Knowledge graph updating method and device, computer equipment and storage medium

Abstract

The invention discloses a knowledge graph updating method and device, computer equipment and a storage medium, relates to the technical field of data processing, and is used for solving the problems of low auditing efficiency and difficulty in ensuring the stability of node quality in related technologies. The method comprises the following steps: in response to a newly added target node in a knowledge graph of a target asset, obtaining a first data sequence of a first index associated with the target asset and a second data sequence of a second index associated with the target node; dividing the first data sequence and the second data sequence to obtain a plurality of first data subsequences and a plurality of second data subsequences; respectively calculating a first correlation coefficient between each first data sub-sequence and the corresponding second data sub-sequence, and calculating a second correlation coefficient between the first data sequence and the second data sequence according to the first correlation coefficient; and if the second correlation coefficient is greater than a first preset threshold, updating the target node to the knowledge graph.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a knowledge graph updating method and device, computer equipment and a storage medium. BACKGROUND

[0002] In the research of financial asset allocation, as a technology that can efficiently organize and correlate massive information, the knowledge graph is widely used to reveal the complex internal relationship between asset prices and various influencing factors. However, due to the problems of incomplete data and dynamic evolution of knowledge content in the knowledge graph itself, its content needs to be continuously updated and maintained. In the updating process, especially when customizing nodes of users to the knowledge graph, in order to ensure the overall content integrity and logical consistency of the knowledge graph, these customizing nodes must be audited. In related technologies, the auditing method is mainly manual auditing by domain experts, which is not only low in efficiency, but also difficult to ensure the stability of node quality. SUMMARY

[0003] Therefore, a knowledge graph updating method, device, computer equipment and storage medium are provided to solve the problem of low auditing efficiency and difficulty in ensuring the stability of node quality in related technologies.

[0004] In a first aspect, the present application provides a knowledge graph updating method, which comprises: In response to adding a target node in the knowledge graph of a target asset, obtaining a first data sequence of a first index associated with the target asset and a second data sequence of a second index associated with the target node, wherein the first data sequence and the second data sequence are aligned on a time axis; Dividing the first data sequence and the second data sequence to obtain a plurality of first data subsequences and a plurality of second data subsequences; Calculating a first correlation coefficient between each first data subsequence and the corresponding second data subsequence, and calculating a second correlation coefficient between the first data sequence and the second data sequence according to the first correlation coefficient; If the second correlation coefficient is greater than a first preset threshold, updating the target node to the knowledge graph.

[0005] In one embodiment, the dividing of the first data sequence and the second data sequence to obtain a plurality of first data subsequences and a plurality of second data subsequences comprises: Sampling the first time sequence and the second time sequence according to a preset window length and a preset sliding step to obtain the plurality of first data subsequences and the plurality of second data subsequences.

[0006] In one embodiment, the calculating, according to the first correlation coefficient, a second correlation coefficient between the first data sequence and the second data sequence comprises: counting a target number of the first correlation coefficients greater than a second preset threshold value in the first correlation coefficients between each of the first data subsequences and the corresponding second data subsequences; calculating a ratio between the target number and a total number of the first correlation coefficients to obtain the second correlation coefficient.

[0007] In one embodiment, before the acquiring the first data sequence of the first index associated with the target asset and the second data sequence of the second index associated with the target node, the method further comprises: acquiring target attribute information and target category information corresponding to the target node, wherein the target attribute information at least includes a second index name and a second index introduction text, and the target category information is used to indicate that the second index is an expression combination category or a data file category; performing a format specification review on the target attribute information to obtain a first review result, and performing a special review corresponding to the target category information to obtain a second review result; determining a formal review result of the target node according to the first review result and the second review result, and if the formal review result is passed, acquiring the first data sequence and the second data sequence.

[0008] In one embodiment, the performing the special review corresponding to the target category information to obtain the second review result comprises: if the target category information indicates that the second index is the expression combination category, performing a logic reasonableness review on a target expression of the second index to obtain the second review result; if the target category information indicates that the second index is the data file type, performing a content integrity review on a target data file associated with the second index to obtain the second review result.

[0009] In one embodiment, the updating the target node to the knowledge graph comprises: merging the target node to the knowledge graph based on a relationship edge between the target node and an existing node in the knowledge graph; allocating a target identifier to the target node, and storing the second data sequence based on the target identifier to bind the target node and the second data sequence through the target identifier.

[0010] In one embodiment, after the updating the target node to the knowledge graph, the method further comprises: in response to receiving the viewing request of the target node, highlighting the target node and the nodes and relationship edges associated with the target node; According to the target identifier, the second data sequence is obtained, and a data trend chart is generated based on the second data sequence.

