Method, apparatus, device, and medium for generating knowledge graph of production unit
Patent Information
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- SIEMENS AG
- Filing Date
- 2023-09-25
- Publication Date
- 2026-07-01
AI Technical Summary
Existing methods for generating knowledge graphs of production units are hindered by knowledge barriers, including the lack of experienced industry experts, ineffective methods for collecting and verifying expert knowledge, and the limitations of top-down approaches that are difficult to promote and reuse.
A method and apparatus for generating knowledge graphs of production units that acquire knowledge from sources, convert it into display content for a visual interface, and generate a knowledge topological graph and knowledge graph based on user-triggered instructions, allowing for bottom-up generation without relying on expert guidance or industry standards.
This approach enables efficient and user-friendly generation of knowledge graphs, improving the efficiency of knowledge graph creation and allowing for flexible, collaborative assembly of multi-source knowledge.
Smart Images

Figure CN2023121217_03042025_PF_FP_ABST
Abstract
Description
Method, apparatus, device, and medium for generating knowledge graph of production unitFIELD
[0001] The present invention relates to the technical field of knowledge graph, in particular to a method, apparatus, device, and medium for generating knowledge graph of production unit.BACKGROUND
[0002] Industry knowledge is the core of industrial application, and many digital applications need to follow the domain scenarios, so we often explore and develop industrial digital services under the guidance of industry experts. But the enterprises also face many challenges in domain application / service and even advanced collaborative analysis, these challenges include but are not limited to staff turnover and funding restrictions, and the knowledge barriers are obstacles too.
[0003] Typical examples of knowledge barriers include: although workers can obtain equipment vibration and electrical data, they often do not understand the meaning of these data. On the other hand, on-site equipment experts understand the mechanical structure of the equipment, but may not be familiar with the data characteristics of the equipment.SUMMARY
[0004] Embodiments of the present invention propose a method, apparatus, device, and medium for generating knowledge graph of production unit.
[0005] In a first aspect, a method for generating knowledge graph of production unit is provided. The method comprising:
[0006] acquiring knowledge from a knowledge source of a production unit;
[0007] converting the knowledge into a display content suitable for display on a visual interface;
[0008] generating a first knowledge topological graph suitable for display on the visual interface, in response to a user-triggered instruction issued on the display content; and
[0009] generating a first knowledge graph based on the first knowledge topological graph.
[0010] In a second aspect, an apparatus for generating knowledge graph of production unit is provided. The apparatus comprising:
[0011] an acquiring module, configured to acquire knowledge from a knowledge source of a production unit;
[0012] a converting module, configured to convert the knowledge into a display content suitable for display on a visual interface;
[0013] a first generating module, configured to generate a first knowledge topological graph suitable for display on the visual interface, in response to a user-triggered instruction issued on the display content; and
[0014] a second generating module, configured to generate a first knowledge graph based on the first knowledge topological graph.
[0015] In a third aspect, an electronic device is provided. The electronic device comprising a processor and a memory, wherein an application program executable by the processor is stored in the memory for causing the processor to execute a method for generating knowledge graph of production unit as described in any of the above.
[0016] In a fourth aspect, a computer-readable medium comprising computer-readable instructions stored thereon is provided, wherein the computer-readable instructions, when executed by a processor, implement a method for generating knowledge graph of production unit as described in any of the above.
[0017] In a fifth aspect, a computer program product comprising a computer program, when the computer program is executed by a processor for executing a method for generating knowledge graph of production unit as described in any of the above.
[0018] According to the above technical solutions, generating knowledge graphs in a user-friendly way in a visual interface based on knowledge obtained from knowledge sources is conducive to bottom-up generation of knowledge graphs, without relying on expert guidance or industry standards, improving the efficiency of generating knowledge graphs.BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to make technical solutions of examples of the present disclosure clearer, accompanying drawings to be used in description of the examples will be simply introduced hereinafter. Obviously, the accompanying drawings to be described hereinafter are only some examples of the present disclosure. Those skilled in the art may obtain other drawings according to these accompanying drawings without creative labor.
[0020] Fig. 1 is a flowchart of a method for generating knowledge graph of production unit according to an embodiment of the present invention.
