A knowledge publishing method and intelligent management system

Abstract

The present invention relates to a knowledge publishing method and an intelligent management system. The knowledge publishing method converts an acquired electronic document of a knowledge file to be published into an electronic document in a target format, and then extracts key information, entity relationships and summary information in the electronic document in the target format, generates a landing page containing all key information according to the extracted key information, generates a knowledge cloud map according to the electronic document in the target format, the extracted key information and the entity relationship, and establishes the target format electronic document, the landing page and the knowledge cloud map as a knowledge attribution association between the knowledge as a whole and its affiliated department, and then publishes the knowledge as a whole after the knowledge attribution association has been established as knowledge, thereby realizing automatic knowledge extraction and automatic publishing of the knowledge to be published, improving the efficiency of knowledge publishing, and facilitating users to search and query the published knowledge.

Description

Technical Field

[0001] The present invention relates to the field of engineering knowledge management, and in particular to a knowledge publishing method and an intelligent management system. Background Art

[0002] In engineering design fields such as petrochemical engineering and construction engineering, engineering design companies usually collect project documents such as finished product documents of completed projects, process documents during project execution, and project auxiliary documents such as design manuals, professional books, design knowledge points and field materials, and then use the collected project documents and auxiliary documents to establish the engineering design company's own knowledge system.

[0003] For project documents, companies usually collect electronic files of these project documents and simply classify projects, units, disciplines and file types according to the electronic file numbers. However, the simple classification method based on numbers cannot reflect other details in the project documents, nor can it conduct targeted retrieval of the content in the project documents, which is not conducive to the effective use of the company's intellectual assets. In addition, project-related documents usually have PDF, Word, Excel, PPT, DWG, pictures and other formats, which are huge in number and located in different systems. It is more difficult to retrieve and use different types of documents in different systems. Summary of the invention

[0004] The first technical problem to be solved by the present invention is to provide a knowledge publishing method for the above-mentioned prior art. The knowledge publishing method can automatically process the knowledge to be published and then publish it, realize unified retrieval of documents of different types stored in different systems, and facilitate users to query and manage the published knowledge.

[0005] The second technical problem to be solved by the present invention is to provide an intelligent management system that applies the above-mentioned knowledge publishing method.

[0006] The technical solution adopted by the present invention to solve the first technical problem is: a knowledge publishing method, characterized in that it includes the following steps:

[0007] Step S1, obtaining the electronic document of the knowledge file of the knowledge to be released;

[0008] Step S2, converting the electronic document of the knowledge file to be published into an electronic document in a target format;

[0009] Step S3, extracting key information from the target format electronic document;

[0010] Step S4, extracting entity relationships in the target format electronic document, and constructing a knowledge graph using the extracted entity relationships;

[0011] Step S5, extracting summary information from the target format electronic document;

[0012] Step S6, generating a landing page containing all key information based on the extracted key information;

[0013] Step S7, generating a knowledge cloud graph according to the target format electronic document, the extracted key information and entity relationships;

[0014] Step S8, according to the type and source of the knowledge to be released, the target format electronic document, landing page and knowledge cloud map are associated with the department to which they belong as a whole, and the knowledge attribution association is established;

[0015] Step S9, publishing the entire knowledge after establishing knowledge attribution association as knowledge.

[0016] Furthermore, in the knowledge publishing method, in step S3, the key information extraction process includes the following steps:

[0017] Step S31, determining the type of the target format electronic document:

[0018] When the target electronic document is a graphic document or a table document, the process proceeds to step S32; otherwise, the process proceeds to step S4;

[0019] Step S32, converting the target format electronic document into a target document of a preset type;

[0020] Step S33, regionally locating and dividing the key information in the target document; wherein, when the preset type is a graphic document, the key information is a graphic label; when the preset style is a table document, the key information is a table header;

[0021] Step S34, defining a key information area in the target document, and setting a coordinate range of the key information area in a plane rectangular coordinate system according to a drawing size corresponding to the target document;

[0022] Step S35, dividing the key information area into a target document type area, a key information identification area and a key value area; wherein each area after the key information area is divided corresponds to a different coordinate range in a plane rectangular coordinate system, and each area after the key information area is divided has one-to-one corresponding key information;

[0023] Step S36, extracting information from the target document type area, the key information identification area and the key value area to obtain corresponding key information.