[0011] In a second aspect, the present application provides a knowledge graph updating device, which comprises: The acquisition module is configured to, in response to adding a target node in a knowledge graph of a target asset, acquire a first data sequence of a first index associated with the target asset and a second data sequence of a second index associated with the target node, wherein the first data sequence and the second data sequence are aligned on a time axis. The division module is configured to divide the first data sequence and the second data sequence to obtain a plurality of first data subsequences and a plurality of second data subsequences. The calculation module is configured to calculate a first correlation coefficient between each first data subsequence and a corresponding second data subsequence, and calculate a second correlation coefficient between the first data sequence and the second data sequence according to the first correlation coefficient. The updating module is configured to, if the second correlation coefficient is greater than a first preset threshold, update the target node to the knowledge graph.

[0012] In a third aspect, the present application provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the knowledge graph updating method of the first aspect when executing the computer program.

[0013] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the knowledge graph updating method of the first aspect.

[0014] The knowledge graph updating method, device, computer device and storage medium can improve the correlation calculation accuracy by dividing the complete data sequence into a plurality of subsequences for segment correlation calculation and calculating the overall correlation between the first data sequence and the second data sequence based on the segment correlation calculation, so as to not only realize the automation of updating the knowledge graph, but also fundamentally guarantee the stability of the node quality and the data consistency of the knowledge graph when determining whether to update the target node to the knowledge graph based on the overall correlation. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 An application environment diagram of the knowledge graph updating method in an embodiment; Figure 2 A flowchart of a knowledge graph updating method in an embodiment; Figure 3 A structural block diagram of a knowledge graph updating device in an embodiment; Figure 4 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0016] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. The specific operation method in the method embodiment can also be applied to the device embodiment or the system embodiment. It should be noted that in the description of the present application, "multiple" is understood as "at least two". The association relationship of the associated objects is described, which means that there can be three relationships, for example, A and / or B, which means that there are three cases: A exists alone, A and B exist together, and B exists alone. A and B are connected, which means that A and B are directly connected and A and B are connected through C. In addition, in the description of the present application, "first", "second", etc. are used only for the purpose of distinguishing the description, and cannot be understood as indicating or implying relative importance, nor can it be understood as indicating or implying order.

[0017] In the present application, the acquisition, transmission, storage and use of data all meet the requirements of relevant national laws and regulations.

[0018] Before introducing the knowledge graph updating method provided by the present application, in order to facilitate understanding, the technical background of the present application will be described in detail below.

[0019] In the related art, the auditing method of the newly added nodes of the knowledge graph is mainly the manual auditing of the field experts. This method not only has low efficiency and cannot cope with the massive and high-speed generated nodes, but also has strong subjectivity of manual auditing, lacks unified and objective standards, and the auditing results of different experts can be quite different, so it is difficult to guarantee the stability of the node quality.

[0020] In view of this, the present application provides a knowledge graph updating method, device, computer device and storage medium, to solve the problems of low auditing efficiency and difficulty in guaranteeing the stability of node quality in the related art.

[0021] The application environment to which the technical solution of the present application can be applied will be briefly introduced below. It should be noted that the following introduction of the application environment is only used to illustrate the present application and is not limited. In specific implementation, the technical solution provided by the present application can be flexibly applied according to actual needs.

[0022] The knowledge graph updating method provided by the present application can be applied to, for example, Figure 1The application environment shown mainly includes a terminal 102 and a server 104. Among them, the terminal 102 and the server 104 can interact information through a communication network, and the communication network can adopt a communication mode including a wireless communication mode and a wired communication mode.

[0023] For example, the terminal 102 can access the network and communicate with the server 104 through a cellular mobile communication technology, which can include a 5th Generation Mobile Networks (5G) technology.

[0024] For example, the terminal 102 can access the network and communicate with the server 104 through a short-range wireless communication mode, which can include a Wireless Fidelity (Wi-Fi) technology.

[0025] The number of the above-mentioned devices is not limited in the present application, such as Figure 1 As shown, only the terminal 102 and the server 104 are taken as examples for description, and the above-mentioned devices and their respective functions are briefly introduced as follows.

[0026] The terminal 102 is a device that can provide voice and / or data connectivity for a target object, including a handheld terminal device with wireless connection function, a vehicle-mounted terminal device, etc.

[0027] For example, the terminal 102 includes but is not limited to a mobile phone, a tablet computer, a notebook computer, a palm computer, a Mobile Internet Device (MID), a wearable device, a Virtual Reality (VR) device, an Augmented Reality (AR) device, a wireless terminal device in industrial control, a wireless terminal device in unmanned driving, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, or a wireless terminal device in smart home, etc.

[0028] In addition, the terminal 102 can be installed with a knowledge graph management platform, which can be a software (for example, an application program, a browser, etc.), a webpage, an applet, etc. In the present application, the target object can log in the knowledge graph management platform on the terminal 102, and after the target object can interact information with the server 104 for data query, the corresponding result is fed back to the target object.

[0029] Exemplarily, the server 104 can be loaded with a back-end microservice corresponding to the knowledge graph management platform described above, and store various knowledge graphs and related data thereof. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0030] The technical solutions provided by the present application will be described in detail below in combination with the accompanying drawings and specific embodiments.