[0021] Fig. 2 is an exemplary schematic diagram of visual interface according to an embodiment of the present invention.
[0022] Fig. 3 is an exemplary schematic diagram displaying editing records in visual interface according to an embodiment of the present invention.
[0023] Fig. 4 is a first exemplary schematic diagram of combining knowledge topological graphs according to an embodiment of the present invention.
[0024] Fig. 5 is a second exemplary schematic diagram of combining knowledge topological graphs according to an embodiment of the present invention.
[0025] Fig. 6 is a schematic diagram of index structure for identifying quality of knowledge graph according to an embodiment of the present invention.
[0026] Fig. 7 is a schematic diagram of the process of generating knowledge graph of production unit according to an embodiment of the present invention.
[0027] Fig. 8 is a structural diagram of an apparatus for generating knowledge graph of production unit according to an embodiment of the present invention.
[0028] Fig. 9 is a structural diagram of an electronic device according to an embodiment of the present invention.
[0029] List of reference numbers: DETAILED DESCRIPTION
[0030] In order to make the purpose, technical scheme, and advantages of the invention clearer, the following examples are given to further explain the invention in detail.
[0031] In order to be concise and intuitive in description, the scheme of the invention is described below by describing several representative embodiments. Many details in the embodiments are only used to help understand the scheme of the invention. However, it is obvious that the technical scheme of the invention can be realized without being limited to these details. In order to avoid unnecessarily blurring the scheme of the invention, some embodiments are not described in detail, but only the framework is given. Hereinafter, "including" refers to "including but not limited to" , "according to. . . " refers to " at least according to. . ., but not limited to. . . " . When the number of an element is not specifically indicated below, it means that the element can be one or more, or can be understood as at least one.
[0032] How to understand these knowledge barriers? The most typical example is that although we can get the vibration and electrical data of the equipment, we do not understand the meaning of these data. Which data are normal, and which are abnormal. On the other hand, field equipment experts understand the mechanical structure of the equipment but are not familiar with the data characteristics of the equipment. Widespread and expanded to the production scene, involving many productions equipment, production scheduling requirements and energy, all of them are closely related to knowledge. Since knowledge is so important, why is there still a lack of knowledge integration? After research, it has been found that there are many reasons leading to knowledge barriers, including but not limited to:
[0033] Q1: Lack of experienced industry experts and effective methods to collect and verify expert’s knowledge. The loss of industry talents (experts) is an important reason. From the technical point of view, what is more important is the lack of effective technical methods. For a long time, people have placed their hopes on AI to acquire and organize knowledge systems through semantic recognition and extraction. However, the maintenance of thesaurus and the development and optimization of extraction algorithms are a long-term investment process, so there is a lack of knowledge applications.
[0034] Q2: The existing top-down approach to building knowledge graph is not easy to promote and reuse. Generally, we build knowledge models or ontologies based on expert guidance or industry specifications, following top-down programming methods, but such top-down knowledge database-based methods are not suitable for the change of scenario, such as
[0035] Q3: How to verify the knowledge graph? It is also a business question that needs to be answered from a technical perspective. If we cannot find a good technical solution to verify the effectiveness and correctness of knowledge, it is difficult to promote it from a business perspective.
[0036] Embodiments of the present provide a functional model-based solution to collect enterprise domain know-how and try to make knowledge as a new business expansion. The key innovation points include: 1: Provide functional model-based engineering building-blocks to promote knowledge accumulation. 2: Strategy mode of knowledge verification and adjustment. And these functional model-based engineering building-blocks can be easily integrated into project services and product services, to help us acquire domain knowledge and extract relevant and valuable industry models.
[0037] The existing knowledge graph digital service has two directions.
[0038] One direction is about KG building, most of this direction is based on Graph DB client tool to build the knowledge entity and relationship and some experts using RDF methods to create ontology. This direction, in the early stage, this approach needs to cross the gap between IT development and industry knowledge, and in the later stage, it requires continuous database maintenance, otherwise it is difficult to reuse and expand.
