generate the structure of the current knowledge graph, extract the edge set, node combination and relationship type set of the structure of the current knowledge graph, and obtain the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph; Step 102, setting a similarity model between the data to be updated and the knowledge graph data, calculating the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than a preset similarity threshold, updating the corresponding node of the current knowledge graph to the data to be updated; Specifically, the similarity model between the data to be updated and the knowledge graph data includes: , in, For the current knowledge graph at time The embedding matrix The data to be updated at time The embedding matrix The similarity of is the edge set, For relationship type The first weight of For time The weight of For Node The first importance weight of For relationship type The second weight, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. is a node set, For Node The normalized weight of For Node The normalized weight of For Node The first importance weight of is the ReLU function or the Sigmoid function; Specifically, the sum of all weights in the similarity model between the data to be updated and the knowledge graph data is equal to 1, and then each parameter is randomly initialized, and the initialization learning rate is set. The first weight of the relationship type, the weight of time, etc. in the similarity model between the data to be updated and the knowledge graph data are fit by the gradient descent method until convergence.

[0028] Step 103, setting a knowledge graph optimization model, and optimizing the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method and the knowledge graph optimization model is minimized.

[0029] Specifically, the knowledge graph optimization model include: , in, For in time The edge set when For relationship type The third weight, For Node The second importance weight of For Node The second importance weight of For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, is the time consistency function, is the regularization function, For Node To Node distance.

[0030] Specifically, the time consistency function include: , in, is the adjustment factor of the temporal consistency function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, For in time The node collection at that time.

[0031] Specifically, the regularization function include: , in, is the adjustment factor of the regularization term, For in time is the regularization term of the structure of the current knowledge graph, For in time The structure of the current knowledge graph, For Node The weight of the local subgraph embedding vector of For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For Node The local subgraph embedding vector in the data to be updated, used to represent the node The context information in the data to be updated, For relationship type The weight of the local subgraph embedding vector of For in time The set of relationship types when For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. For relationship type The local subgraph embedding vector in the data to be updated is used to represent the context information of the relationship in the data to be updated. is the weight of sparsity regularization, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time The node collection at that time.

[0032] Example 2 like Figure 2 As shown, the embodiment of the present invention further proposes an intelligent knowledge graph automatic updating and optimization system, wherein the knowledge graph is a knowledge graph of satellite intelligence data, including: The topology module is used to perform topology analysis on the current knowledge graph, generate the structure of the current knowledge graph, extract the edge set, node combination and relationship type set of the structure of the current knowledge graph, and obtain the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph; A similarity calculation module is used to set a similarity model between the data to be updated and the knowledge graph data, calculate the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than a preset similarity threshold, update the corresponding node of the current knowledge graph to the data to be updated; Specifically, the similarity model between the data to be updated and the knowledge graph data includes: , in, For the current knowledge graph at time The embedding matrix The data to be updated at time The embedding matrix The similarity of is the edge set, For relationship type The first weight of For time The weight of For Node The first importance weight of For relationship type The second weight, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. is a node set, For Node The normalized weight of For Node The normalized weight of For Node The first importance weight of It is the ReLU function or the Sigmoid function.

[0033] The optimization module is used to set the knowledge graph optimization model and optimize the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method and the knowledge graph optimization model is minimized.

[0034] Specifically, the knowledge graph optimization model include: , in, For in time The edge set when For relationship type The third weight, For Node The second importance weight of For Node The second importance weight of For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, is the time consistency function, is the regularization function, For Node To Node distance.

[0035] Specifically, the time consistency function include: , in, is the adjustment factor of the temporal consistency function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, For in time The node collection at that time.

[0036] Specifically, the regularization function include: , in, is the adjustment factor of the regularization term, For in time is the regularization term of the structure of the current knowledge graph, For in time The structure of the current knowledge graph, For Node The weight of the local subgraph embedding vector of For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For Node The local subgraph embedding vector in the data to be updated, used to represent the node The context information in the data to be updated, For relationship type The weight of the local subgraph embedding vector of For in time The set of relationship types when For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. For relationship type The local subgraph embedding vector in the data to be updated is used to represent the context information of the relationship in the data to be updated. is the weight of sparsity regularization, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time The node collection at that time.