[0031] Figure 2 For a flowchart of the knowledge graph updating method in an embodiment, the flowchart can be executed by a knowledge graph updating device. The device can be implemented in a software manner, a hardware manner, or a combination of software and hardware. As shown in Figure 2 The flowchart includes the following steps: S201, in response to adding a target node in a knowledge graph of a target asset, acquiring a first data sequence of a first index associated with the target asset and a second data sequence of a second index associated with the target node; S202, dividing the first data sequence and the second data sequence to obtain a plurality of first data subsequences and a plurality of second data subsequences; S203, calculating a first correlation coefficient between each first data subsequence and the corresponding second data subsequence, and calculating a second correlation coefficient between the first data sequence and the second data sequence according to the first correlation coefficient; S204, if the second correlation coefficient is greater than a first preset threshold, updating the target node to the knowledge graph.

[0032] In the present application, a user can create a knowledge graph for different assets through a knowledge graph management platform on the terminal 102. The knowledge graph is composed of nodes and relationship edges between the nodes. The knowledge graph management platform adds a first label (such as personal, private, etc., specific labels are subject to circumstances, which are not limited here) to the knowledge graph created by the user, indicating that only the user who created the knowledge graph can edit it (such as deleting nodes, deleting relationship edges, modifying nodes, adding nodes, etc.).

[0033] The user can also share the created knowledge graph with another user. At this time, the knowledge graph management platform changes the first label of the knowledge graph to a second label (such as sharing, non-personal, etc., specific labels are subject to circumstances, which are not limited here), indicating that another user can view the knowledge graph but cannot edit it. Another user can also save the knowledge graph in another way to save the knowledge graph as a knowledge graph corresponding to another user containing the first label, so that another user can also edit the saved knowledge graph.

[0034] In addition, different users can create different knowledge graphs for the same asset on the knowledge graph management platform, distinguish different knowledge graphs by naming, and view different knowledge graphs by list selection.

[0035] The knowledge graph management platform also sets an asset administrator for each asset and sets different permissions for different roles (asset administrator, ordinary user). For example, the asset administrator can view all knowledge graphs created on the knowledge graph management platform, and the ordinary user can only view the knowledge graphs created by himself or shared by others. The asset administrator can also publish the created knowledge graph. At this time, the knowledge graph management platform adds a third label (such as public, general, etc., the specific label is subject to circumstances, not limited here) to the published knowledge graph, indicating that different users can view the published knowledge graph, but only the asset administrator can directly edit the published knowledge graph, and the ordinary user can only add nodes to the published public knowledge graph and submit for audit, and after the audit is passed, the nodes can be merged into the published knowledge graph. The specific permission setting method is subject to circumstances, not limited here.

[0036] Among them, the asset is associated with the data sequence of the corresponding first index, such as a stock associated with a stock price, the nodes in the knowledge graph of the asset include type nodes and index nodes, the type nodes are not associated with indexes, and the index nodes are associated with the data sequence of the corresponding second index. The nodes added by the user in the published knowledge graph are usually index nodes.

[0037] For example, in the knowledge graph of a stock, the type nodes can include industry type nodes (such as technology industry, financial industry, etc.), event type nodes (such as M country central bank interest rate hike, release of new products, etc.), and the index nodes can include financial index nodes (such as quarterly revenue, P / E ratio, etc.), macro index nodes (such as M country 10-year treasury yield, M country currency index, etc.), and their connection relationships can include: stock - belongs to (technology industry) - has (smartphone business) - occurs (release of new mobile phones) - affects (quarterly revenue) - affects (stock); stock - is affected by (P / E ratio), stock - belongs to (financial industry) - is affected by (M country central bank interest rate hike) - leads to (M country 10-year treasury yield rises) - negatively affects (P / E ratio) - affects (stock).

[0038] Exemplarily, the general user logs in the knowledge graph management platform and can view the published knowledge graph of the target asset. At this time, the general user newly creates a target node (target node) and the relationship edge between the target node and the existing node in the knowledge graph, and initiates a request for adding the target node in the knowledge graph of the target asset by clicking the submission. In response to receiving the request for adding the target node in the knowledge graph of the target asset, the first data sequence of the first index associated with the target asset and the second data sequence of the second index associated with the target node are obtained.

[0039] For example, the target asset is a stock, and the target node is a research and development intensity node. The stock is associated with a closing price sequence, and the research and development intensity node is associated with a research and development intensity value sequence. The closing price sequence is taken as the first data sequence, and the research and development intensity value sequence is taken as the second data sequence, or the closing price sequence in a preset time range (the specific range is determined as appropriate, which is not limited here) is taken as the first data sequence, and the research and development intensity value sequence in the preset time range is taken as the second data sequence. The specific acquisition method is determined as appropriate, which is not limited here.

[0040] The first data sequence and the second data sequence are aligned on the time axis. The alignment method can be a missing value filling method, such as forward filling and backward filling. The specific alignment method is determined as appropriate, which is not limited here.

