A method of knowledge management and representation

Abstract

The application provides a knowledge management and expression method, relates to the technical field of data information management, and comprises the steps of judging whether a mode layer and knowledge data are matched, generating a graph structure framework, obtaining core features of the mode layer, and generating a knowledge graph containing knowledge point nodes, correlation relationship edges and visual elements, etc.The application aims at the weak explainability of the existing knowledge graph, converts abstract knowledge features into intuitive visual elements through multi-dimensional visual rules, records specific data by matching node feature forms, and ensures logical compliance through node text matching degree and knowledge correlation index checking; solves the noise sensitivity problem, filters invalid correlations through matching degree and correlation index threshold, eliminates isolated nodes, effectively reduces entity connection errors, relationship ambiguity and other noises, and guarantees the accuracy and reliability of the graph; and the graph is split into connected subgraphs to realize block management, greatly reduces query delay, and improves use efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data information management, and particularly relates to a knowledge management and expression method. BACKGROUND

[0002] In the field of knowledge management, the core requirement is to transform scattered fragmented information into a knowledge network with structured organization, correlation mapping and reasoning ability, so as to support efficient landing of business decision, information retrieval and intelligent application. As a key technology for representing and reasoning structured knowledge, the knowledge graph has been widely applied to search engines, intelligent question answering, recommendation systems and other fields. Its essence is a structured semantic knowledge base that clearly describes the correlation between entities and attribute information in the real world through a graph structure, providing a basic framework for the systematic storage, transmission and application of knowledge.

[0003] The current knowledge graph has the following problems: weak interpretability, simple vectorization representation cannot intuitively express business; noise sensitivity, entity connection error, relationship extraction ambiguity; low data query efficiency, high query delay of large-scale graph, therefore, a knowledge management and expression method is proposed. SUMMARY

[0004] To solve the above technical problems, the present application provides a knowledge management and expression method, which can make business knowledge explicit through different forms of expression, making the relationship between business knowledge and knowledge and the reasoning process more transparent, especially in fields or business scenarios with strong professionalism, the interpretability and traceability are stronger; the value of business knowledge is reflected by managing the strength of the relationship, the usage frequency of the knowledge point and the importance of the usage scenario, thereby promoting the efficient use of business knowledge in actual business scenarios.

[0005] The technical scheme adopted by the present application is as follows: a knowledge management and expression method, comprising the following steps:

[0006] S1, acquiring knowledge data entities, constructing a mode layer of a knowledge graph based on the knowledge data entities, and calculating the matching degree of the knowledge data entity node text and the mode layer node text, when the matching degree of the knowledge data entity node text and the mode layer node text is greater than a set matching threshold, it is determined that the mode layer is adapted to the knowledge data entities;

[0007] S2, mapping the position of the mode layer to generate a graph structure framework, calculating the knowledge correlation index of the mode layer, when the knowledge correlation index is greater than a correlation threshold, it is determined that the graph structure framework is correlated; meanwhile, the core features of the mode layer are acquired, the core features including usage frequency and scene importance;

[0008] S3. Generate a knowledge graph containing knowledge point nodes, relational edges, and visualization elements based on the graph structure framework, pattern layer knowledge relevance index, and core features; preset multi-dimensional visualization rules and determine the correspondence between the multi-dimensional visualization rules and the visualization element knowledge graph;

[0009] S4. Exclude isolated nodes in the knowledge graph of visualized elements in S3, randomly select a non-isolated node, search for its connected nodes by traversal, and mark it as a connected subgraph; repeat the traversal process until all non-isolated nodes are marked, and obtain multiple connected subgraphs.

[0010] S5. Calculate the spatial attributes of the knowledge data entity graph in each connected subgraph. The spatial attributes of the knowledge data entity graph include the number of nodes associated with each node, the average length of the shortest path from a node to other nodes, and the number of times a node acts as a bridge for the shortest path from other nodes. Input the calculated number of nodes associated with each node, the average length of the shortest path from a node to other nodes, the number of times a node acts as a bridge for the shortest path from other nodes, the usage frequency obtained in step S2, and the scene importance into the LR model to calculate the knowledge data entity score. Compare the calculated knowledge data entity score with the set knowledge data entity threshold, mark important knowledge data entities, and divide them into hierarchical entities based on important knowledge data entities.