[0037] Example 3 An embodiment of the present invention also proposes a storage medium storing multiple instructions, wherein the instructions are used to implement the intelligent knowledge graph automatic updating and optimization method.

[0038] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0039] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the method steps: Step 101, topologically analyze the current knowledge graph, generate the structure of the current knowledge graph, extract the edge set, node combination and relationship type set of the structure of the current knowledge graph, and obtain the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph; Step 102, setting a similarity model between the data to be updated and the knowledge graph data, calculating the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than a preset similarity threshold, updating the corresponding node of the current knowledge graph to the data to be updated; Specifically, the similarity model between the data to be updated and the knowledge graph data includes: , in, For the current knowledge graph at time The embedding matrix The data to be updated at time The embedding matrix The similarity of is the edge set, For relationship type The first weight of For time The weight of For Node The first importance weight of For relationship type The second weight, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. is a node set, For Node The normalized weight of For Node The normalized weight of For Node The first importance weight of It is the ReLU function or the Sigmoid function.

[0040] Step 103, setting a knowledge graph optimization model, and optimizing the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method and the knowledge graph optimization model is minimized.

[0041] Specifically, the knowledge graph optimization model include: , in, For in time The edge set when For relationship type The third weight, For Node The second importance weight of For Node The second importance weight of For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, is the time consistency function, is the regularization function, For Node To Node distance.

[0042] Specifically, the time consistency function include: , in, is the adjustment factor of the time consistency function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, For in time The node collection at that time.

[0043] Specifically, the regularization function include: , in, is the adjustment factor of the regularization term, For in time is the regularization term of the structure of the current knowledge graph, For in time The structure of the current knowledge graph, For Node The weight of the local subgraph embedding vector of For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For Node The local subgraph embedding vector in the data to be updated, used to represent the node The context information in the data to be updated, For relationship type The weight of the local subgraph embedding vector of For in time The set of relationship types when For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. For relationship type The local subgraph embedding vector in the data to be updated is used to represent the context information of the relationship in the data to be updated. is the weight of sparsity regularization, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time The node collection at that time.

[0044] Example 4 An embodiment of the present invention also proposes an electronic device, including a processor and a storage medium connected to the processor, wherein the storage medium stores multiple instructions, and the instructions can be loaded and executed by the processor so that the processor can execute an intelligent knowledge graph automatic update and optimization method.

[0045] Specifically, the electronic device of this embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.

[0046] Among them, the storage medium can be used to store software programs and modules, such as an intelligent knowledge graph automatic update and optimization method in an embodiment of the present invention, corresponding program instructions / modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, realizing the above-mentioned intelligent knowledge graph automatic update and optimization method. The storage medium may include a high-speed random storage medium, and may also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include a storage medium remotely arranged relative to the processor, and these remote storage media can be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0047] The processor can call the information and application programs stored in the storage medium through the transmission system to execute the method steps of Example 1.

[0048] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0049] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0050] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0051] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0052] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0053] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0054] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. An intelligent knowledge graph automatic updating and optimization method, wherein: The knowledge graph is a knowledge graph of satellite intelligence data, and is characterized by including: Perform topology analysis on the current knowledge graph, generate the structure of the current knowledge graph, extract the edge set, node combination and relationship type set of the structure of the current knowledge graph, and obtain the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph; Set a similarity model between the data to be updated and the knowledge graph data, calculate the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than the preset similarity threshold, update the corresponding node of the current knowledge graph to the data to be updated; A knowledge graph optimization model is set to optimize the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method to minimize the knowledge graph optimization model.

2. The intelligent knowledge graph automatic updating and optimization method according to claim 1, characterized in that: The similarity model between the data to be updated and the knowledge graph data includes: ， in, For the current knowledge graph at time The embedding matrix The data to be updated at time The embedding matrix The similarity of is the edge set, For relationship type The first weight of For time The weight of For Node The first importance weight of For relationship type The second weight, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. is a node set, For Node The normalized weight of For Node The normalized weight of For Node The first importance weight of It is the ReLU function or the Sigmoid function.