[0041] Then, the first data sequence and the second data sequence are divided to obtain a plurality of first data subsequences and a plurality of second data subsequences, and the first correlation coefficient between each first data subsequence and the corresponding second data subsequence is calculated, and the second correlation coefficient between the first data sequence and the second data sequence is calculated according to the first correlation coefficient. The first correlation coefficient can be a Pearson correlation coefficient or a rank correlation coefficient. The specific method is determined as appropriate, which is not limited here.

[0042] Finally, it is determined whether the second correlation coefficient is greater than a first preset threshold value (the specific value is determined as appropriate, which is not limited here). If yes, the target node is updated to the knowledge graph of the target asset. If not, the user request for adding the target node is rejected, or it is prompted that manual review is required.

[0043] By the above method, the complete data sequence is divided into a plurality of subsequences for segment correlation calculation, and the overall correlation between the first data sequence and the second data sequence is calculated based on the segment correlation calculation. This can improve the correlation calculation accuracy, so that when the overall correlation is determined based on the overall correlation to determine whether the target node is updated to the knowledge graph, not only the automation of updating the knowledge graph can be realized, but also the stability of the node quality and the data consistency of the knowledge graph are fundamentally guaranteed.

[0044] In an embodiment, the first data sequence and the second data sequence are divided into a plurality of first data subsequences and a plurality of second data subsequences, including but not limited to: The first time sequence and the second time sequence are sampled according to the preset window length and the preset sliding step to obtain the plurality of first data subsequences and the plurality of second data subsequences. When the preset window length is equal to the preset sliding step, the plurality of first data subsequences and the plurality of second data subsequences are obtained, which are non-overlapping. When the preset window length is greater than the preset sliding step, the plurality of first data subsequences and the plurality of second data subsequences are obtained, which have overlapping parts. The specific values of the preset window length and the preset sliding step are determined according to the situation, which are not limited here. By the above method, the continuous and possibly overlapping data subsequences are generated based on the sliding window sampling method, which can more densely capture the local correlation of the nodes and the assets in different time segments, thereby improving the robustness of the global correlation analysis.

[0045] In an embodiment, the second correlation coefficient between the first data sequence and the second data sequence is calculated according to the first correlation coefficient, including but not limited to: The target number of the first correlation coefficients greater than the second preset threshold (the specific value is determined according to the situation, which is not limited here) between each first data subsequence and the corresponding second data subsequence is counted.

[0046] The ratio between the target number and the total number of the first correlation coefficients is calculated to obtain the second correlation coefficient between the first data sequence and the second data sequence.

[0047] Optionally, the target number can also be directly determined as the second correlation coefficient between the first data sequence and the second data sequence. At this time, the first preset threshold is determined according to the total number of the first correlation coefficients, and the specific determination method of the second correlation coefficient is determined according to the situation, which is not limited here.

[0048] By the above method, the global correlation coefficient is synthesized based on the number of local correlation strengths, which can ensure the significant correlation between the nodes and the assets, improve the stability and reliability of the knowledge graph node access, and avoid false judgments caused by accidental or short-term fluctuations.

[0049] In an embodiment, before obtaining the first data sequence of the first index related to the target asset and the second data sequence of the second index related to the target node, the following steps are further included: The target attribute information and the target category information corresponding to the target node are obtained, wherein the target attribute information at least includes the second index name and the second index introduction text, and the target category information is used to indicate whether the second index is an expression combination category or a data file category. performing a format specification review on the target attribute information to obtain a first review result, and performing a special review corresponding to the target category information to obtain a second review result.

[0050] According to the first review result and the second review result, a formal review result of the target node is determined. For example, when the first review result and the second review result are both passed, it is determined that the formal review result of the target node is passed; when the first review result or the second review result is not passed, it is determined that the formal review result of the target node is not passed.

[0051] If the formal review result of the target node is passed, the first data sequence and the second data sequence are obtained; if the formal review result is not passed, the user request for adding the target node is rejected, or it is prompted that manual review is required.

[0052] Through the above method, before performing data correlation analysis, the target attribute information and the target category information of the node are subjected to dual review of format specification and category speciality, ensuring the compliance of the node in terms of basic data quality and business logic, thereby improving the efficiency of knowledge graph node review, avoiding invalid or format error data from entering the subsequent calculation process from the source, and ensuring the quality and stability of the knowledge graph.

[0053] In one embodiment, the target attribute information is subjected to a format specification review to obtain a first review result, including but not limited to: performing a first format specification review on the second index name to obtain a first specification review result. The first format specification review at least includes a name duplication review, a string length review, and an abnormal character review, which are not limited here.

