Methods, devices, equipment, and storage media for updating enterprise knowledge graphs

Abstract

This application provides a method, apparatus, device, and storage medium for updating an enterprise knowledge graph. The enterprise knowledge graph includes at least one enterprise node and multiple attribute information nodes associated with the enterprise node. The method includes: determining the industry status of the enterprise node in the enterprise knowledge graph, where the industry status includes a predicted state and an unpredicted state; identifying a target enterprise node whose industry status is unpredicted; extracting a subgraph of the enterprise knowledge graph from the target enterprise node; obtaining a prediction result of the industry type of the target enterprise based on the attribute information in the subgraph; and updating the enterprise knowledge graph based on the prediction result, updating the industry status of the target enterprise node to a predicted state. This application, by setting an industry status for enterprise nodes, facilitates the identification of enterprise nodes with unpredicted industry types, avoids predicting all enterprise nodes in the graph, and improves the efficiency of updating the enterprise knowledge graph.

Description

Technical Field

[0001] This application relates to the field of knowledge graph technology, and in particular to a method, apparatus, device and storage medium for updating an enterprise knowledge graph. Background Technology

[0002] A knowledge graph is a knowledge base with a directed graph structure that is applied in various fields. For example, information from different companies can be organized into a corporate knowledge graph, making it easier to perform statistics and view data on multiple companies.

[0003] An enterprise knowledge graph can include multiple enterprise nodes representing enterprise entities and multiple industry nodes representing industry types. As a company's business develops, its industry type changes, necessitating timely updates to the connections between enterprise nodes and industry nodes in the enterprise knowledge graph. Since the enterprise knowledge graph contains enterprise nodes representing multiple companies, it becomes crucial to optimize the update efficiency by providing targeted updates for each company. Summary of the Invention

[0004] This application provides a method, apparatus, device, and storage medium for updating enterprise knowledge graphs, so as to improve the updating efficiency of enterprise knowledge graphs.

[0005] Firstly, this application provides a method for updating an enterprise knowledge graph, wherein the enterprise knowledge graph includes at least one enterprise node and at least one attribute information node associated with the enterprise node, the attribute information node being used to represent the enterprise's attribute information; the method includes:

[0006] The industry status of each enterprise node in the enterprise knowledge graph is determined. Each enterprise node has an identifier for determining the industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted.

[0007] Identify the target enterprise node whose industry status is unpredictable among the enterprise nodes, and extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph;

[0008] Based on the attribute information corresponding to the attribute information node of the target enterprise node in the enterprise knowledge graph subgraph, the industry type of the target enterprise is predicted, and the prediction result is obtained.

[0009] The enterprise knowledge graph is updated based on the prediction results, and the industry status of the target enterprise node is updated to the predicted status.

[0010] Secondly, this application provides an apparatus for updating an enterprise knowledge graph, the enterprise knowledge graph including at least one enterprise node and at least one attribute information node associated with the enterprise node, the attribute information node being used to represent the attribute information of the enterprise; the apparatus includes:

[0011] The industry status determination module is used to determine the industry status of each enterprise node in the enterprise knowledge graph. Each enterprise node has an identifier for determining the industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted.

[0012] The subgraph extraction module is used to identify the target enterprise node whose industry status is unpredictable among the enterprise nodes, and extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph.

[0013] The industry type prediction module is used to predict the industry type of the target enterprise based on the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph, and obtain the prediction result.

[0014] The graph update module is used to update the enterprise knowledge graph based on the prediction results and update the industry status of the target enterprise node to the predicted status.

[0015] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0016] The memory stores computer-executed instructions;

[0017] The processor executes computer execution instructions stored in the memory to implement the enterprise knowledge graph update method as described in the first aspect.

[0018] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the enterprise knowledge graph update method as described in the first aspect.

[0019] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the enterprise knowledge graph updating method as described in the first aspect.

[0020] This application provides a method, apparatus, device, and storage medium for updating an enterprise knowledge graph. By assigning an industry status identifier to each enterprise node, it determines whether the industry type of each enterprise node has been predicted. If the industry status of an enterprise node is marked as unpredicted, a subgraph of the enterprise knowledge graph for that node is retrieved from the enterprise knowledge graph. Based on the attribute information in the subgraph, the industry type of the enterprise node is predicted. The industry node corresponding to that industry type is then connected to the enterprise node, completing the update of the enterprise knowledge graph. This avoids the process of determining the industry type for all enterprise nodes in the enterprise knowledge graph, as is done in existing technologies. Only unpredicted enterprise nodes are identified, reducing computational load, saving update time, and improving the efficiency of enterprise knowledge graph updates. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0022] Figure 1 A flowchart illustrating a method for updating an enterprise knowledge graph, provided as an embodiment of this application;

[0023] Figure 2 A schematic diagram of an enterprise knowledge graph provided in an embodiment of this application;

[0024] Figure 3 A schematic diagram of an enterprise knowledge graph subgraph provided in an embodiment of this application;

[0025] Figure 4 A flowchart illustrating a method for updating an enterprise knowledge graph, provided as an embodiment of this application;

[0026] Figure 5 A flowchart illustrating a method for updating an enterprise knowledge graph, provided as an embodiment of this application;

[0027] Figure 6 A structural block diagram of an enterprise knowledge graph updating device provided in an embodiment of this application;

[0028] Figure 7 A structural block diagram of an enterprise knowledge graph updating device provided in an embodiment of this application;

[0029] Figure 8 A structural block diagram of an electronic device provided in an embodiment of this application;

[0030] Figure 9 This is a structural block diagram of an electronic device provided in an embodiment of this application.

[0031] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0033] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0034] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0035] In the description of this application, it should be understood that the terms "first," "second," "third," etc., are used only to distinguish similar objects and are not necessarily used to describe a specific order or sequence, nor should they be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0036] It should be noted that, due to space limitations, this application specification does not exhaustively list all possible implementation methods. Those skilled in the art, after reading this application specification, should be able to deduce that, as long as the technical features do not contradict each other, any combination of technical features can constitute an optional implementation method. The following provides a detailed description of each embodiment.