[0011] Further, in step S1, obtaining knowledge data entities involves extracting knowledge data entities from the knowledge data using an entity recognition algorithm and extracting the relationships between the knowledge data entities using a relation extraction algorithm; the entity recognition algorithm is the BERT-CRF algorithm, and the relation extraction algorithm is the PCNN algorithm.

[0012] Further, in step S1, the formula for calculating the node text matching degree is:

[0013]

[0014] In the formula, The number of the preset time period. For the first The matching degree of node text within a preset time period. It is a natural constant. For the first The string matching degree for a preset time period. For string reference matching degree, For the first Semantic similarity over a preset time period For semantic reference similarity, For the first Attribute validation values ​​for a preset time period. This is a reference validation value for the attribute. is an attribute verification correction factor.

[0015] Further, in step S2, the calculation formula of the knowledge correlation index is:

[0016]

[0017] wherein, is the number of a preset time period, is the knowledge correlation index of the th preset time period, is a natural constant, is the entity matching degree of the th preset time period, is the entity reference matching degree, is the relationship consistency of the th preset time period, is the relationship reference consistency, is the concept coverage of the th preset time period.

[0018] Further, in step S2, the multi-dimensional visualization rule is:

[0019] Node size mapping: 5 fixed node sizes are preset, the maximum usage frequency of all knowledge data entities is obtained, the frequency interval is divided by dividing the maximum usage frequency by 5, and the knowledge point node matching the corresponding size is used for the knowledge point node falling into different intervals; if the maximum usage frequency <10, all knowledge point nodes use the size corresponding to the maximum interval;

[0020] Node color mapping: 5 gradient colors are preset, the maximum scene importance index of all knowledge data entities is obtained, the importance interval is divided by dividing the maximum index by 5, and the knowledge point node matching the corresponding color is used for the scene importance falling into different intervals; if the maximum index <10, all knowledge point nodes use the color corresponding to the maximum interval;

[0021] Node line number mapping: 3 kinds of node description lines, i.e., thin line, medium thick line and thick line, are preset, which correspond to the low, medium and important three levels of the scene importance of the knowledge data entity respectively, and the corresponding line is matched after determining the level;

[0022] Association relationship line type mapping: 4 kinds of association relationship line types are preset, and the corresponding line type is matched after determining the type according to the business process dependence degree;

[0023] Characteristic form configuration: a small form is configured for each knowledge point node to record the specific numerical value or field description of the usage frequency and scene importance;

[0024] The obtained visualization element corresponds to the core feature of the knowledge point in the knowledge graph.

[0025] Further, in step S4, obtaining the connected subgraph specifically comprises the following steps:

[0026] S41: traverse the visual element knowledge graph in S3, exclude isolated nodes without any association relationship, and mark the remaining non-isolated nodes to form a non-isolated node set;

[0027] S42: randomly select an initial node from the non-isolated node set, search all nodes directly or indirectly connected to the initial node by traversing, and mark the initial node and its connected nodes as a connected subgraph;

[0028] S43: repeat the operation of S42 until all nodes in the non-isolated node set are marked, and a plurality of connected subgraphs are obtained.

[0029] Further, in step S5, the method for dividing the hierarchical entity is: taking an important knowledge data entity as the core, dividing the entity set with hierarchical structure according to the relationship distance of the node and the important knowledge data entity, if there is no knowledge data entity with a score greater than or equal to 0.5 in a connected subgraph, then the knowledge data entity with the highest score in the connected subgraph is set as the important knowledge data entity; taking the important knowledge data entity as the core level and setting it as level 0, traversing the 1-degree relationship nodes of the important knowledge data entity and setting them as level 1, traversing the 1-degree relationship nodes of the level 1 nodes and setting them as level 2, until there are no new relationship nodes to traverse, and the hierarchical entity containing different levels is obtained.

[0030] Further, after obtaining the hierarchical entity containing different levels, the breadcrumb navigation is used to keep the knowledge path coordination in the drilling process.