[0054] For example, the second index name is compared with the existing index names in the knowledge graph, the string length of the second index name is calculated, and the second index name is subjected to regular expression matching. If no duplicate index name is found, the string length is less than a preset length (the specific value is determined according to the situation and is not limited here), and no abnormal character is matched, it is determined that the first specification review result is passed; if a duplicate name is found, the string length is greater than or equal to the preset length, or an abnormal character is matched, it is determined that the first specification review result is not passed, and the user request for adding the target node is rejected or it is prompted that manual review is required. The comparison method of the index name can be: The similarity between the second index name and each existing index name is calculated respectively. If there is an index name with a similarity greater than a first preset similarity (the specific value is determined according to the situation, which is not limited here), it is determined that there is a duplicate index name. If there is no index name with a similarity greater than the first preset similarity, it is determined that there is no duplicate index name. Or; First, all existing index names are added to the Bloom filter after being processed by a hash function. Then, the Bloom filter is queried based on the second index name. If there is no return, it is directly determined that there is no duplicate name. If there is a possible return, the similarity between the second index name and each existing index name is calculated respectively. If there is an index name with a similarity greater than the first preset similarity, it is determined that there is a duplicate index name. If there is no index name with a similarity greater than the first preset similarity, it is determined that there is no duplicate index name. The specific comparison method is determined according to the situation, which is not limited here.

[0055] Through the above method, based on the three reviews of name duplication, length control and abnormal characters, the uniqueness, simplicity and standardization of node names can be ensured, and data redundancy, storage chaos and retrieval conflicts caused by naming problems can be eliminated from the source, effectively maintaining the standardization and quality of the knowledge graph.

[0056] The second index introduction text is subjected to a second format specification review, and a second specification review result is obtained. The second format specification review at least includes a duplication review and a word count review of the introduction text, and the specific method is determined according to the situation, which is not limited here.

[0057] For example, the second index introduction text is compared with the existing index introduction text in the knowledge graph, and the target word count (excluding symbols) of the second index introduction text is counted. If there is no duplicate introduction text, and the target word count is greater than or equal to a preset word count (the specific value is determined according to the situation, which is not limited here), it is determined that the second specification review result is passed. If there is a duplicate introduction text, or the target word count is less than the preset word count, it is determined that the second specification review result is not passed, and the user request for adding the target node is rejected or it is prompted that manual review is required. The comparison method of the introduction text can be: The second index introduction text is compared with the existing index introduction text, and the semantic similarity is calculated. If there is an introduction text with a semantic similarity greater than a second preset similarity (the specific value is determined according to the situation, which is not limited here), it is determined that there is a duplicate introduction text. If there is no introduction text with a semantic similarity greater than the second preset similarity, it is determined that there is no duplicate introduction text. The specific comparison method is determined according to the situation, which is not limited here.

[0058] By the above method, the repeatability of the introduction text is detected, and it is ensured that the number of words is not less than the preset number of words, so that the description information of the node can have necessary uniqueness and detail, thereby effectively improving the value and understandability of the knowledge graph, and avoiding information redundancy and low quality problems caused by repeated content or too brief description.

[0059] Therefore, based on the first normative review result and the second normative review result, the first review result of the target attribute information is determined. For example, if the first normative review result and the second normative review result are both passed, it is determined that the first review result is passed; if the first normative review result or the second normative review result is not passed, it is determined that the first review result is not passed.

[0060] By the above method, the name and introduction text of the index are subjected to format normative review, ensuring the standardization and readability of the basic metadata.

[0061] In one embodiment, it is exemplarily illustrated that the special review corresponding to the target category information is performed to obtain a second review result, including but not limited to: If the target category information indicates that the second index is an expression combination category, a logical reasonableness review is performed on the target expression of the second index to obtain a second review result.

[0062] For example, the historical data corresponding to the target expression is obtained, and a simulation operation is performed based on the historical data and the target expression to obtain a simulation operation result. If the simulation operation result is within a preset reasonable range (the specific range is determined according to the situation, which is not limited here), it is determined that the second review result is passed; if the simulation operation result is not within the preset reasonable range, it is determined that the second review result is not passed, and the user request for adding the target node is rejected or it is prompted that manual review is required.

[0063] If the target category information indicates that the second index is a data file type, a content integrity review is performed on the target data file associated with the second index to obtain a second review result.

[0064] For example, the target data file is parsed to obtain a target format corresponding to the target data file, target data content, and a time coverage range calculated based on the target data content; If the target format is a specified format (such as Excel, the specific format is subject to circumstances, not limited here), the target data content contains structured basic information, and the time coverage range is greater than or equal to the target time range (the specific range is subject to circumstances, not limited here), then the second review result is passed, wherein the basic information at least includes a timestamp sequence and a corresponding data sequence; if the target format is not a specified format, or the target data content does not completely contain structured basic information, or the time coverage range is less than the target time range, then the second review result is not passed, and the user request for adding the target node is rejected or it is prompted that manual review is required.

[0065] Through the above method, the logical rationality of the expression of the expression combination type index is reviewed, and logical errors are eliminated from the source. For data file type indexes, the integrity of the data file is reviewed to prevent data loss, thereby improving the standardization, logical rigor, and reliability of the node.

[0066] In one embodiment, illustratively, updating the target node to the knowledge graph includes but is not limited to: Merging the target node into the knowledge graph based on the relationship edges between the target node and the existing nodes in the knowledge graph.