[0037] Enterprise knowledge graphs can represent information such as the industry type, address, establishment date, and products of a company. They allow for the rapid acquisition of relevant company information. Timely updates to the enterprise knowledge graph are crucial for accurately obtaining information about each company. Each change in a company's information may alter its industry type; therefore, it is necessary to predict a company's industry type based on its existing information and update the enterprise knowledge graph accordingly.

[0038] In an enterprise knowledge graph, each enterprise corresponds to an enterprise subgraph, which includes an enterprise node, all nodes directly connected to that enterprise node, and the connections between these nodes. Each time the enterprise knowledge graph is updated, the enterprise subgraphs of each enterprise are retrieved, the industry type of the enterprise is predicted, and the industry nodes corresponding to the predicted industry type are connected to the corresponding enterprise node, thus completing the knowledge graph update. This process requires traversing and querying all enterprise nodes in the knowledge graph to determine the industry type of each node. However, some enterprise nodes do not change their industry type, and traversing and calculating for each node wastes a significant amount of time, resulting in low update efficiency and failing to guarantee the real-time updates of the enterprise knowledge graph.

[0039] This application provides a method, apparatus, device, and storage medium for updating an enterprise knowledge graph, which aims to solve the above-mentioned technical problems in the prior art.

[0040] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0041] Figure 1 This is a flowchart illustrating a method for updating an enterprise knowledge graph according to an embodiment of this application. The enterprise knowledge graph includes at least one enterprise node and at least one attribute information node associated with the enterprise node. The attribute information node represents the enterprise's attribute information. This method is executed by an enterprise knowledge graph updating device. Figure 1 As shown, the method includes the following steps:

[0042] S101. Determine the industry status of each enterprise node in the enterprise knowledge graph; whereby each enterprise node has a corresponding identifier for determining the industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted.

[0043] For example, a pre-drawn enterprise knowledge graph can be included, which may contain multiple nodes connected by edges. Nodes in the enterprise knowledge graph may include enterprise nodes, attribute information nodes, and industry nodes. Enterprise nodes represent enterprise entities and can be represented by enterprise name or enterprise number; one enterprise node represents one enterprise. Attribute information nodes can be connected to enterprise nodes and are used to represent the attribute information of the enterprise node. One attribute information can correspond to one attribute information node. Attribute information refers to various information related to the enterprise, such as enterprise location, products produced by the enterprise, and projects undertaken by the enterprise. Industry nodes can also be connected to enterprise nodes and are used to represent the industry type to which the enterprise node belongs. Industry type refers to the type of industry involved in the enterprise's production; for example, industry type could be database security, cloud application security, natural language processing, and artificial intelligence.

[0044] In the process of drawing an enterprise knowledge graph, attribute information nodes corresponding to enterprise attribute information can be connected to enterprise nodes, and enterprise nodes can also be connected to industry nodes corresponding to the enterprise's industry type. In this embodiment, the industry type to which an enterprise node belongs can be automatically predicted based on the attribute information nodes connected to it, and the industry nodes corresponding to the predicted industry type can be connected to the enterprise node to update the enterprise knowledge graph. Figure 2 This is a schematic diagram of an enterprise knowledge graph. Figure 2 The graph includes two enterprise nodes, Enterprise Node 1 and Enterprise Node 2, representing two enterprises. Enterprise Node 1 and Enterprise Node 2 share the same industry type and are both connected to Enterprise Node 1. Enterprise Node 1 has four attribute information nodes: Attribute Information Node 1, Attribute Information Node 2, Attribute Information Node 3, and Attribute Information Node 4. Enterprise Node 2 has two attribute information nodes: Attribute Information Node 5 and Attribute Information Node 6. Each enterprise node corresponds to an industry status identifier. When an enterprise node is added to the enterprise knowledge graph, an industry status is assigned to each enterprise node. The industry status represents the predicted industry type of the enterprise node; that is, the industry status indicates whether the industry type of the enterprise node has been predicted. Predicting the industry type of an enterprise node means determining the industry type of the enterprise corresponding to the enterprise node based on the attribute information nodes connected to the enterprise node. Industry status includes predicted and unpredictable states. Industry status can change at any time. For example, when adding a new enterprise node to the enterprise knowledge graph, the industry status of the enterprise node is unpredictable; when connecting an enterprise node to an industry node, the industry status of the enterprise node can change to predicted.

[0045] When updating the enterprise knowledge graph, a preset enterprise knowledge graph is obtained, and the industry status of each enterprise node is determined from the enterprise knowledge graph. The enterprise nodes in the enterprise knowledge graph can be traversed to obtain the industry status of each enterprise node and determine whether the industry status of each enterprise node is a predicted state or an unpredictable state.

[0046] S102. Identify the target enterprise node in each enterprise node whose industry status is unpredictable, and extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph.

[0047] For example, if the industry status of an enterprise node is predicted, it is considered that the industry type of the enterprise node has been determined; if the industry status of an enterprise node is unpredictable, it is considered that the industry type of the enterprise node has not yet been determined, and it is necessary to predict the industry type of the enterprise node based on the attribute information of the enterprise node, and then connect the enterprise node with the industry node corresponding to the industry type.

[0048] After obtaining the industry status of each enterprise node in the enterprise knowledge graph, we identify enterprise nodes with an unpredictable industry status, resulting in multiple such nodes. These unpredictable enterprise nodes are then designated as the target enterprise nodes.

[0049] Enterprise nodes whose industry status is unpredictable are identified as target enterprise nodes. Enterprise knowledge graph subgraphs corresponding to the target enterprise nodes are extracted from the enterprise knowledge graph. An enterprise knowledge graph subgraph is a part of the overall enterprise knowledge graph; each enterprise node corresponds to one enterprise knowledge graph subgraph.