[0031] The beneficial effects of the present application are:

[0032] The present application aims at the weak explainability of the existing knowledge graph, converts abstract knowledge features into intuitive visual elements through multi-dimensional visualization rules, records specific data through node feature form, and ensures logical compliance through node text matching degree and knowledge correlation index verification;

[0033] The noise sensitivity problem is solved, invalid associations are filtered through matching degree and correlation index threshold, isolated nodes are excluded, entity connection errors, relationship ambiguity and other noises are effectively reduced, and the accuracy and reliability of the graph are ensured; in view of the high query delay pain point, the graph is divided into connected subgraphs for block management, the LR model is used to calculate the entity score to mark the important knowledge data entity, and the important knowledge data entity and the associated nodes are preferentially loaded during query, which greatly reduces the query delay and improves the use efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 This is a flowchart of the knowledge management and representation method according to an embodiment of the present invention;

[0035] Figure 2 This is a flowchart of a connected subgraph according to an embodiment of the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] like Figure 1 and Figure 2 As shown in the figure, a knowledge management and representation method according to an embodiment of the present invention includes the following steps:

[0038] S1. Obtain knowledge data entities, construct a pattern layer of the knowledge graph based on the knowledge data entities, and calculate the matching degree between the knowledge data entity node text and the pattern layer node text. When the matching degree between the knowledge data entity node text and the pattern layer node text is greater than a set matching threshold, it is determined that the pattern layer is compatible with the knowledge data entity.

[0039] Knowledge data is acquired by extracting knowledge data entities through entity recognition algorithms and extracting the relationships between knowledge data entities through relation extraction algorithms; the entity recognition algorithm is BERT-CRF algorithm and the relation extraction algorithm is PCNN algorithm.

[0040] The BERT-CRF algorithm uses a pre-trained BERT model to semantically encode knowledge data text and combines it with a CRF layer to annotate entity boundaries; the PCNN relation extraction algorithm uses a segmented convolutional neural network to extract local features from text fragments containing entity pairs.

[0041] The node text matching score is calculated as follows: First, the node text in the knowledge data and the node text in the pattern layer are segmented to obtain the string matching score. Then, the entity semantic information in the two types of node text is identified, and the semantic similarity is obtained based on the cosine similarity of the vector space. Next, the attribute values ​​corresponding to the target node text attributes are extracted and attribute verification is performed to obtain the attribute verification value. Finally, the string matching score, semantic similarity, and attribute verification value are fused using a preset formula to obtain the node text matching score. The calculation formula is as follows:

[0042]

[0043] In the formula, The number of the preset time period. For the first The text matching degree of nodes within a preset time period. It is a natural constant. For the first The string matching degree for a preset time period. For string reference matching degree, For the first Semantic similarity over a preset time period For semantic reference similarity, For the first Attribute validation values ​​for a preset time period. This is a reference validation value for the attribute. The attribute validation correction factor is set to a matching threshold of 0.8.

[0044] S2. Perform position mapping on the pattern layer to generate a graph structure framework, calculate the knowledge relevance index of the pattern layer, and determine that the graph structure framework is qualified for association when the knowledge relevance index is greater than the association threshold; at the same time, obtain the core features of the pattern layer, including usage frequency and scene importance.

[0045] The core features, including frequency of use and importance of scenarios, were obtained by extracting them from the operation logs of the knowledge management platform. Specifically:

[0046] Usage frequency includes the call records of knowledge data entities, including the number of times the entity is queried, viewed, and referenced by users; and the number of times the entity is automatically called when embedded in business processes.

[0047] The importance of a scenario includes the number of users and the number of times it is shared.

[0048] The knowledge relevance index is calculated as follows: First, concept coverage, entity matching degree, and relationship consistency are calculated; then, the knowledge relevance index is calculated using a preset formula, which is:

[0049]

[0050] In the formula, The number of the preset time period. For the first A knowledge relevance index for a preset time period. It is a natural constant. For the first Entity matching degree over a preset time period For entity reference matching degree, For the first Consistency of relationships across preset time periods For the sake of relational reference consistency, For the first The concept coverage of the preset time period is 0.8.

[0051] S3, generating a knowledge graph containing knowledge point nodes, association relationship edges and visual elements according to the atlas structure framework, the mode layer knowledge association index and the core features; determining a corresponding relationship between the multi-dimensional visualization rules and the visual element knowledge graph according to preset multi-dimensional visualization rules.