[0067] Assigning a target identifier to the target node, and storing the second data sequence based on the target identifier using a graph database, to bind the target node and the second data sequence through the target identifier.

[0068] Through the above method, a standardized node access mechanism is established, and the node and its data sequence are embedded into the knowledge graph on the premise of ensuring the uniqueness of the global identifier, realizing the systematic integration and traceable management of massive heterogeneous data.

[0069] In one embodiment, illustratively, after updating the target node to the knowledge graph, it further includes: In response to receiving a viewing request for the target node, highlighting the target node and the nodes and relationship edges associated with the target node. For example, displaying the updated knowledge graph of the target asset in the visualization interface, and the user initiates a viewing request by clicking the target node in the updated knowledge graph, and in response to receiving the viewing request for the target node, highlighting the target node and the nodes and relationship edges associated with the target node.

[0070] At the same time, obtaining the target identifier of the target node, and according to the target identifier, obtaining the target data sequence corresponding to the target node, and using a chart engine to render and generate a data trend chart (such as a line chart, a column chart, etc., the specific format is subject to circumstances, not limited here) based on the target data sequence, and displaying the data trend chart.

[0071] Optionally, the user comment list corresponding to the target node can also be obtained through the target identifier, and the user comment list is displayed.

[0072] For example, the data trend chart and the user comment list are displayed in a sub-interface of the visualization interface, and the sub-interface includes at least a first area for displaying the data trend chart, and a second area for displaying the user comment list, and a rich text edit box is provided at the bottom of the user comment list to enable a user to post a new comment, and the new comment is stored in association with the target identifier. The specific display method is subject to circumstances and is not limited here.

[0073] Through the above method, based on the linkage display of the highlighted display of the node and the multi-dimensional data, the deep information penetration of any node in the knowledge graph can be realized, and the user can quickly cut into the specific numerical trend and community opinion from the global topological relationship, thereby effectively improving the explorability of the complex knowledge graph and the data interpretation efficiency.

[0074] It should be understood that, although Figure 2 the steps in the flowchart are shown in order according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, Figure 2 at least part of the steps in the flowchart can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0075] In one embodiment, as shown in Figure 3 , a knowledge graph updating apparatus is provided, comprising: an acquisition module 301, a division module 302, a calculation module 303, and an updating module 304, wherein: The acquisition module 301 is configured to, in response to adding a target node in a knowledge graph of a target asset, acquire a first data sequence of a first index associated with the target asset and a second data sequence of a second index associated with the target node, wherein the first data sequence and the second data sequence are aligned on a time axis; The division module 302 is configured to divide the first data sequence and the second data sequence to obtain a plurality of first data subsequences and a plurality of second data subsequences; The calculation module 303 is configured to calculate a first correlation coefficient between each first data subsequence and a corresponding second data subsequence, and calculate a second correlation coefficient between the first data sequence and the second data sequence according to the first correlation coefficient; The update module 304 is used to update the target node to the knowledge graph if the second correlation coefficient is greater than the first preset threshold.

[0076] In one embodiment, the partitioning module 302 is used for: The first time series and the second time series are sampled according to the preset window length and the preset sliding step size to obtain multiple first data subsequences and multiple second data subsequences.

[0077] In one embodiment, the calculation module 303 is used for: Among the first correlation coefficients between each first data subsequence and its corresponding second data subsequence, the number of target numbers whose first correlation coefficients are greater than a second preset threshold is calculated. The second correlation coefficient is obtained by calculating the ratio between the number of targets and the total number of the first correlation coefficients.

[0078] In one embodiment, the acquisition module 301 is further configured to: Obtain the target attribute information and target category information corresponding to the target node. The target attribute information includes at least the name of the second indicator and the description text of the second indicator. The target category information is used to indicate whether the second indicator is an expression combination category or a data file category. Perform a format standardization review on the target attribute information to obtain the first review result, and then perform a special review on the target category information to obtain the second review result; Based on the first and second review results, determine the formal review result of the target node. If the formal review result is passed, then obtain the first data sequence and the second data sequence.

[0079] In one embodiment, the acquisition module 301 is further configured to: If the target category information indicates that the second indicator is an expression combination category, then the logical rationality of the target expression of the second indicator is reviewed to obtain the second review result; If the target category information indicates that the second indicator is a data file type, then the content integrity of the target data file associated with the second indicator is reviewed to obtain the second review result.

[0080] In one embodiment, the update module 304 is used for: Based on the relationship edges between the target node and existing nodes in the knowledge graph, the target node is merged into the knowledge graph; Assign a target identifier to the target node and store a second data sequence based on the target identifier, so as to bind the target node to the second data sequence through the target identifier.

[0081] In one embodiment, the update module 304 is further configured to: In response to receiving the viewing request of the target node, the target node and the nodes and relationship edges associated with the target node are highlighted; According to the target identifier, a second data sequence is acquired, and a data trend chart is generated based on the second data sequence.