[0024] Improved, in the knowledge publishing method, in step S3, it also includes: labeling the target format electronic document according to the extracted key information to obtain a labeled target format electronic document with labels. Labeling the target format electronic document can further narrow the search scope by using the labels when searching for knowledge, thereby improving the accuracy of the search results.

[0025] As a further improvement, in the knowledge publishing method, the key information in step S3 includes at least one of a project name, a unit name, a design phase, a profession and a knowledge type, and the label is information related to the project.

[0026] Optionally, in the knowledge publishing method, in step S1, the knowledge to be published is acquired through a crawler or a system interface.

[0027] Improved, in the present invention, the knowledge publishing method also includes a process for processing the creation version information of the published knowledge.

[0028] As a further improvement, in the knowledge publishing method, the target format electronic document is a PDF document.

[0029] The technical solution adopted by the present invention to solve the second technical problem is: an intelligent management system, characterized in that any one of the knowledge publishing methods described is applied.

[0030] In order to facilitate users to acquire knowledge and reduce unnecessary time costs, the intelligent management system of the invention also performs intelligent knowledge push.

[0031] Compared with the prior art, the advantages of the present invention are: the knowledge publishing method of the invention converts the acquired electronic document of the knowledge file to be published into an electronic document in a target format, and then extracts the key information, entity relationship and summary information in the electronic document in the target format, and generates a landing page containing all the key information based on the extracted key information, and generates a knowledge cloud map based on the electronic document in the target format, the extracted key information and the entity relationship, and establishes the target format electronic document, the landing page and the knowledge cloud map as the knowledge as a whole and the knowledge to which it belongs. The knowledge as a whole with the knowledge attribution association established is then published as knowledge, thereby realizing the automatic extraction and automatic publication of the knowledge to be published, improving the efficiency of knowledge publishing, and facilitating users to search and query the published knowledge. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic diagram of a knowledge publishing method flow in an embodiment of the present invention;

[0033] Figure 2 is a schematic diagram of an intelligent management system in an embodiment of the present invention;

[0034] Figure 3 for Figure 2 A schematic diagram of the framework of the system management module in the intelligent management system shown;

[0035] Figure 4 for Figure 2 The framework diagram of the knowledge management module in the intelligent management system shown. DETAILED DESCRIPTION

[0036] The present invention is further described in detail below with reference to the accompanying drawings.

[0037] This embodiment provides a knowledge publishing method, which can be specifically applied to knowledge publishing in the field of petrochemical engineering. Figure 1 As shown, the knowledge publishing method of this embodiment includes the following steps S1 to S9:

[0038] Step S1, obtaining the electronic document of the knowledge file of the knowledge to be released through a crawler or a system interface; wherein, in actual situations, the electronic document of the knowledge file of the knowledge to be released may be a PDF document, a Word document, an Excel document, a picture document or a DWG document; the electronic document of the knowledge file of the knowledge to be released in this embodiment is assumed to be a Word document, and the name of the Word document is A.doc;

[0039] Step S2, converting the electronic document of the knowledge file to be published into an electronic document in a target format; wherein the electronic document in the target format is assumed to be a PDF document, then after the Word document A.doc in this embodiment is converted into a converted PDF document, the name of which is A.pdf;

[0040] Step S3, extracting key information from the target format electronic document; wherein the key information here includes project name, unit name, design stage, specialty and knowledge type and other project-related information;

[0041] Specifically in this embodiment, executing step S3 is to extract key information such as project name, unit name, design stage, major and knowledge type in the A.pdf file;

[0042] Specifically, in this embodiment, the key information extraction process includes the following steps S31 to S36:

[0043] Step S31, determining the type of the target format electronic document:

[0044] When the target electronic document is a graphic document or a table document, the process proceeds to step S32; otherwise, the process proceeds to step S4;

[0045] Step S32, converting the target format electronic document into a target document of a preset type; for example, for a graphic document, the target format electronic document can be converted into an A1 format and the layout mode of the document can be a horizontal or vertical target document B.pdf; for a table document, the target format electronic document can be converted into an A4 format and the layout mode of the document can be a horizontal or vertical target document B.pdf;