[0039] The other direction is key info extraction, the extracted info including entity and relationship, most of the methods depend on the semantic thesaurus, and was supported by rule-based, probabilistic models, factorization methods and embedding models, these methods and models created the kg need long-term investment and hard to validated.
[0040] Embodiments of the present invention attempts to better assist people in different fields in collecting and solidifying industry knowledge from the perspective of visualization tools based on construction tools.
[0041] Fig. 1 is a flowchart of a method for generating a knowledge graph of a production unit according to an embodiment of the present invention. As shown in Figure 1, the method comprises:
[0042] Step 101: acquiring knowledge from a knowledge source of a production unit.
[0043] Here, production units can include manufacturing enterprises (such as automotive manufacturing, electronic information product manufacturing, mechanical manufacturing, metal product manufacturing, food and beverage manufacturing, building material manufacturing, etc. ) and non-manufacturing enterprises (such as service, retail, financial, and educational enterprises) . Among them, the production unit can be implemented as a group level company, company, manufacturer, production line, production workshop or production station, etc.
[0044] The knowledge source can be an open-source database or a private database. For example, knowledge can be implemented as device manuals (PDF, word, or image) , maintenance records (CSV, XLS, ppt / pptx) , OT operation control logic (configuration files, DB, and API) , system knowledge output (API files or test files) , and so on.
[0045] Step 102: converting the knowledge into a display content suitable for display on a visual interface.
[0046] In one embodiment, converting the knowledge into a display content suitable for display on a visual interface comprises: determining a type attribute of the knowledge; converting the knowledge into a display content suitable for display on a visual interface based on a conversion tool corresponding to the type attribute. Therefore, by converting knowledge into display content, it is easy for users to operate and generate a knowledge graph.
[0047] For example, when knowledge is a file, it can be converted into content suitable for display on web pages through conversion tools such as S2X, E2X / E2J. When knowledge provides content for APIs, it can be transformed into content suitable for display on web pages through methods such as JSON / J2XML. When knowledge provides content for databases, it can be transformed into content suitable for display on web pages through methods such as Schema JSON / J2XML, DB Access, etc. When knowledge is an image, it can be transformed into content suitable for display on web pages through methods such as Img2Cnavas.
[0048] Step 103: generating a first knowledge topological graph suitable for display on the visual interface, in response to a user-triggered instruction issued on the display content.
[0049] Here, based on the instruction issued by user browsing the displayed content, a first knowledge topological graph adapted to display on the visual interface is generated.
[0050] For example, users can select a vocabulary from the displayed content to serve as entity 1, and correspondingly generate node 1 of entity 1 in the first knowledge topological graph. Users can choose another vocabulary from the displayed content to serve as entity 2, and accordingly generate node 2 of entity 2 in the first knowledge topological graph. Moreover, users can establish a connection between node 1 and node 2 by dragging and dropping node 1 and node 2, and edit the relationship between node 1 and node 2.
[0051] Step 104: generating a first knowledge graph based on the first knowledge topological graph.
[0052] Therefore, based on knowledge obtained from knowledge sources, generating knowledge graphs in a user-friendly way in a visual interface is conducive to bottom-up generation of knowledge graphs, without relying on expert guidance or industry standards, improving the efficiency of generating knowledge graphs.
[0053] In one embodiment, the user-triggered instruction comprises at least one of the following:
[0054] (1) selection instruction for selecting part or all of the display content;
[0055] (2) node generation instruction for generating a node in the first knowledge topological graph based on a selected content;
[0056] (3) node connection instruction for connecting nodes in the first knowledge topological graph;
[0057] (4) relationship editing instruction for editing relationships between nodes in the first knowledge topological graph;
[0058] (5) node deletion instruction for deleting a node in the first knowledge topological graph.
[0059] Therefore, the function of generating knowledge graphs has been enriched by proposing multiple types of user triggered instructions.
[0060] The above exemplary description provides typical examples of user-triggered instructions, and those skilled in the art may realize that this description is only exemplary and is not intended to limit the scope of protection of the present invention's embodiments.