[0050] In an enterprise knowledge graph, there can be multiple enterprise nodes, each represented by an enterprise identifier. Each enterprise node corresponds to a subgraph of the enterprise knowledge graph, and the enterprise knowledge graph can be composed of these subgraphs. A subgraph of an enterprise node consists of that enterprise node and all nodes connected to it. Nodes directly connected to an enterprise node include at least one attribute information node. For example, multiple attribute information nodes and multiple industry nodes can be connected to a single enterprise node. The subgraph of the enterprise knowledge graph can represent the attribute information of the enterprise corresponding to the enterprise node. If an industry node exists in the subgraph, it can also indicate the industry type to which the enterprise node currently belongs. Figure 3 This is a schematic diagram of a subgraph of an enterprise knowledge graph. Figure 3In the diagram, Enterprise Node 1 represents a company, and Attribute Information Nodes 1 and 2 represent the company's location. Attribute Information Node 1 could be City A, and Attribute Information Node 2 could be District B of City A. Attribute Information Nodes 3 and 4 could be the names of the products the company produces, and Industry Node 1 could be the industry type to which the company's products belong. In other words, the company's location, products, and industry type can be determined through the enterprise knowledge graph subgraph.

[0051] Identify the target enterprise node whose industry state is unpredictable, and then determine its enterprise knowledge graph subgraph from the enterprise knowledge graph. For example, if there are two enterprise nodes with an unpredictable industry state, namely enterprise node one and enterprise node two, then find all nodes connected to these two enterprise nodes from the enterprise knowledge graph. Enterprise node one and the nodes connected to it can form one enterprise subgraph, and enterprise node two and the nodes connected to it can form another enterprise subgraph.

[0052] Each enterprise node can be pre-divided into its own enterprise knowledge graph subgraph in the enterprise knowledge graph. After the target enterprise node is determined, the corresponding enterprise knowledge graph subgraph can be directly obtained from the enterprise knowledge graph.

[0053] In this embodiment, extracting the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph includes: determining the target enterprise node as the first node, and determining the node connected to the first node as the second node according to the connection relationship between the nodes in the enterprise knowledge graph; wherein the second node includes at least one attribute information node; and determining the first node, the second node, and the connection relationship between the first node and the second node as the enterprise knowledge graph subgraph corresponding to the first node.

[0054] Specifically, target enterprise nodes with an unpredictable industry status are identified and designated as the first node. The connections between nodes are obtained from the enterprise knowledge graph. Based on these connections, nodes directly connected to the first node are identified. These connected nodes are designated as the second nodes of that first node. Each second node includes at least one attribute information node; that is, one first node can correspond to multiple second nodes, and each second node can include multiple attribute information nodes and one or more industry nodes. Different first nodes can correspond to different second nodes.

[0055] It's worth noting that the enterprise knowledge graph subgraph includes not only the first node and its second node, but also the connections between them. These connections can include terms like "production," "belong to," "location," "located in," "construction," and "declaration." For example, if the first node is a company and the second node is the company's product, the connection between them is "production," meaning the company represented by the first node produces the product represented by the second node. As another example, if the second node is an industry type, the connection between them could be "belong to."

[0056] The advantage of this setup is that by searching for the second node connected to the first node in the enterprise knowledge graph, the enterprise subgraphs of each enterprise node can be retrieved from the enterprise knowledge graph. This avoids errors in enterprise subgraph retrieval, facilitates the determination of industry type based on enterprise subgraphs, and improves the update accuracy of the enterprise knowledge graph.

[0057] S103. Based on the attribute information corresponding to the attribute information node of the target enterprise node in the enterprise knowledge graph subgraph, predict the industry type of the target enterprise and obtain the prediction result.

[0058] For example, after obtaining the enterprise knowledge graph subgraph corresponding to the target enterprise node, attribute information nodes connected to the target enterprise node are determined from the enterprise knowledge graph subgraph. Each attribute information node represents an attribute of the target enterprise node. Based on the attribute information nodes in the enterprise knowledge graph subgraph, the attribute information of the enterprise node is determined. Based on the attribute information of the target enterprise node, the industry type to which the target enterprise node belongs is predicted, resulting in a prediction result for the industry type to which the target enterprise node belongs. One target enterprise node represents one target enterprise, thus obtaining the prediction result for the industry type to which the target enterprise belongs.

[0059] Rules for determining industry types can be preset, and industry types can be predicted based on these preset rules. For example, the association between attribute information and industry types can be preset. After determining the attribute information, the industry type associated with the attribute information is determined based on the preset association, which serves as the predicted industry type for the enterprise node. Specifically, the association between the enterprise's products in the attribute information and the industry type can be preset. If the enterprise's product is a cloud platform, the associated industry type could be cloud application security; if the enterprise's product is a robot, the associated industry type could be artificial intelligence. The enterprise's products are obtained from the attribute information to predict the industry type of the enterprise node. In this embodiment, the rules for determining industry types are not specifically limited.

[0060] If there are multiple target enterprise nodes in an unpredictable industry state, the industry type can be predicted sequentially for each target enterprise node, or the industry type can be predicted simultaneously for all target enterprise nodes. The enterprise knowledge graph can be updated once the industry type of each target enterprise node is predicted.

[0061] S104. Update the enterprise knowledge graph based on the prediction results, and update the industry status of the target enterprise node to the predicted status.

[0062] For example, after obtaining the prediction result of the industry type to which an enterprise node belongs, the enterprise knowledge graph can be automatically updated. Alternatively, an update reminder can be sent to staff, which can include the confirmed prediction result of the industry type. Staff review the confirmed prediction result and determine whether to update the enterprise knowledge graph. If the staff determines that the prediction result of the industry type is correct, the enterprise knowledge graph is updated; if the staff determines that the prediction result of the industry type has a problem, the original enterprise knowledge graph is maintained. Updating the enterprise knowledge graph can involve changing the industry nodes connected to the enterprise nodes, or adding connections between enterprise nodes and industry nodes to the existing enterprise knowledge graph.

[0063] By connecting a target enterprise node in the enterprise knowledge graph with industry nodes representing its respective industry type, the industry status of that target enterprise node can be automatically updated from unpredicted to predicted. This avoids the need to determine the industry type of the target enterprise node again, reducing the computational load of updating the enterprise knowledge graph and improving update efficiency.

[0064] In this embodiment, the enterprise knowledge graph also includes at least one industry node; the industry node is connected to the enterprise node, and the industry node is used to represent the industry type to which the connected enterprise node belongs; updating the enterprise knowledge graph according to the prediction result includes: detecting whether there is an industry node in the enterprise knowledge graph that corresponds to the prediction result; if so, connecting the target enterprise node and the industry node corresponding to the prediction result with an edge to update the enterprise knowledge graph; if not, adding an industry node corresponding to the prediction result to the enterprise knowledge graph, and connecting the target enterprise node and the industry node corresponding to the prediction result with an edge to update the enterprise knowledge graph.