[0046] Step S33, regionally locating and dividing the key information in the target document; wherein, when the preset type is a graphic document, the key information is a graphic label; when the preset style is a table document, the key information is a table header;

[0047] Step S34, defining a key information area in the target document, and setting a coordinate range of the key information area in a plane rectangular coordinate system according to a drawing size corresponding to the target document;

[0048] For example, if the target document B.pdf is a graphic document, then the converted drawing size corresponding to the graphic document B.pdf is A1, and the key information area of ​​the drawing label in the target document B.pdf is marked as C. Then, according to the drawing size of A1, the coordinate range of the key information area C in the plane rectangular coordinate system is set in the target document B.pdf;

[0049] Step S35, dividing the key information area into a target document type area, a key information identification area and a key value area; wherein each area after the key information area is divided corresponds to a different coordinate range in a plane rectangular coordinate system, and each area after the key information area is divided has one-to-one corresponding key information;

[0050] Based on the key information area C with a set coordinate range in the target document B.pdf, in step S35, the key information area C needs to be divided into a target document type area C1, a key information identification area C2 and a key value area C3; these three key information areas can be continuous areas or non-continuous areas; the number of areas after the key information area is divided can be adjusted as needed; wherein:

[0051] The coordinate range corresponding to the target document type area C1 in the plane rectangular coordinate system is marked as: abscissa x∈(x C1,min ,x C1,max ), the ordinate y∈(y C1,min ,y C1,max );

[0052] The coordinate range corresponding to the key information identification area C2 in the plane rectangular coordinate system is marked as: horizontal coordinate x∈(x C2,min ,xC2,max ), the ordinate y∈(y C2,min ,y C2,max );

[0053] The coordinate range corresponding to the key value area C3 in the plane rectangular coordinate system is marked as: horizontal coordinate x∈(x C3,min ,x C3,max ), the ordinate y∈(y C3,min ,y C3,max );

[0054] Step S36, extract information from the target document type area, key information identification area and key value area to obtain the corresponding key information. Since each area divided in the key information area C has its own key information, the corresponding information in the target document type area C1, key information identification area C2 and key value area C3 is extracted respectively, so that the key information corresponding to each area can be obtained;

[0055] Step S4, extracting entity relationships in the target format electronic document, and constructing a knowledge graph using the extracted entity relationships;

[0056] Assume that in the converted target format electronic document A.pdf, there are related entities such as "safety valve" and "bursting disc", then step S4 extracts the entities "safety valve" and "bursting disc" in A.pdf, and then constructs a knowledge graph based on the extracted related entities "safety valve" and "bursting disc". The knowledge graph can be constructed using the current commonly used knowledge graph construction method;

[0057] Step S5, extracting summary information from the target format electronic document;

[0058] After the entity relationship is extracted, the summary information in the target electronic document A.pdf is also extracted;

[0059] Step S6, generating a landing page containing all the key information based on the extracted key information; wherein, in this embodiment, the landing page contains the key information of the knowledge, and the content contained in the landing page is different according to the different knowledge classifications; for example, the landing page of the standard specification contains key information such as standard classification, keywords, standard number, standard status, referenced standard, alternative standard, replaced standard, issuing department, issuing date, implementation date, drafting unit and standard introduction;

[0060] Step S7, generating a knowledge cloud graph according to the target format electronic document, the extracted key information and entity relationships;

[0061] That is, according to the target format electronic document A.pdf, the extracted key information such as the project name, unit name, design stage, specialty and knowledge type, and the entity relationship between "safety valve" and "bursting disc", a knowledge cloud graph is generated, and the generation of the knowledge cloud graph can adopt the existing conventional method;

[0062] Step S8, according to the type and source of the knowledge to be released, the target format electronic document, landing page and knowledge cloud map are associated with the department to which they belong as a whole knowledge, and a knowledge attribution association is established; wherein the knowledge attribution department association can be automatically established according to the source and type of the knowledge, and the public knowledge belongs to all departments, and the non-public knowledge is assigned to the corresponding department according to the established knowledge label information such as profession and variety; that is, the "knowledge attribution association" established here is the association between the knowledge as a whole and the department to which the knowledge as a whole belongs;

[0063] Step S9: publish the knowledge as a whole after establishing the knowledge attribution association. According to actual needs, version information can also be created for the published knowledge, so as to query and learn the version corresponding to each published knowledge.