[0061] In one embodiment, the method comprising: displaying a list containing multiple preset relationships in the visual interface; wherein the relationship editing instruction is used to select a relationship from the list. Therefore, the difficulty of generating a knowledge graph can be reduced by selecting preset relationships,
[0062] In one embodiment, the method comprising: displaying an input box in the visual interface; sending a prompt message to input a relationship between nodes in the input box; receiving a string input in the input box; wherein the relationship editing instruction is used to determine the string as a relationship between nodes. Therefore, the accuracy of the knowledge graph can be improved by inputting relationships represented by strings.
[0063] In one embodiment, wherein generating a first knowledge graph based on the first knowledge topological graph comprises: converting the first knowledge topological graph into a script supported by a database for storing a knowledge graph; writing the script into the database to generate the first knowledge graph. Therefore, a convenient and fast knowledge graph generation method is implemented based on script operations.
[0064] In one embodiment, comprising: acquiring a second knowledge topological graph; displaying the second knowledge topological graph in the visual interface; combing the first knowledge topological graph and the second knowledge topological graph in the visual interface; generating a second knowledge graph based on a combined knowledge topological graph. Therefore, different knowledge topological graphs can also be combined to generate a combined knowledge topological graph. Combining knowledge graphs can enrich the content of knowledge graphs. The combination of knowledge topological graphs promotes the collaborative assembly of multi-source knowledge and increases the flexibility of knowledge system.
[0065] In one embodiment, wherein comprises at least one of the following:
[0066] (1) knowledge sources of the second knowledge topological graph and the first knowledge topological graph belong to different departments in the production unit;
[0067] (2) knowledge source of the second knowledge topological graph belongs to another production unit that is different from the production unit;
[0068] (3) knowledge sources of the second knowledge topological graph and the first knowledge topological graph belong to the same department in the production unit at different times;
[0069] (4) the user who participates in generating the second knowledge topological graph is different from the user who issues the user-triggered instruction.
[0070] Therefore, multiple types of collaborative assembly can be achieved, which is suitable for various application scenarios.
[0071] In one embodiment, wherein combing the first knowledge topological graph and the second knowledge topological graph comprises: determining the same nodes in the first knowledge topological graph and the second knowledge topological graph; merging the same nodes in the combined knowledge topological graph. Therefore, the efficiency of collaborative assembly can be improved by merging common nodes.
[0072] In one embodiment, the method comprising: determining a maturity index of the second knowledge graph based on the number of nodes added compared to the first knowledge graph in the second knowledge graph; determining a combination index of the second knowledge graph based on the number of times the same nodes are merged; determining a citation index of the second knowledge graph based on citation number of the second knowledge graph; evaluating quality of the second knowledge graph based on the maturity index, combination index, and citation index. Therefore, a measurement index system for the quality of knowledge graphs is proposed, which can measure the quality of knowledge graphs from multiple dimensions.
[0073] Fig. 2 is an exemplary schematic diagram of visual interface according to an embodiment of the present invention. The visual interface includes command bar 11, first display area 12, and second display area 13. Command bar 11 contains multiple controls, such as control 31 for importing knowledge (such as maintenance records, equipment manuals, etc. ) and control 32 for saving knowledge topological graph. After knowledge is imported, display the content of the knowledge in the first display area 12. For example, the displayed content includes: " Problem description: spreader reports "51" motor overcurrent fault at the lift at the entrance of BA5. Check on site that the walking slave wheel bearing is broken. (replace the new rubber coating wheel assembly, and the last maintenance time is 20xx. 12.10. ) " and "Disposal way: Manually reattach the body; Replace the slave wheel after pulling the spreader out of the lifter " .
[0074] Users can select content (such as words) in the first display area 12. Correspondingly, a node corresponding to the selected content is generated in the second display area 13. For example, if the user selects "BA5" and "Motor overcurrent fault" , corresponding "BA5" node 36 and "Motor overcurrent fault" node 37 will be generated in the second display area 13. Then, the user can perform a drag and drop operation to connect the line between "BA5" node 36 and "Motor overcurrent fault" node 37, and edit the relationship between "BA5" node 36 and "Motor overcurrent fault" node 37. Similarly, generating node 35 for "driven wheel bearing" and node 38 for "Manually reattach the body; Replace the slave wheel after pulling the spreader out of the lifter " , the relationship between nodes 35 and 38, as well as the relationship between nodes 36 for "BA5" and 35 for "driven wheel bearing" . Convert the knowledge topological graph containing all nodes and relationships in the first display area 13 into a script, and write the script into a database to generate the knowledge graph.