[0082] The specific limitations of the knowledge graph updating apparatus can refer to the limitations of the knowledge graph updating method in the foregoing, and will not be described herein. Each module in the knowledge graph updating apparatus can be realized by software, hardware, or a combination thereof, in whole or in part. Each module can be embedded in or independent of a processor in a computer device in a hardware form, or can be stored in a memory in a computer device in a software form, so as to be called and executed by a processor to perform the operations corresponding to each module.

[0083] In one embodiment, a computer device, which can be a server, has an internal structure diagram as shown in Figure 4 The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store knowledge graph updating data. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a knowledge graph updating method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball, or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0084] Those skilled in the art can understand that Figure 4 The structure shown in the foregoing

[0085] In one embodiment, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented: In response to adding the target node in the knowledge graph of the target asset, a first data sequence of a first index associated with the target asset and a second data sequence of a second index associated with the target node are obtained, wherein the first data sequence and the second data sequence are aligned on a time axis; The first data sequence and the second data sequence are divided to obtain a plurality of first data subsequences and a plurality of second data subsequences; A first correlation coefficient between each first data subsequence and a corresponding second data subsequence is calculated respectively, and a second correlation coefficient between the first data sequence and the second data sequence is calculated according to the first correlation coefficient; If the second correlation coefficient is greater than a first preset threshold, the target node is updated to the knowledge graph.

[0086] In one embodiment, the processor executing the computer program further implements the following steps: The first time sequence and the second time sequence are sampled according to a preset window length and a preset sliding step to obtain a plurality of first data subsequences and a plurality of second data subsequences.

[0087] In one embodiment, the processor executing the computer program further implements the following steps: The target number of first correlation coefficients greater than a second preset threshold among the first correlation coefficients between each first data subsequence and a corresponding second data subsequence is counted; The ratio between the target number and the total number of first correlation coefficients is calculated to obtain a second correlation coefficient.

[0088] In one embodiment, the processor executing the computer program further implements the following steps: Target attribute information and target category information corresponding to the target node are obtained, wherein the target attribute information at least includes a second index name and a second index introduction text, and the target category information is used to indicate that the second index is an expression combination category or a data file category; A format specification review of the target attribute information is performed to obtain a first review result, and a special review corresponding to the target category information is performed to obtain a second review result; According to the first review result and the second review result, a formal review result of the target node is determined, and if the formal review result is passed, the first data sequence and the second data sequence are obtained.

[0089] In one embodiment, the processor executing the computer program further implements the following steps: If the target category information indicates that the second index is an expression combination category, a logic reasonableness review of a target expression of the second index is performed to obtain a second review result; If the target category information indicates that the second index is a data file type, a content integrity review is performed on a target data file associated with the second index, and a second review result is obtained.

[0090] In one embodiment, the processor, when executing the computer program, also implements the following steps: Based on the relationship edges between the target node and the existing nodes in the knowledge graph, the target node is merged into the knowledge graph. A target identifier is assigned to the target node, and the second data sequence is stored based on the target identifier, so as to bind the target node and the second data sequence through the target identifier.

[0091] In one embodiment, the processor, when executing the computer program, also implements the following steps: In response to receiving a viewing request for the target node, the target node and the nodes and relationship edges associated with the target node are highlighted; According to the target identifier, the second data sequence is obtained, and a data trend chart is generated based on the second data sequence.

[0092] In one embodiment, a computer-readable storage medium is provided, and the computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the following steps are implemented: In response to adding a target node in the knowledge graph of the target asset, a first data sequence of a first index associated with the target asset and a second data sequence of a second index associated with the target node are obtained, wherein the first data sequence and the second data sequence are aligned on a time axis; The first data sequence and the second data sequence are divided to obtain a plurality of first data subsequences and a plurality of second data subsequences; The first correlation coefficients between each first data subsequence and the corresponding second data subsequence are calculated respectively, and the second correlation coefficient between the first data sequence and the second data sequence is calculated according to the first correlation coefficients; If the second correlation coefficient is greater than a first preset threshold, the target node is updated to the knowledge graph.

[0093] In one embodiment, the computer program, when executed by the processor, also implements the following steps: The first time sequence and the second time sequence are sampled according to a preset window length and a preset sliding step to obtain a plurality of first data subsequences and a plurality of second data subsequences.

[0094] In one embodiment, the computer program, when executed by the processor, also implements the following steps: The target number of first correlation coefficients greater than a second preset threshold among the first correlation coefficients between each first data subsequence and the corresponding second data subsequence is counted. A ratio between the target number and the first total number of correlation coefficients is calculated to obtain a second correlation coefficient.

[0095] In one embodiment, the computer program, when executed by the processor, further implements the following steps: Obtaining target attribute information and target category information corresponding to the target node, wherein the target attribute information at least includes a second index name and a second index introduction text, and the target category information is used to indicate that the second index is an expression combination category or a data file category; Performing a format specification review on the target attribute information to obtain a first review result, and performing a special review corresponding to the target category information to obtain a second review result; According to the first review result and the second review result, determining a formal review result of the target node, and if the formal review result is passed, obtaining the first data sequence and the second data sequence.