[0065] Specifically, after determining the predicted industry type of an enterprise node, the system searches the enterprise knowledge graph to see if a corresponding industry node exists. If it does, the enterprise node is connected to that industry node, updating the enterprise knowledge graph. If not, the system first adds the corresponding industry node to the enterprise knowledge graph, then connects the target enterprise node and the industry node corresponding to the predicted result with an edge, thus updating the enterprise knowledge graph. Alternatively, the user can be prompted to manually add non-existent industry nodes. Connections between nodes in the enterprise knowledge graph can be edge connections. If an enterprise node is already connected to an industry node, and a new industry type prediction is determined based on the enterprise node's attribute information, the connection between the enterprise node and the current industry node can be broken, and the enterprise node can be connected to the industry node corresponding to the newly determined prediction.

[0066] A single enterprise node can connect to multiple industry nodes. For example, if an enterprise node is connected to industry node 1, but its industry status is unpredictable, its industry type can be predicted based on its current attribute information. If the determined industry type differs from the industry type of the previously connected industry nodes, the enterprise node can be connected to the industry nodes of the determined industry type. The original industry nodes can still remain connected to the enterprise node, thus meeting the practical production needs of an enterprise with multiple industry types.

[0067] The advantage of this setup is that by connecting enterprise nodes with industry nodes of the same industry type, the enterprise knowledge graph can be updated in a timely manner. Users can then visually observe these changes when viewing the enterprise knowledge graph, thus improving viewing efficiency.

[0068] This application provides a method for updating an enterprise knowledge graph. By assigning an industry status identifier to each enterprise node, it determines whether the industry type of each enterprise node has been predicted. If the industry status of an enterprise node is marked as unpredicted, a subgraph of the enterprise knowledge graph for that node is obtained. Based on the attribute information in the subgraph, the industry type of the enterprise node is predicted. The industry node corresponding to that industry type is then connected to the enterprise node, completing the update of the enterprise knowledge graph. This avoids the process of determining the industry type for all enterprise nodes in the enterprise knowledge graph, as in existing technologies. Only unpredicted enterprise nodes are identified, reducing computational load, saving update time, and improving the efficiency of enterprise knowledge graph updates.

[0069] Figure 4This is a flowchart illustrating a method for updating an enterprise knowledge graph, which is an optional embodiment based on the above embodiments.

[0070] In this embodiment, the industry type of the target enterprise is predicted based on the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph of the target enterprise node, and the prediction result is obtained. This can be further refined as follows: determining the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph of the target enterprise node; converting the attribute information into triple data based on the preset connection relationship between the target enterprise node and the attribute information node in the enterprise knowledge graph subgraph of the target enterprise node, wherein the preset connection relationship is used to indicate the attribute relationship between the target enterprise corresponding to the target enterprise node and the attribute information corresponding to the attribute information node; and obtaining the prediction result of the industry type of the target enterprise based on the triple data and the preset enterprise industry prediction model. Figure 4 As shown, the enterprise knowledge graph updating method in this embodiment includes the following steps:

[0071] S401. Determine the industry status of each enterprise node in the enterprise knowledge graph; whereby each enterprise node has a corresponding identifier for determining the industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted.

[0072] For example, this step can refer to step S101 above, and will not be repeated here.

[0073] S402. Identify the target enterprise node in each enterprise node whose industry status is unpredictable, and extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph.

[0074] For example, this step can refer to step S102 above, and will not be repeated here.

[0075] S403. Determine the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph corresponding to the target enterprise node.

[0076] For example, each attribute information node can represent an attribute information, and the attribute information represented by each attribute information node is pre-stored. The attribute information nodes in the enterprise knowledge graph subgraph are determined, obtaining all attribute information of the enterprise nodes in that subgraph. Based on the attribute information of the enterprise nodes and a preset industry type prediction rule, the prediction result of the industry type to which the enterprise node belongs is determined. For example, the preset industry type prediction rule is that if the attribute information includes a preset content, then the predicted industry type to which the enterprise node belongs is considered to be the database security industry type.

[0077] S404. Based on the preset connection relationship between the target enterprise node and the attribute information node in the enterprise knowledge graph subgraph corresponding to the target enterprise node, the attribute information is converted into triple data. The preset connection relationship is used to indicate the attribute relationship between the target enterprise corresponding to the target enterprise node and the attribute information corresponding to the attribute information node.

[0078] For example, the attribute information represented by each attribute information node in the enterprise knowledge graph subgraph is determined, and each attribute information is converted into triple data. Pre-defined connections exist between the connected nodes in the knowledge graph. For instance, a pre-defined connection could be the attribute relationship between the target enterprise node and the attribute information node, where the attribute relationship could be "belongs to," "produces," or "originates." The connection relationship between the target enterprise node and the attribute information node in the enterprise knowledge graph subgraph is determined, and triple data is generated based on the target enterprise node, attribute information node, and their connection relationships. That is, the triple data can represent the relationship between the enterprise and attribute information, such as a originate relationship or a production relationship. The format of the triple data can be (target enterprise node, connection relationship, attribute information node). For example, based on the enterprise knowledge graph subgraph of Enterprise 1, the attribute information obtained is that Enterprise 1 produces firewall products; therefore, the triple data could be (Enterprise 1, production, firewall).

[0079] S405. Based on the triplet data and the preset enterprise industry prediction model, obtain the prediction result of the industry type of the target enterprise.

[0080] For example, a pre-set enterprise industry prediction model can be used. Inputting triplet data into this model will output a prediction of the industry type of the target enterprise node, i.e., the prediction of the target enterprise's industry type. In this embodiment, the enterprise industry prediction model can be an RGCN (Relational Graph Convolutional Network) model. The RGCN model is pre-trained; its input can be triplet data, and its output can be the prediction of the industry type. For example, if the input triplet data is (Enterprise 1, Production, Firewall), the RGCN model can output the industry type as "Database Security". In this embodiment, the model structure of the RGCN model is not specifically limited; it can consist of convolutional layers, pooling layers, and fully connected layers, etc.