[0075] Fig. 3 is an exemplary schematic diagram displaying editing records in a visual interface according to an embodiment of the present invention.
[0076] In Figure 3, the visual interface also includes an editing record area 14. The editing record area 14 contains editing records sorted in chronological order for the knowledge topological graph. For example, the editing record area 14 contains three editing records, namely editing records 141-143. When selecting an editing record in the editing record area 14, further display the edited knowledge topological graph in the editing details display area, as well as auxiliary information such as editing content and editing time.
[0077] Fig. 4 is a first exemplary schematic diagram of combining knowledge topological graphs according to an embodiment of the present invention. Fig. 5 is a second exemplary schematic diagram of combining knowledge topological graphs according to an embodiment of the present invention.
[0078] In Figure 4, the user can drag and drop the second knowledge topological graph 40 to approach the first knowledge topological graph 20 and generate combined knowledge topological graph 50. Same nodes (node 24) in the first knowledge topological graph 20 and the second knowledge topological graph 40 are merged into one.
[0079] The node ID in the second knowledge topological graph 40 and the node ID in the first knowledge topological graph 20 may be the same. Therefore, in the combined knowledge topological graph 50, it is necessary to solve the conflict problem between node IDs. In one embodiment, the IDs in the first knowledge topological graph 20 are kept unchanged, and the maximum value of the IDs in the first knowledge topological graph 20 is determined. Then, add the maximum value to all newly added node IDs that originally belonged to the second knowledge topological graph 40 as the new ID. For example, assume that the node with the highest ID value in the first knowledge topological graph 20 is node 26, and its ID is 150. So, in combined knowledge topological graph 50, add 150 to respective ID values of the newly added nodes 41-45.
[0080] Based on combined knowledge topological graph 50, a second knowledge graph (referred to as the combinatorial knowledge graph) can be generated. The implementation method of the present invention also proposes a quality index system for combining knowledge graphs. The specific ways to determine the second knowledge graph include:
[0081] (1) Based on the added nodes of the second knowledge graph compared to the first knowledge graph, determine maturity index A1 of the second knowledge graph. For example, if the second knowledge graph has more new nodes compared to the first knowledge graph, it means that the second knowledge graph enriches more content, which is beneficial for the quality of the second knowledge graph.
[0082] (2) Determine combination index A2 of the second knowledge graph based on the number of times the same nodes are merged. For example, if the second knowledge graph has more merged nodes, it means that the second knowledge graph is more accurate, which is beneficial for the quality of the second knowledge graph.
[0083] (3) Based on the citation number of the second knowledge graph, determine citation index A3 of the second knowledge graph. For example, if the number of references in the second knowledge graph increases, it means that the second knowledge graph is popular, indirectly representing a higher quality of the second knowledge graph.
[0084] (4) Based on the maturity index, combination index, and citation index, evaluate the quality A of the second knowledge graph. For example, A=a1 *A1+a2 *A2+a3 *A3. Among them, a1, a2, and a3 are the preset weights, respectively.
[0085] Fig. 6 is a schematic diagram of the index structure for identifying the quality of the knowledge graph according to according to an embodiment of the present invention.
[0086] The quality of the combined knowledge graph is determined by the maturity index 61, combination index 62, and citation index 63.
[0087] Embodiments of the invention tries to from a tool perspective, based on building blocks, better help people in different fields to collect and solidify industry knowledge. The advantages including: Avoid collecting expert domain information from the perspective of developers and also avoid the domain engineers learning knowledge solidification methods or development models. Instead of demanding integrated top-down design, flexible collective creation, continuous editing, and optimization have been replaced. Validate knowledge maturity based on tracking and comparing knowledge model reuse.