[0052] The corresponding relationship between the multi-dimensional visualization rules and the visual element knowledge graph is that the core features of the mode layer are converted into visual elements of the knowledge graph through preset quantitative multi-dimensional visualization rules. The specific corresponding relationship calculation method of the multi-dimensional visualization rules is as follows:

[0053] Node size mapping calculation: preset 5 fixed sizes ), obtaining the maximum usage frequency of all knowledge data entities, matching according to the usage frequency interval, and the frequency interval is calculated as follows:

[0054] Interval

[0055] Wherein, is the usage frequency (number of calls in unit time) of a single knowledge data entity, is the maximum usage frequency of all entities, · is the ceiling function; if , all nodes uniformly adopt to avoid too small size to be identified.

[0056] Node color mapping calculation: preset 5 gradient colors ), matching according to the scene importance interval, and the importance interval is calculated as follows:

[0057] Interval

[0058] Wherein, is the scene importance index of a single knowledge data entity, and the scene importance index includes the number of users and the number of sharing times, is the maximum importance index of all entities; if , all nodes uniformly adopt to ensure that the importance can be distinguished.

[0059] Node line number mapping: preset 3 line numbers , matching according to the scene importance index, the scene importance index , the scene importance index , and the scene importance index .

[0060] Relationship line type mapping: The relationship line type mapping presets four line types: solid line, dashed line, dotted line, and double solid line. The relationship line types include four types of relationships: process to process, process to data, process to message, and process to AI.

[0061] When the data types of the two related knowledge data entities are both processes, the process-to-process relationship is triggered, and the line mapping of the relationship is set to a solid line.

[0062] When the data type of one knowledge data entity is process and the data type of another knowledge data entity is data, the linear mapping of the relationship is set to a dashed line;

[0063] When the data type of one knowledge data entity is process and the data type of another knowledge data entity is message, the linear mapping of the relationship is set to dotted line;

[0064] When one knowledge data entity has a data type of process and another knowledge data entity has a data type of AI model, the linear mapping of the relationship is set to double solid lines.

[0065] Feature form configuration: Configure a small form for each knowledge point node to record specific values ​​or field descriptions of usage frequency and scenario importance;

[0066] In the acquired knowledge graph, the visual elements correspond to the core features of the knowledge points.

[0067] S4. Exclude isolated nodes in the knowledge graph of visualized elements in S3, randomly select a non-isolated node, and search for its connected nodes by traversing it, marking it as a connected subgraph; repeat the traversal process until all non-isolated nodes are marked, resulting in multiple connected subgraphs.

[0068] S5. Calculate the spatial attributes of the knowledge data entity graph in each connected subgraph. The spatial attributes of the knowledge data entity graph include the number of nodes associated with each node, the average length of the shortest path from a node to other nodes, and the number of times a node acts as a bridge for the shortest path from other nodes. Input the calculated number of nodes associated with each node, the average length of the shortest path from a node to other nodes, the number of times a node acts as a bridge for the shortest path from other nodes, the usage frequency obtained in step S2, and the scene importance into the LR model to calculate the knowledge data entity score. Compare the calculated knowledge data entity score with the set knowledge data entity threshold, mark important knowledge data entities, and divide them into hierarchical entities based on important knowledge data entities.

[0069] The method for quantifying the spatial attributes of a knowledge data entity graph is as follows:

[0070] Number of edges associated with a node (denoted as ) The calculation logic involves traversing all associated edges within the connected subgraph and counting the edges related to the first edge. The number of edges directly connected to a node, calculated as:

[0071]

[0072] where, is the node set of a single connected subgraph; is an indicator function: if node is directly connected to node , ; otherwise .

[0073] The average length of the shortest path from node i to other nodes (denoted as ), calculated by using Dijkstra algorithm to calculate the shortest path length from node i to each other node j and then taking the arithmetic average, calculated as:

[0074]

[0075] where, is the total number of nodes in the connected subgraph ; is the shortest path length from node i to node j; if i and j have no connected path, .