3. The intelligent knowledge graph automatic updating and optimization method according to claim 1, characterized in that: The knowledge graph optimization model include: ， in, For in time The edge set when For relationship type The third weight, For Node The second importance weight of For Node The second importance weight of For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, is the time consistency function, is the regularization function, For Node To Node distance.

4. The intelligent knowledge graph automatic updating and optimization method according to claim 3, characterized in that: Temporal consistency function include: ， in, is the adjustment factor of the time consistency function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, For in time The node collection at that time.

5. The intelligent knowledge graph automatic updating and optimization method according to claim 3, characterized in that: The regularization function include: ， in, is the adjustment factor of the regularization term, For in time is the regularization term of the structure of the current knowledge graph, For in time The structure of the current knowledge graph, For Node The weight of the local subgraph embedding vector of For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For Node The local subgraph embedding vector in the data to be updated, used to represent the node The context information in the data to be updated, For relationship type The weight of the local subgraph embedding vector of For in time The set of relationship types when For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. For relationship type The local subgraph embedding vector in the data to be updated is used to represent the context information of the relationship in the data to be updated. is the weight of sparsity regularization, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time The node collection at that time.

6. An intelligent knowledge graph automatic updating and optimization system, wherein: The knowledge graph is a knowledge graph of satellite intelligence data, and is characterized by including: The topology module is used to perform topology analysis on the current knowledge graph, generate the structure of the current knowledge graph, extract the edge set, node combination and relationship type set of the structure of the current knowledge graph, and obtain the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph; A similarity calculation module is used to set a similarity model between the data to be updated and the knowledge graph data, calculate the similarity between the embedding matrix of the data to be updated and the embedding matrix of the current knowledge graph, and if the similarity is lower than a preset similarity threshold, update the corresponding node of the current knowledge graph to the data to be updated; The optimization module is used to set the knowledge graph optimization model and optimize the updated current knowledge graph, wherein the knowledge graph optimization model is fitted by the least squares method and the knowledge graph optimization model is minimized.

7. The intelligent knowledge graph automatic updating and optimization system according to claim 6, characterized in that: The similarity model between the data to be updated and the knowledge graph data includes: ， in, For the current knowledge graph at time The embedding matrix The data to be updated at time The embedding matrix The similarity of is the edge set, For relationship type The first weight of For time The weight of For Node The first importance weight of For relationship type The second weight, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the data to be updated, For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. is a node set, For Node The normalized weight of For Node The normalized weight of For Node The first importance weight of It is ReLU function or Sigmoid function.

8. The intelligent knowledge graph automatic updating and optimization system according to claim 6, characterized in that: The knowledge graph optimization model include: ， in, For in time The edge set when For relationship type The third weight, For Node The second importance weight of For Node The second importance weight of For Node The embedding vector in the current knowledge graph, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, is the time consistency function, is the regularization function, For Node To Node distance.

9. The intelligent knowledge graph automatic updating and optimization system according to claim 8, characterized in that: Temporal consistency function include: ， in, is the adjustment factor of the temporal consistency function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time Time Node The embedding vector in the current knowledge graph, For in time Time Node The embedding vector in the current knowledge graph, For in time The node collection at that time.

10. The intelligent knowledge graph automatic updating and optimization system according to claim 8, characterized in that: The regularization function include: ， in, is the adjustment factor of the regularization term, For in time is the regularization term of the structure of the current knowledge graph, For in time The structure of the current knowledge graph, For Node The weight of the local subgraph embedding vector of For Node The local subgraph embedding vector in the current knowledge graph is used to represent the node Contextual information in the current knowledge graph, For Node The local subgraph embedding vector in the data to be updated, used to represent the node The context information in the data to be updated, For relationship type The weight of the local subgraph embedding vector of For in time The set of relationship types when For relationship type The local subgraph embedding vector in the current knowledge graph is used to represent the contextual information of the relationship in the current knowledge graph. For relationship type The local subgraph embedding vector in the data to be updated is used to represent the context information of the relationship in the data to be updated. is the weight of sparsity regularization, For Node The embedding vector in the current knowledge graph, is the ReLU function or the Sigmoid function, For in time The node collection at that time.

Images