[0088] Fig. 7 is a schematic diagram of the process of generating knowledge graph of production unit according to an embodiment of the present invention. Operator 70 can generate a knowledge graph based on engineering building blocks. The knowledge source adapter 71 can import multiple data sources such as file 711, API 712, image 713, and database 714. The convert pipeline 715 formats the respective knowledge from multiple data sources to form display content suitable for display on the Web, such as XML or HTML.
[0089] Operator 70 extracts entities and relationships 722 from the content displayed on visualization page (such as Webpage) of the visual interface based on the grab operation 721. Store entities and relationships 722 in cache 723. In instance manager 731 of manager 73, multiple users are allowed to log in for collaborative management. In the hierarchical structure 732 of knowledge topology, a knowledge topology structure diagram (such as XML file format) is established based on entities and relationships 722 obtained from cache 723. Display the knowledge topology structure diagram in the visual interface. In assembly 733, the knowledge topology structure diagrams generated by each user can be assembled. In Record and Track 734, save the user's editing records for knowledge topological graphs. In storage pipeline 735, save the edited knowledge topological graphs. Generate knowledge graphs 74 based on the edited knowledge topological graphs, and save the knowledge graphs in the database74.
[0090] Moreover, in quality analysis 736, knowledge graph 74 is analyzed by determining maturity index, combination index, and citation index for knowledge graph 74; Evaluate the quality of knowledge graph 74 based on maturity index, combination index, and citation index.
[0091] In summary, embodiments of the present invention try to help users cross the gap between industry knowledge and application development through the guidance of the engineering tool, while avoiding the need for programmers to build top-down knowledge structures that limit the reuse of knowledge systems. The key processing process can include:
[0092] Step1. Access knowledge source by adapter. In actual production and factory management, there is potential knowledge that needs to be extracted urgently, such as equipment manuals (. PDF, . DOC or image) , and maintenance records, OT operation control logic (config file, DB and API) , system knowledge output (API file or postman test file) , this hidden knowledge can be actively accepted through the adapter.
[0093] Step2. Extract knowledge entity and element by grab module. Semantic technology is advanced, but in the industrial field, knowledge accumulation through semantic learning alone requires relatively large human and time investment; Imagine that if our industry personnel aggregate and correlate knowledge through rapid grasping actions, it will reduce the gap between developers and the industry and improve the feasibility and popularity of knowledge accumulation.
[0094] Step3. Model based management and verification mechanism to help and support the exertion of group wisdom. The formation of a knowledge system is a gradual process that integrates the wisdom of multiple people. It helps us manage and validate knowledge through management, collaborative construction, and tracking growth patterns, as Fig3.
[0095] In Figure 7, there are three important modules, knowledge source adapter 71, editing module 72 and manager 73.
[0096] Knowledge source adapter 71: This module focuses on aggregating knowledge from multiple sources in industry and unifying them into a format that supports visual presentation (such as XML or HTML) . As the basic of Editing module 72. based on the customer chosen or imported the knowledge carrier (carrier type) to carry out actions store or monitor the imported info and the importance is to convert the carrier to XML / HTML / JSON and so on, the format which to support the frontend embedded presentation.
[0097] Editing module 72: Forming elements of knowledge entities through front-end crawling operations. This module is designed to help people in various fields select knowledge elements of industries or scenarios through simple crawling (selection) operations. By default, the relationship between elements is constructed based on the order of fetching (automatically track the last selected element) . At the same time, it supports users to modify / edit the description and the connection endpoint by selecting the relationship line by drag-and-drop operation, as Fig1. The core of Editing module 72is to help user to build up the knowledge structure based on grab knowledge element. Its target is to solve the gap of domain experts and developers and save effort and costs during the construction process.
[0098] Manager 73: Knowledge Manager supports user to manage knowledge across various industries / projects and supports team collaboration to build knowledge entities. The key part is Knowledge Assembly and Comparison. Knowledge Assembly: Whether built multiple times or collaborating with multiple people, we have to face the assembly of knowledge structures. We avoid semantic matching or machine learning matching, try to merge or assembly the knowledge structure by drag-and-drop operation easily. Comparison: For a long time, we have always been asked how to verify knowledge. We can understand the customer's pain points from two aspects. Firstly, provide friendly knowledge presentation method, not the KG-DB tool, not need special query statements. This point, we can solve based on the previous modules’ explanation. Secondly, the existing construction methods seem like a dark box operation for users, lacking flexibility, leading to a lack of confidence on the path of scale up. Based on invention, we can easily support the knowledge structure buildup and assembly, then from the perspective of validation, comparison module can provide the tracking and appraise strategy.