[0076] The number of times a node serves as a bridge in the shortest path of other nodes (denoted as ), the calculation logic is to traverse all node pairs in the subgraph and count the number of times the shortest path contains node ; the calculation formula is:

[0077]

[0078] where, is an indicator function: if the shortest path from node to node passes through node , ; otherwise ; is to avoid repeated statistics of node pairs and .

[0079] In addition, because , , , the frequency of use , and the importance of the scene have large dimensional differences, they need to be standardized by standard score to eliminate the dimensional influence. The feature standardization formula (denoted as after standardization) is:

[0080]

[0081] In the formula, Indexing for feature dimensions ( correspond , correspond , correspond , correspond , correspond ); For the first The node of the first Original eigenvalues; For the first in the historical knowledge graph The mean of the dimensional features; For the first in the historical knowledge graph Standard deviation of dimensional features.

[0082] The linear prediction function of the LR model (denoted as ) The 5-dimensional standardized features are weighted and summed with the model parameters to obtain the linear prediction value. The calculation formula is as follows:

[0083]

[0084] In the formula, This represents the bias term in the LR model; For the first Weight parameters for dimensional features; For the first The node of the first Dimensional standardization features.

[0085] linear prediction value Mapping to the [0, 1] interval yields the knowledge data entity score:

[0086]

[0087] In the formula, For the first The knowledge data entity score of each node (value range [0, 1]); It is a natural constant; The above is the linear prediction value.

[0088] Specifically, step S4, obtaining the connected subgraph, includes the following steps:

[0089] S41: Traverse the knowledge graph of visualized elements in S3, exclude isolated nodes with no relation to each other, and mark the remaining non-isolated nodes to form a set of non-isolated nodes;

[0090] S42: randomly select an initial node from the non-isolated node set, mark the initial node and its connected nodes as a connected subgraph by traversing all nodes directly or indirectly connected to the initial node;

[0091] S43: repeat the operation of S42 until all nodes in the non-isolated node set are marked, and obtain multiple connected subgraphs.

[0092] Specifically, in step S5, the method for dividing the hierarchical entity is: taking an important knowledge data entity as the core, dividing the entity set with hierarchical structure according to the relationship distance of nodes and the important knowledge data entity, if there is no knowledge data entity with a score ≥0.5 in a connected subgraph, then set the knowledge data entity with the highest score in the connected subgraph as the important knowledge data entity; take the important knowledge data entity as the core level and set it as level 0, traverse the 1-degree relationship nodes of the important knowledge data entity and set them as level 1, traverse the 1-degree relationship nodes of the level 1 nodes and set them as level 2, until there are no new relationship nodes to traverse, and obtain the hierarchical entity containing different levels.

[0093] To ensure smooth drilling, the user clicks on a certain level node to trigger drilling, and the system preferentially loads the connected subgraph where the node is located; during the drilling process, the knowledge path is kept in cooperation through breadcrumb navigation, the characteristics are kept in cooperation through node visualization elements, and the association relationship is kept in cooperation through relationship lines.

[0094] The application obtains the knowledge data entity with a knowledge data entity score greater than 0.5 through step S5, marks it as an important knowledge data entity, and applies it to the cooperation of progressive loading and dynamic display for smooth drilling.

[0095] When initially loaded, the current node and 1-degree associated nodes are displayed, the height between levels is set as h, the nodes within the level are automatically displayed in multiple lines, and the height is set as h / 3; when secondarily loaded, the user clicks on the 1-degree node, the 2-degree associated nodes of the node and the corresponding relationship are loaded; if the number of connected subgraph nodes is >5000, the important knowledge data entity and the 1-2 degree nodes are preferentially displayed, and the remaining nodes are loaded through button control to avoid lag. The important knowledge data entity node is preferentially loaded, and the non-important knowledge data entity node is delayed.

[0096] The trigger condition for real-time updating of the knowledge graph is business logic change. If the business change is greater than or equal to 10%, real-time updating is triggered. Business changes include changes in the number of nodes and fluctuations in core feature values. After triggering, the changed knowledge point nodes or associated relationship edges are updated in real time, and the knowledge correlation index is recalculated to ensure the accuracy of the updated graph correlation. The intermediate processing of the timing update is the full update of the knowledge graph at the preset time period every day, including the import of new knowledge data entities and associated relationship edges, the deletion of invalid knowledge data entities and associated relationship edges, the re-statistics and annotation of all knowledge data entity core features.