[0081] Training samples can be collected in advance to train the RGCN model. The training samples can be attribute information and the corresponding actual industry types. The attribute information from the training samples is input into the RGCN model to be trained. If the predicted industry type output by the RGCN model is consistent with the actual industry type of the training samples, then the RGCN model training is considered complete.

[0082] By using the RGCN model to determine the industry type of enterprise nodes in the enterprise subgraph, the prediction results can be automatically determined, reducing the process of manual judgment by staff. The determination of industry type is more accurate and efficient, thereby improving the accuracy and efficiency of updating the enterprise knowledge graph.

[0083] In this embodiment, based on triplet data and a preset enterprise industry prediction model, the prediction result of the target enterprise's industry type is obtained, including: inputting triplet data into the preset enterprise industry prediction model; wherein, at least two industry types are preset in the enterprise industry prediction model; extracting features from the triplet data according to the preset convolutional layer in the enterprise industry prediction model to obtain the confidence level of the target enterprise belonging to each industry type; and determining the prediction result of the target enterprise's industry type based on the confidence level.

[0084] Specifically, the enterprise industry prediction model can be an RGCN model, which can specifically be a BERT-RGCN (Bidirectional Encoder Representation from Transformers-Relational Graph Convolutional Network) model. Multiple industry types can be pre-defined. Triple data is input into the enterprise industry prediction model, which calculates the confidence score of each triple data point belonging to each industry type. The confidence score represents the probability that the triple data point belongs to each industry type. For example, convolutional layers in the enterprise industry prediction model can be used to extract features from the triple data, and the confidence score can be calculated based on the extracted features. The calculated confidence scores are then sorted by size, and the highest confidence score is determined. The industry type corresponding to the highest confidence score is determined as the predicted industry type of the target enterprise in the enterprise knowledge graph subgraph. For example, the final output of the RGCN model could be:

[0085] [{"ent_name":"Company One","industry_pred":"Artificial Intelligence Industry","pred_prob":"0.91"}

[0086] Based on this output, it can be determined that the predicted industry type of Company 1 is artificial intelligence, with a confidence level of 0.91.

[0087] The enterprise industry forecasting model can also push the confidence level of each industry type to the front-end interface for staff to view, and staff can manually select the industry type to which the enterprise node belongs.

[0088] The advantages of this setup are that it calculates the confidence level of the target enterprise node belonging to each industry type based on a pre-trained enterprise industry prediction model, and determines the predicted industry type of the target enterprise based on the confidence level. This achieves automatic prediction of industry type, and the determination based on the confidence level effectively avoids errors in industry type identification. It also reduces the manual determination process by staff, saving manpower and time, and improving the efficiency and accuracy of industry type prediction.

[0089] S406. Update the enterprise knowledge graph based on the prediction results, and update the industry status of the target enterprise node to the predicted status.

[0090] For example, this step can refer to step S103 above, and will not be repeated here.

[0091] This application provides a method for updating an enterprise knowledge graph. By assigning an industry status identifier to each enterprise node, it determines whether the industry type of each enterprise node has been predicted. If the industry status of an enterprise node is marked as unpredicted, a subgraph of the enterprise knowledge graph for that node is obtained. Based on the attribute information in the subgraph, the industry type of the enterprise node is predicted. The industry node corresponding to that industry type is then connected to the enterprise node, completing the update of the enterprise knowledge graph. This avoids the process of determining the industry type for all enterprise nodes in the enterprise knowledge graph, as in existing technologies. Only unpredicted enterprise nodes are identified, reducing computational load, saving update time, and improving the efficiency of enterprise knowledge graph updates.

[0092] Figure 5 This is a flowchart illustrating a method for updating an enterprise knowledge graph, which is an optional embodiment based on the above embodiments.

[0093] In this embodiment, after identifying the target enterprise nodes whose industry status is "unpredictable" among the enterprise nodes, the following can be added: The target enterprise nodes with the "unpredictable" industry status can be written into a preset first database; wherein, the first database is used to record enterprise nodes whose industry type has not been predicted. After predicting the industry type of the target enterprise and obtaining the prediction result, the following can be added: The target enterprise node can be deleted from the first database, and the prediction result of the target enterprise node and its industry type can be associated and written into a preset second database; wherein, the second database is used to record enterprise nodes whose industry type has been predicted and their corresponding industry types. Figure 5 As shown, the method for updating the enterprise knowledge graph in this embodiment includes the following steps:

[0094] S501. Determine the industry status of each enterprise node in the enterprise knowledge graph; whereby each enterprise node has a corresponding identifier for determining the industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted.

[0095] For example, this step can refer to step S101 above, and will not be repeated here.

[0096] S502. Identify the target enterprise nodes in each enterprise node whose industry status is unpredictable, extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph, and write the target enterprise nodes whose industry status is unpredictable into the preset first database; wherein, the first database is used to record enterprise nodes that have not made predictions for their respective industry types.

[0097] For example, if the industry status of an enterprise node is determined to be predicted, then the industry type of that enterprise node is not predicted; if the industry status of an enterprise node is determined to be unpredictable, then the unpredictable enterprise node is identified as the target enterprise node. From the enterprise knowledge graph, the enterprise knowledge graph subgraph corresponding to the target enterprise node is obtained. Based on the attribute information represented by the attribute information nodes in the enterprise knowledge graph subgraph, the prediction result of the industry type of the target enterprise node is determined.

[0098] A primary database is pre-set. This database records enterprise nodes whose industry type has not been predicted; that is, it records target enterprise nodes whose industry status is "unpredictable." The primary database can be a MySQL database, a relational database. After identifying all target enterprise nodes in the "unpredictable" industry status, these nodes are written to the primary database. For example, they can be written using the enterprise name or identifier. The contents of the primary database can be displayed on the front-end interface, allowing staff to view the target enterprise nodes whose industry type needs to be predicted.

[0099] Based on the target enterprise node stored in the MySQL database, query the enterprise knowledge graph subgraph of that target enterprise node in the enterprise knowledge graph. The newly added data in the MySQL database, i.e., the target enterprise node, can be transformed through entity linking and entity fusion using the d2rq tool, and the transformed data is then updated in the pre-defined Neo4j database. Neo4j is a high-performance graph database. Neo4j queries can be used to query the enterprise knowledge graph subgraph from the enterprise knowledge graph. The advantage of this setup is that by writing the target enterprise node into the first database, the target enterprise node for which the enterprise knowledge graph subgraph needs to be determined can be clearly identified, avoiding the need to retrieve the enterprise knowledge graph subgraph from already predicted enterprise nodes, thus improving the efficiency of determining the enterprise knowledge graph subgraph.