[0099] There are several advantages compared to the current solutions. (1) Reduce the difficulty of industry knowledge accumulation and across the gap between development and industry domain. (2) Increase flexibility in building a knowledge system, promote the collaborative assembly and combination of multi-source knowledge, and construct effective knowledge models. (3) Provide model comparison strategy index for users to judge the growth ability, stability, and applicability of the knowledge system. The technical features of embodiments of the present invention which contribute to the advantages are listed as below: (1) Provide engineering building blocks, including the unify convert and grab module to transfer the knowledge materials to knowledge elements. (2) Based on tracking the knowledge elements, provide indexes to measure the knowledge system.
[0100] Embodiments of the present invention enhance knowledge digital services, strengthen domain knowledge accumulation, and knowledge-based service.
[0101] Fig. 8 is a structural diagram of an apparatus for generating a knowledge graph of a production unit according to an embodiment of the present invention. As shown in Figure 8, apparatus for generating a knowledge graph of a production unit 800 comprising: an acquiring module 801, configured to acquire knowledge from a knowledge source of a production unit; a converting module 802, configured to convert the knowledge into a display content suitable for display on a visual interface; a first generating module 803, configured to generate a first knowledge topological graph suitable for display on the visual interface, in response to a user-triggered instruction issued on the display content; and a second generating module (804) , configured to generate a first knowledge graph based on the first knowledge topological graph.
[0102] In one embodiment, the user-triggered instruction comprises at least one of the following: a selection instruction for selecting part or all of the display content; a node generation instruction for generating a node in the first knowledge topological graph based on a selected content; a node connection instruction for connecting nodes in the first knowledge topological graph; a relationship editing instruction for editing relationships between nodes in the first knowledge topological graph; a node deletion instruction for deleting a node in the first knowledge topological graph.
[0103] Embodiments of the present invention also propose an electronic device with a processor memory architecture.
[0104] Fig. 9 is a structural diagram of an electronic device according to an embodiment of the present invention. As shown in Figure 9, electronic device 900 includes a processor 901, a memory 902, and a computer program stored on memory 902 that can run on processor 901. When the computer program is executed by processor 901, the method for generating knowledge graph of production unit as described in either of the above is implemented. Among them, memory 902 can be implemented as various storage media such as electrically erasable programmable read-only memory (EEPROM) , flash memory, programmable program read-only memory (PROM) , etc. Processor 801 can be implemented to include one or more central processors or one or more field programmable gate arrays, wherein the field programmable gate array integrates one or more central processor cores. Specifically, the central processing unit or core can be implemented as a CPU, MCU, DSP, and so on.
[0105] It should be noted that not all steps and modules in the above processes and structural diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution sequence of each step is not fixed and can be adjusted as needed. The division of each module is only for the convenience of describing the functional division used. In actual implementation, a module can be divided into multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be in the same device or different devices.
[0106] The hardware modules in each implementation can be implemented mechanically or electronically. For example, a hardware module can include specially designed permanent circuits or logic devices (such as dedicated processors, such as FPGA or ASIC) to complete specific operations. Hardware modules can also include programmable logic devices or circuits temporarily configured by software (such as general-purpose processors or other programmable processors) for performing specific operations. As for the specific use of mechanical methods, either dedicated permanent circuits or temporarily configured circuits (such as software configuration) to implement hardware modules, it can be determined based on cost and time considerations.
[0107] The above is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this invention shall be included within the scope of protection of this invention.