[0097] The visualization element rules remain consistent during the drilling process. When the knowledge graph is dynamically updated, the core feature changes need to be updated synchronously, and the knowledge correlation index needs to be verified again after the update to ensure the adaptation of the corresponding relationship and data.

[0098] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the skilled in the art can combine and combine the different embodiments or features of the embodiments or examples described in the present specification without contradiction.

[0099] Any process or method descriptions in flow charts or described elsewhere herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for performing specific logic functions or steps in the process. The scope of preferred embodiments of the present application includes additional implementation in which the functions described in the illustrated or discussed order are performed in a different order, including substantially simultaneously or in reverse order, as will be understood by those skilled in the art of the embodiments to which the present application belongs.

[0100] The logic and / or steps represented in flow diagrams or otherwise described herein, for example, can be considered as a sequence of instructions to implement logic functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, processor- containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. In the context of this specification, a "computer-readable medium" can be any means that can contain, store, communicate, propagate or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a machine-readable storage device (e.g., magnetic, optical or other) a machine-readable storage diskette (e.g., floppy disk, optical disk, etc.), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), optical fibers, and a portable compact disc read-only memory (CDROM). Further, the computer-readable medium can even be paper or other suitable medium upon which the program is printed, as the program can be electronically captured, for example via the optical scanner of a device or other electronic capture device, and then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and stored in a computer memory.

[0101] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, various steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies, known in the art, or combinations thereof, can be used: a discrete logic circuit having logic gates for implementing logic functions upon data signals, an application specific integrated circuit having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), or the like.

[0102] Those of skill in the art could readily implement the above described example methods with all or a portion of the disclosed steps carried out by a program for use with a computer system or similar electronic device, where the program is intended for use as intermediate steps to achieve the results indicated herein. Note that the software is not intended to be limited to the state of this technology, as it typically is defined to include any type of computer instruction or set of instructions, that can be employed to cause a processor to perform a computer implemented process.

[0103] In addition, each function unit in each embodiment of the present application can be integrated in one processing module, or each unit can exist physically separately, or two or more units can be integrated in one module. The integrated module can be realized in the form of hardware, or in the form of software function module. When the integrated module is realized in the form of software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0104] Although the embodiments of the present application have been shown and described above, it should be understood by those skilled in the art that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A method of knowledge management and representation, characterized by, The method comprises the following steps: S1, acquiring knowledge data entities, constructing a mode layer of a knowledge graph based on the knowledge data entities, and calculating the matching degree of knowledge data entity node text and mode layer node text, wherein when the matching degree of the knowledge data entity node text and the mode layer node text is greater than a set matching threshold, it is determined that the mode layer is adapted to the knowledge data entities; S2, performing position mapping on the mode layer to generate a graph structure framework, calculating a mode layer knowledge correlation index, wherein when the knowledge correlation index is greater than a correlation threshold, it is determined that the graph structure framework is correlated; meanwhile, the core features of the mode layer are acquired, and the core features include usage frequency and scene importance; S3, generating a knowledge graph containing knowledge point nodes, correlation relationship edges and visual elements according to the graph structure framework, the mode layer knowledge correlation index and the core features; presetting multi-dimensional visualization rules, and determining the correspondence between the multi-dimensional visualization rules and the visual element knowledge graph; S4, excluding isolated nodes in the visual element knowledge graph in S3, randomly selecting a non-isolated node, marking its connected nodes as a connected subgraph through traversal search, and repeating the traversal process until all non-isolated nodes are marked to obtain multiple connected subgraphs; S5, calculating the spatial attributes of the knowledge data entity graph in each connected subgraph, wherein the spatial attributes of the knowledge data entity graph include the number of node correlation edges, the average length of the shortest path from a node to other nodes, and the number of times a node serves as a bridge of the shortest path of other nodes, inputting the number of node correlation edges, the average length of the shortest path from a node to other nodes, and the number of times a node serves as a bridge of the shortest path of other nodes, the usage frequency and the scene importance obtained in step S2 into an LR model to calculate the score of the knowledge data entity; comparing the calculated knowledge data entity score with a set knowledge data entity threshold to mark important knowledge data entities, and dividing the hierarchical entities according to the important knowledge data entities.