[0100] S503. Based on the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph of the target enterprise node, predict the industry type of the target enterprise, obtain the prediction result, delete the target enterprise node from the first database, and associate the prediction result of the target enterprise node and the industry type of the target enterprise into the preset second database; wherein, the second database is used to record the enterprise node whose industry type has been predicted and the corresponding industry type.

[0101] For example, based on the attribute information nodes represented by the attribute information nodes in the enterprise knowledge graph subgraph, the predicted industry type of the target enterprise is determined. After predicting the industry type of the target enterprise, the target enterprise node is deleted from the first database. This not only saves space in the first database but also avoids repeated predictions of the industry type for the target enterprise node, improving the update efficiency of the enterprise knowledge graph.

[0102] After determining the predicted industry type of the target enterprise, the target enterprise node and its corresponding industry type can be associated and stored in a pre-defined second database. This second database can also be a MySQL database. It records enterprise nodes whose predicted industry type has been determined and the industry type corresponding to those nodes. In this embodiment, recording an enterprise node in the database means recording its enterprise identifier, such as its name. The second database can be pushed to the front-end interface for staff to view. After viewing the content of the second database, staff can manually determine whether to update the enterprise knowledge graph based on its content—that is, whether to connect the target enterprise node with the determined industry type node in the enterprise knowledge graph. This avoids errors in determining the industry type, which could lead to incorrect updates to the enterprise knowledge graph and effectively improves the accuracy of the knowledge graph updates.

[0103] S504. Update the enterprise knowledge graph based on the prediction results, and update the industry status of the target enterprise node to the predicted status.

[0104] For example, this step can refer to step S103 above, and will not be repeated here.

[0105] In this embodiment, the method further includes: in response to the new enterprise node instruction, adding a new enterprise node in the enterprise knowledge graph, and adding an industry status identifier for the unpredicted state to the new enterprise node.

[0106] Specifically, staff can issue new enterprise node creation commands on the front-end interface, adding new enterprise nodes to the enterprise knowledge graph at any time. Upon receiving a new enterprise node creation command, the system determines the enterprise name or identifier from the command and adds the node to the knowledge graph. The command can also include the enterprise's attribute information. Based on this, the system identifies the attribute information nodes to be added to the knowledge graph and adds them accordingly. Each time a new enterprise node is added, it is assigned an initial industry status identifier, which is always in an unpredicted state. That is, after adding a new enterprise node, its industry type must be determined before connecting it to the corresponding industry node.

[0107] The advantage of this setup is that after adding a new enterprise node, the enterprise knowledge graph can be updated by adding unpredictable industry states, avoiding the omission of industry type predictions for newly added enterprise nodes and the omission of edge connections between enterprise nodes and industry nodes in the enterprise knowledge graph.

[0108] In this embodiment, the method further includes: responding to the adjustment instruction of attribute information, determining the enterprise node corresponding to the attribute information to be adjusted, and determining the industry status of the enterprise node corresponding to the attribute information to be adjusted as an unpredicted state.

[0109] Specifically, staff can issue attribute information adjustment commands at any time through the front-end interface, thereby updating the attribute information of enterprise nodes. For example, if an enterprise increases the number of product types it produces, it can identify the attribute information nodes corresponding to the added product types and connect these nodes to the enterprise node. As the number of products an enterprise produces increases, the industry type to which the enterprise belongs may also increase. In other words, adjusting enterprise attribute information will affect the industry type to which the enterprise belongs.

[0110] Once a change is detected in the attribute information nodes connected to an enterprise node in the knowledge graph, the industry status of that enterprise node can be updated to an unpredictable state. A change in the attribute information nodes connected to an enterprise node refers to the addition, removal, or alteration of the attribute information nodes connected to that node. If the enterprise node's industry status was unpredictable before the attribute information adjustment, that unpredictable state is maintained; if the enterprise node's industry status was predicted before the attribute information adjustment, the industry status is changed from predicted to unpredictable.

[0111] The advantage of this setup is that when the attribute information of an enterprise node changes, the industry nodes connected to that enterprise node may also change. By changing the industry status of an enterprise node to an unpredictable state, it is easier to predict the industry type of enterprise nodes whose attribute information has changed, thereby improving the accuracy of enterprise knowledge graph updates and avoiding omissions in enterprise knowledge graph updates.

[0112] In this embodiment, the addition of enterprise nodes and the adjustment of attribute information can be performed before or after any step.

[0113] This application provides a method for updating an enterprise knowledge graph. By assigning an industry status identifier to each enterprise node, it determines whether the industry type of each enterprise node has been predicted. If the industry status of an enterprise node is marked as unpredicted, a subgraph of the enterprise knowledge graph for that node is obtained. Based on the attribute information in the subgraph, the industry type of the enterprise node is predicted. The industry node corresponding to that industry type is then connected to the enterprise node, completing the update of the enterprise knowledge graph. This avoids the process of determining the industry type for all enterprise nodes in the enterprise knowledge graph, as in existing technologies. Only unpredicted enterprise nodes are identified, reducing computational load, saving update time, and improving the efficiency of enterprise knowledge graph updates.

[0114] Figure 6 This is a structural block diagram of an enterprise knowledge graph updating device provided in an embodiment of this application. The enterprise knowledge graph includes at least one enterprise node and at least one attribute information node associated with the enterprise node. The attribute information node is used to represent the attribute information of the enterprise. For ease of explanation, only the parts relevant to the embodiments of this disclosure are shown. (Refer to...) Figure 6 The device includes: an industry status determination module 601, a sub-map extraction module 602, an industry type prediction module 603, and a map update module 604.

[0115] The industry status determination module 601 is used to determine the industry status of each enterprise node in the enterprise knowledge graph. Each enterprise node has an identifier for determining the industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted.

[0116] The subgraph extraction module 602 is used to determine the target enterprise node in the enterprise node whose industry status is unpredictable, and extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph.