[0096] In one embodiment, the computer program, when executed by the processor, further implements the following steps: If the target category information indicates that the second index is an expression combination category, performing a logical reasonableness review on a target expression of the second index to obtain a second review result; If the target category information indicates that the second index is a data file type, performing a content integrity review on a target data file associated with the second index to obtain a second review result.

[0097] In one embodiment, the computer program, when executed by the processor, further implements the following steps: Based on the relationship edges between the target node and the existing nodes in the knowledge graph, the target node is merged into the knowledge graph; Assigning a target identifier to the target node, and storing the second data sequence based on the target identifier, so as to bind the target node and the second data sequence through the target identifier.

[0098] In one embodiment, the computer program, when executed by the processor, further implements the following steps: In response to receiving a viewing request of the target node, highlighting the target node and the nodes and relationship edges associated with the target node; According to the target identifier, obtaining the second data sequence, and generating a data trend chart based on the second data sequence.

[0099] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiment methods can be included. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0100] Any combination of the technical features of the above embodiments can be made, and in order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0101] The above embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A knowledge graph updating method, characterized in that, The method includes: In response to adding a target node in the knowledge graph of the target asset, a first data sequence of a first indicator associated with the target asset and a second data sequence of a second indicator associated with the target node are obtained, wherein the first data sequence and the second data sequence are aligned on the time axis; The first data sequence and the second data sequence are divided to obtain multiple first data subsequences and multiple second data subsequences; Calculate the first correlation coefficient between each of the first data subsequences and the corresponding second data subsequences, and calculate the second correlation coefficient between the first data sequence and the second data sequence based on the first correlation coefficient. If the second correlation coefficient is greater than the first preset threshold, then the target node is updated to the knowledge graph.

2. The method according to claim 1, characterized in that, The step of dividing the first data sequence and the second data sequence to obtain multiple first data subsequences and multiple second data subsequences includes: The first time series and the second time series are sampled according to a preset window length and a preset sliding step size to obtain the plurality of first data subsequences and the plurality of second data subsequences.

3. The method according to claim 1, characterized in that, The step of calculating the second correlation coefficient between the first data sequence and the second data sequence based on the first correlation coefficient includes: Among the first correlation coefficients between each first data subsequence and its corresponding second data subsequence, the number of target numbers whose first correlation coefficients are greater than a second preset threshold is calculated. The second correlation coefficient is obtained by calculating the ratio between the target quantity and the total number of the first correlation coefficients.

4. The method according to claim 1, characterized in that, Before obtaining the first data sequence of the first indicator associated with the target asset and the second data sequence of the second indicator associated with the target node, the method further includes: Obtain the target attribute information and target category information corresponding to the target node, wherein the target attribute information includes at least the second indicator name and the second indicator description text, and the target category information is used to indicate that the second indicator is an expression combination category or a data file category; The target attribute information is subjected to a format standardization review to obtain a first review result, and the target category information is subjected to a special review to obtain a second review result; Based on the first review result and the second review result, the formal review result of the target node is determined. If the formal review result is passed, the first data sequence and the second data sequence are obtained.

5. The method according to claim 4, characterized in that, The execution of the special review corresponding to the target category information yields a second review result, including: If the target category information indicates that the second indicator is the expression combination category, then the target expression of the second indicator is subjected to a logical rationality review to obtain the second review result; If the target category information indicates that the second indicator is a data file type, then the content integrity of the target data file associated with the second indicator is reviewed to obtain the second review result.

6. The method according to claim 1, characterized in that, The step of updating the target node to the knowledge graph includes: Based on the relationship edges between the target node and existing nodes in the knowledge graph, the target node is merged into the knowledge graph; Assign a target identifier to the target node and store the second data sequence based on the target identifier, so as to bind the target node to the second data sequence through the target identifier.

7. The method according to claim 6, characterized in that, After updating the target node to the knowledge graph, the method further includes: In response to receiving a viewing request for the target node, the target node, as well as the nodes and relationship edges associated with the target node, are highlighted. Based on the target identifier, the second data sequence is obtained, and a data trend chart is generated based on the second data sequence.

8. A knowledge graph updating device, characterized in that, The device includes: The acquisition module is used to acquire a first data sequence of a first indicator associated with the target asset and a second data sequence of a second indicator associated with the target node in response to the addition of a target node in the knowledge graph of the target asset, wherein the first data sequence and the second data sequence are aligned on the time axis; A partitioning module is used to partition the first data sequence and the second data sequence to obtain multiple first data subsequences and multiple second data subsequences; The calculation module is used to calculate the first correlation coefficient between each of the first data subsequence and the corresponding second data subsequence, and to calculate the second correlation coefficient between the first data sequence and the second data sequence based on the first correlation coefficient. An update module is used to update the target node to the knowledge graph if the second correlation coefficient is greater than a first preset threshold.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images