2. The knowledge management and representation method of claim 1, wherein: In step S1, the knowledge data entities in the knowledge data are extracted by an entity recognition algorithm, and the correlation between the knowledge data entities is extracted by a relation extraction algorithm; the entity recognition algorithm is a BERT-CRF algorithm, and the relation extraction algorithm is a PCNN algorithm.

3. The knowledge management and representation method of claim 1, wherein, In step S1, the calculation formula of the node text matching degree is: In the formula, is the number of the preset time period, is the node text matching degree of the preset time period, is a natural constant, is the string matching degree of the preset time period, is the string reference matching degree, is the semantic similarity of the preset time period, is the semantic reference similarity, is the attribute verification value of the preset time period, is the attribute reference verification value, is the attribute verification correction factor.

4. The knowledge management and representation method of claim 1, wherein, In step S2, the calculation formula of the knowledge correlation index is: In the formula, is the number of the preset time period, is the knowledge relevance index of the preset time period, is a natural constant, is the entity matching degree of the preset time period, is the entity reference matching degree, is the relationship consistency of the preset time period, is the relationship reference consistency, is the concept coverage of the preset time period.

5. The knowledge management and representation method of claim 1, wherein, In step S2, the multi-dimensional visualization rules are: Node size mapping: presetting 5 fixed node sizes, acquiring the maximum usage frequency of all knowledge data entities, dividing the frequency interval by dividing the maximum usage frequency by 5, and matching the corresponding size of the knowledge point node falling into different intervals according to the usage frequency; if the maximum usage frequency is less than 10, all knowledge point nodes adopt the size corresponding to the maximum interval; Node color mapping: presetting 5 gradient colors, acquiring the maximum scene importance index of all knowledge data entities, dividing the importance interval by dividing the maximum index by 5, and matching the corresponding color of the knowledge point node falling into different intervals according to the scene importance; If the maximum index is less than 10, all knowledge point nodes adopt the color corresponding to the maximum interval; Node line number mapping: three kinds of node description lines, namely, preset thin line, medium thick line and thick line, correspond to the low, medium and important levels of the scene importance of knowledge data entities respectively, and the corresponding line is matched after the level is determined; Association relationship line type mapping: four kinds of association relationship line types are preset, and the corresponding line type is matched after the type is determined according to the business process dependence degree; Feature form configuration: a small form is configured for each knowledge point node to record the specific value or field description of the usage frequency and scene importance; The obtained knowledge graph corresponds the visual elements to the core features of knowledge points.

6. The knowledge management and representation method of claim 1, wherein, In step S4, the obtaining of the connected subgraph specifically includes the following steps: S41: traversing the knowledge graph of the visual elements in S3, excluding isolated nodes without any association relationship, and marking the remaining non-isolated nodes to form a non-isolated node set; S42: randomly selecting an initial node from the non-isolated node set, searching all nodes directly or indirectly connected to the initial node by traversal, and marking the initial node and its connected nodes as a connected subgraph; S43: repeating the operation of S42 until all nodes in the non-isolated node set are marked, and a plurality of connected subgraphs are obtained.

7. The knowledge management and representation method of claim 1, wherein, In step S5, the method for dividing the hierarchical entities is: taking important knowledge data entities as the core, dividing the entity set with hierarchical structure according to the relationship distance between nodes and important knowledge data entities, if there is no knowledge data entity with a score greater than or equal to 0.5 in a connected subgraph, the knowledge data entity with the highest score in the connected subgraph is set as the important knowledge data entity; Taking the important knowledge data entity as the core level and setting it as level 0, traversing the 1-degree relationship nodes of the important knowledge data entity and setting them as level 1, traversing the 1-degree relationship nodes of the level 1 nodes and setting them as level 2, until there is no new relationship node to traverse, and the hierarchical entities containing different levels are obtained.

8. The knowledge management and representation method of claim 7, wherein: After obtaining the hierarchical entities containing different levels, the knowledge path is kept coordinated through breadcrumb navigation in the drilling process.

Images