Claims
1.A method for generating knowledge graph of production unit, comprising:acquiring (101) knowledge from a knowledge source of a production unit;converting (102) the knowledge into a display content suitable for display on a visual interface;generating (103) a first knowledge topological graph suitable for display on the visual interface, in response to a user-triggered instruction issued on the display content; andgenerating (104) a first knowledge graph based on the first knowledge topological graph.2.The method of claim 1, wherein converting (102) the knowledge into a display content suitable for display on a visual interface comprises:determining a type attribute of the knowledge;converting the knowledge into a display content suitable for display on a visual interface based on a conversion tool corresponding to the type attribute.3.The method of claim1, wherein the user-triggered instruction comprises at least one of the following:a selection instruction for selecting part or all of the display content;a node generation instruction for generating a node in the first knowledge topological graph based on a selected content;a node connection instruction for connecting nodes in the first knowledge topological graph;a relationship editing instruction for editing relationships between nodes in the first knowledge topological graph;a node deletion instruction for deleting a node in the first knowledge topological graph.4.The method of claim 3, comprising:displaying a list containing multiple preset relationships in the visual interface;wherein the relationship editing instruction is used to select a relationship from the list.5.The method of claim 3, comprising:displaying an input box in the visual interface;sending a prompt message to input a relationship between nodes in the input box;receiving a string input in the input box;wherein the relationship editing instruction is used to determine the string as a relationship between nodes.6.The method of any one of claims 1-5, wherein generating (104) a first knowledge graph based on the first knowledge topological graph comprises:converting the first knowledge topological graph into a script supported by a database for storing a knowledge graph;writing the script into the database to generate the first knowledge graph.7.The method of any one of claims 1-5, comprising:acquiring a second knowledge topological graph;displaying the second knowledge topological graph in the visual interface;combing the first knowledge topological graph and the second knowledge topological graph in the visual interface;generating a second knowledge graph based on a combined knowledge topological graph.8.The method of claim 7, wherein comprises at least one of the following:knowledge sources of the second knowledge topological graph and the first knowledge topological graph belong to different departments in the production unit;knowledge source of the second knowledge topological graph belongs to another production unit that is different from the production unit;knowledge sources of the second knowledge topological graph and the first knowledge topological graph belong to the same department in the production unit at different times;a user who participates in generating the second knowledge topological graph is different from the user who issues the user-triggered instruction.9.The method of claim 7, wherein combing the first knowledge topological graph and the second knowledge topological graph comprises:determining the same nodes in the first knowledge topological graph and the second knowledge topological graph;merging the same nodes in the combined knowledge topological graph.10.The method of claim 9, comprising:determining a maturity index of the second knowledge graph based on the number of nodes added compared to the first knowledge graph in the second knowledge graph;determining a combination index of the second knowledge graph based on the number of times the same nodes are merged;determining a citation index of the second knowledge graph based on citation number of the second knowledge graph;evaluating quality of the second knowledge graph based on the maturity index, combination index, and citation index.11.An apparatus for generating knowledge graph of production unit, comprising:an acquiring module (801) , configured to acquire knowledge from a knowledge source of a production unit;a converting module (802) , configured to convert the knowledge into a display content suitable for display on a visual interface;a first generating module (803) , configured to generate a first knowledge topological graph suitable for display on the visual interface, in response to a user-triggered instruction issued on the display content; anda second generating module (804) , configured to generate a first knowledge graph based on the first knowledge topological graph.12.The apparatus of claim 11, wherein the user-triggered instruction comprises at least one of the following:a selection instruction for selecting part or all of the display content;a node generation instruction for generating a node in the first knowledge topological graph based on a selected content;a node connection instruction for connecting nodes in the first knowledge topological graph;a relationship editing instruction for editing relationships between nodes in the first knowledge topological graph;a node deletion instruction for deleting a node in the first knowledge topological graph.13.An electronic device, comprising a processor (901) and a memory (902) , wherein an application program executable by the processor (901) is stored in the memory (902) for causing the processor (901) to execute a method for generating knowledge graph of production unit according to any one of claims 1-10.14.A computer-readable medium comprising computer-readable instructions stored thereon, wherein the computer-readable instructions for executing a method for generating knowledge graph of production unit according to any one of claims 1-10.15.A computer program product comprising a computer program, upon the computer program is executed by a processor for executing a method for generating knowledge graph of production unit according to any one of claims 1-10.