[0117] The industry type prediction module 603 is used to predict the industry type of the target enterprise based on the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph, and obtain the prediction result.

[0118] The graph update module 604 is used to update the enterprise knowledge graph according to the prediction results and update the industry status of the target enterprise node to the predicted status.

[0119] Figure 7 This application provides a structural block diagram of an enterprise knowledge graph updating device. Figure 6 Based on the illustrated embodiments, as Figure 7 As shown, the industry type prediction module 603 includes an attribute information determination unit 6031, a triplet data determination unit 6032, and a prediction result acquisition unit 6033.

[0120] The attribute information determination unit 6031 is used to determine the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph corresponding to the target enterprise node;

[0121] The triplet data determination unit 6032 is used to convert the attribute information into triplet data according to the preset connection relationship between the target enterprise node and the attribute information node in the enterprise knowledge graph subgraph corresponding to the target enterprise node. The preset connection relationship is used to indicate the attribute relationship between the target enterprise corresponding to the target enterprise node and the attribute information corresponding to the attribute information node.

[0122] The prediction result acquisition unit 6033 is used to obtain the prediction result of the industry type of the target enterprise based on the triplet data and a preset enterprise industry prediction model.

[0123] In one example, the prediction result is obtained in unit 6033, which is specifically used for:

[0124] The triplet data is input into a preset enterprise industry forecasting model; wherein, the enterprise industry forecasting model is preset with at least two industry types;

[0125] Based on the pre-set convolutional layer in the enterprise industry prediction model, feature extraction is performed on the triplet data to obtain the confidence level of the target enterprise belonging to each industry type;

[0126] Based on the confidence level, the predicted industry type of the target company is determined.

[0127] In one example, the subgraph extraction module 602 is specifically used for:

[0128] The target enterprise node is identified as the first node, and the node connected to the first node is identified as the second node based on the connection relationship between the nodes in the enterprise knowledge graph; wherein, the second node includes at least one attribute information node;

[0129] The first node, the second node, and the connection relationship between the first node and the second node are determined as the enterprise knowledge graph subgraph corresponding to the first node.

[0130] In one example, the enterprise knowledge graph includes at least one industry node; the industry node is connected to the enterprise node, and the industry node is used to represent the industry type to which the connected enterprise node belongs;

[0131] The map update module 604 is specifically used for:

[0132] Detect whether there are industry nodes in the enterprise knowledge graph that correspond to the prediction results;

[0133] If so, connect the target enterprise node and the industry node corresponding to the prediction result with an edge to update the enterprise knowledge graph;

[0134] If not, add industry nodes corresponding to the prediction results to the enterprise knowledge graph, and connect the target enterprise node and the industry nodes corresponding to the prediction results with edges to update the enterprise knowledge graph.

[0135] In one example, the device also includes:

[0136] The first database writing module is used to write the target enterprise nodes with an unpredictable industry status into a preset first database after determining the target enterprise nodes with an unpredictable industry status among the enterprise nodes. The first database is used to record enterprise nodes that have not made predictions about their respective industry types.

[0137] In one example, the device also includes:

[0138] The second database writing module is used to delete the target enterprise node from the first database after obtaining the prediction result of the industry type to which the target enterprise node belongs, and associate the prediction result of the target enterprise node and the industry type to which the target enterprise belongs with the prediction result and write it into a preset second database. The second database is used to record the enterprise nodes whose industry type has been predicted and the corresponding industry type.

[0139] In one example, the device also includes:

[0140] The node addition module is used to add a new enterprise node in the enterprise knowledge graph in response to the enterprise node addition command, and add an industry status identifier of unpredictable state to the new enterprise node.

[0141] In one example, the device also includes:

[0142] The information adjustment module is used to respond to the attribute information adjustment command, determine the enterprise node corresponding to the attribute information to be adjusted, and determine the industry status of the enterprise node corresponding to the attribute information to be adjusted as unpredicted.

[0143] Figure 8 A structural block diagram of an electronic device provided in an embodiment of this application, such as... Figure 8 As shown, the electronic device includes a memory 81 and a processor 82. The memory 81 stores instructions executable by the processor 82. The processor 82 is configured to perform the methods provided in the embodiments described above.

[0144] The electronic device also includes a receiver 83 and a transmitter 84. The receiver 83 is used to receive instructions and data sent by other devices, and the transmitter 84 is used to send instructions and data to other devices.

[0145] Figure 9This is a structural block diagram of an electronic device according to an exemplary embodiment. The device may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, vehicle, etc.

[0146] Device 900 may include one or more of the following components: processing component 902, memory 904, power supply component 906, multimedia component 908, audio component 910, input / output (I / O) interface 912, sensor component 914, and communication component 916.

[0147] Processing component 902 typically controls the overall operation of device 900, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 902 may include one or more processors 920 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 902 may include one or more modules to facilitate interaction between processing component 902 and other components. For example, processing component 902 may include a multimedia module to facilitate interaction between multimedia component 908 and processing component 902.

[0148] Memory 904 is configured to store various types of data to support the operation of device 900. Examples of this data include instructions for any application or method operating on device 900, contact data, phonebook data, messages, pictures, videos, etc. Memory 904 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0149] Power supply component 906 provides power to various components of device 900. Power supply component 906 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to device 900.

[0150] Multimedia component 908 includes a screen that provides an output interface between the device 900 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 908 includes a front-facing camera and / or a rear-facing camera. When the device 900 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0151] Audio component 910 is configured to output and / or input audio signals. For example, audio component 910 includes a microphone (MIC) configured to receive external audio signals when device 900 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 904 or transmitted via communication component 916. In some embodiments, audio component 910 also includes a speaker for outputting audio signals.

[0152] I / O interface 912 provides an interface between processing component 902 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0153] Sensor assembly 914 includes one or more sensors for providing status assessments of various aspects of device 900. For example, sensor assembly 914 may detect the on / off state of device 900, the relative positioning of components such as the display and keypad of device 900, changes in the position of device 900 or a component of device 900, the presence or absence of user contact with device 900, the orientation or acceleration / deceleration of device 900, and temperature changes of device 900. Sensor assembly 914 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 914 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 914 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0154] Communication component 916 is configured to facilitate wired or wireless communication between device 900 and other devices. Device 900 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 916 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 916 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0155] In an exemplary embodiment, device 900 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0156] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 904 including instructions, which can be executed by a processor 920 of device 900 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0157] A non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a processor of a terminal device, enable the terminal device to execute the aforementioned method for updating the enterprise knowledge graph of the terminal device.

[0158] This application also discloses a computer program product, including a computer program that, when executed by a processor, implements the method described in this embodiment.

[0159] Various embodiments of the systems and technologies described above in this application can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0160] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or electronic device.

[0161] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0162] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0163] The systems and technologies described herein can be implemented in computing systems that include back-end components (e.g., as data electronic devices), or computing systems that include middleware components (e.g., application electronic devices), or computing systems that include front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0164] Computer systems can include client and electronic devices. Clients and electronic devices are generally geographically separated and typically interact via communication networks. The client-electronic device relationship is created by computer programs running on the respective computers and having a client-electronic device relationship with each other. The electronic device can be a cloud electronic device, also known as a cloud computing electronic device or cloud host, a host product within the cloud computing service system, addressing the shortcomings of traditional physical hosts and VPS services ("Virtual Private Server," or simply "VPS") in terms of management difficulty and weak business scalability. The electronic device can also be an electronic device in a distributed system or an electronic device incorporating blockchain technology. It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application is achieved, and this is not limited herein.

[0165] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the appended claims.

[0166] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for updating an enterprise knowledge graph, characterized in that, The enterprise knowledge graph includes at least one enterprise node and at least one attribute information node associated with the enterprise node, wherein the attribute information node is used to represent the enterprise's attribute information; the method includes: The industry status of each enterprise node in the enterprise knowledge graph is determined. Each enterprise node has an identifier for determining its industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted. Identify the target enterprise node in each enterprise node whose industry status is the unpredicted state, and extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph; Based on the attribute information corresponding to the attribute information node of the target enterprise node in the enterprise knowledge graph subgraph, the industry type of the target enterprise is predicted, and the prediction result is obtained. The enterprise knowledge graph is updated based on the prediction results, and the industry status of the target enterprise node is updated to the predicted status.

2. The method according to claim 1, characterized in that, The step of predicting the industry type of the target enterprise based on the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph, and obtaining the prediction result, includes: Determine the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph corresponding to the target enterprise node; Based on the preset connection relationship between the target enterprise node and the attribute information node in the enterprise knowledge graph subgraph corresponding to the target enterprise node, the attribute information is converted into triple data. The preset connection relationship is used to indicate the attribute relationship between the target enterprise corresponding to the target enterprise node and the attribute information corresponding to the attribute information node. Based on the triplet data and a preset enterprise industry prediction model, the prediction result of the industry type to which the target enterprise belongs is obtained.

3. The method according to claim 2, characterized in that, The step of obtaining the prediction result of the industry type of the target enterprise based on the triplet data and a preset enterprise industry prediction model includes: The triplet data is input into a preset enterprise industry forecasting model; wherein, the enterprise industry forecasting model is preset with at least two industry types; Based on the pre-set convolutional layer in the enterprise industry prediction model, feature extraction is performed on the triplet data to obtain the confidence level of the target enterprise belonging to each industry type; Based on the confidence level, the predicted industry type of the target company is determined.

4. The method according to claim 1, characterized in that, The step of extracting the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph includes: The target enterprise node is identified as the first node, and the node connected to the first node is identified as the second node based on the connection relationship between the nodes in the enterprise knowledge graph; wherein, the second node includes at least one attribute information node; The first node, the second node, and the connection relationship between the first node and the second node are determined as the enterprise knowledge graph subgraph corresponding to the first node.

5. The method according to claim 1, characterized in that, The enterprise knowledge graph also includes at least one industry node; the industry node is connected to the enterprise node, and the industry node is used to represent the industry type to which the connected enterprise node belongs; Updating the enterprise knowledge graph based on the prediction results includes: Detect whether there are industry nodes in the enterprise knowledge graph that correspond to the prediction results; If so, connect the target enterprise node and the industry node corresponding to the prediction result with an edge to update the enterprise knowledge graph; If not, add industry nodes corresponding to the prediction results to the enterprise knowledge graph, and connect the target enterprise node and the industry nodes corresponding to the prediction results with edges to update the enterprise knowledge graph.

6. The method according to claim 1, characterized in that, After identifying the target enterprise nodes whose industry status is unpredictable among the aforementioned enterprise nodes, the following is also included: Target enterprise nodes whose industry status is unpredictable are written into a preset first database; wherein, the first database is used to record enterprise nodes whose industry type has not been predicted.

7. The method according to claim 6, characterized in that, After predicting the industry type of the target company and obtaining the prediction results, the process also includes: The target enterprise node is deleted from the first database, and the prediction results of the target enterprise node and the industry type to which the target enterprise belongs are associated and written into a preset second database; wherein, the second database is used to record enterprise nodes whose industry type has been predicted and the corresponding industry type.

8. A device for updating an enterprise knowledge graph, characterized in that, The enterprise knowledge graph includes at least one enterprise node and at least one attribute information node associated with the enterprise node, wherein the attribute information node is used to represent the enterprise's attribute information; the device includes: The industry status determination module is used to determine the industry status of each enterprise node in the enterprise knowledge graph. Each enterprise node has an identifier for determining the industry status. The industry status includes a predicted status and an unpredicted status. The predicted status indicates that the industry type to which the enterprise belongs has been predicted, and the unpredicted status indicates that the industry type to which the enterprise belongs has not been predicted. The subgraph extraction module is used to identify the target enterprise node in each enterprise node whose industry status is the unpredicted state, and to extract the enterprise knowledge graph subgraph corresponding to the target enterprise node from the enterprise knowledge graph. The industry type prediction module is used to predict the industry type of the target enterprise based on the attribute information corresponding to the attribute information node in the enterprise knowledge graph subgraph, and obtain the prediction result. The graph update module is used to update the enterprise knowledge graph based on the prediction results and update the industry status of the target enterprise node to the predicted status.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the enterprise knowledge graph update method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method for updating the enterprise knowledge graph as described in any one of claims 1-7.

Images