[0064] In order to manage and maintain the target format electronic document, in step S3 of this embodiment, the target format electronic document can also be tagged according to the extracted key information to obtain a tagged target format electronic document with a tag. The tag is set to information related to the project, such as the project name, construction time, construction location, project scale, construction unit, project type, profession, knowledge type and knowledge source.

[0065] In addition, this embodiment also provides an intelligent management system. Specifically, the intelligent management system of this embodiment applies the above-mentioned knowledge publishing method. In addition, the intelligent management system of this embodiment also performs intelligent knowledge push. Specifically, the process of the intelligent management system performing intelligent knowledge push includes the following steps:

[0066] Step 1: pre-build an engineering design knowledge database; wherein the engineering design knowledge database contains at least historical knowledge that has been accumulated by the engineering design company, and the historical knowledge includes professional knowledge and non-professional knowledge related to engineering design; the non-professional knowledge may be knowledge such as production operation, enterprise management, and general knowledge;

[0067] Step 2: Establish a knowledge correspondence relationship based on the constructed engineering design knowledge database; wherein the knowledge correspondence relationship includes personalized feature information and historical knowledge corresponding to the personalized feature information, and the personalized feature information is at least one of the knowledge retrieval behavior information of the personnel, personalized personnel information, and work scene information; the personalized personnel information includes at least one of the user's role, position, job, and profession; the work scene information includes work content information; the knowledge retrieval behavior information includes search terms or sentences; the user here refers to the user in the unit who needs to obtain knowledge;

[0068] Step 3, obtaining the personalized characteristic information of the user; for example, a user A in a petrochemical enterprise needs to search for "safety valve setting principles", at this time, the sentence "safety valve setting principles" retrieved by the user A can be obtained, and the sentence "safety valve setting principles" is the knowledge retrieval behavior information of the user A; then, this knowledge retrieval behavior information is the obtained personalized characteristic information of the user A; semantic analysis can be performed on the knowledge retrieval behavior information, and the analysis process can obtain key knowledge information, which is a keyword or a key sentence;

[0069] Step 4: According to the acquired personalized feature information, the historical knowledge that is located in the knowledge correspondence relationship and corresponds to the personalized feature information is pushed to the user corresponding to the personalized feature information. As mentioned above, assuming that the knowledge retrieval behavior information as the personalized feature information of user A has been acquired by executing step 3, then the historical knowledge that is located in the knowledge correspondence relationship and corresponds to the personalized feature information of the knowledge retrieval behavior information of user A is pushed to the user corresponding to the personalized feature information.

[0070] In this embodiment, the personalized feature information can be set as the knowledge retrieval behavior information of user A as needed. In this case, the intelligent push method of this embodiment further includes the following steps 51 to 53:

[0071] Step 51, extracting entities from the user's knowledge retrieval behavior information;

[0072] That is, for the above-mentioned person in the unit - user A, entities can be extracted from the knowledge retrieval behavior information of user A, and the entity "safety valve" can be extracted;

[0073] Step 52, searching for all associated entities associated with the extracted entity in a pre-constructed knowledge graph; wherein the knowledge graph is obtained based on historical knowledge in a constructed engineering design knowledge database;

[0074] By pre-processing the historical knowledge in the engineering design knowledge database to obtain a knowledge graph, and then extracting the entity "safety valve" based on the knowledge retrieval behavior information of user A, all related entities associated with the entity "safety valve" can be obtained. For example, all related entities associated with the entity "safety valve" are the entity "bursting disc";

[0075] Step 53: Push the historical knowledge that is located in the knowledge correspondence and is related to all associated entities to the user corresponding to the knowledge retrieval behavior information.

[0076] Since the search in the knowledge graph in step 52 has obtained that all associated entities associated with the entity "safety valve" are the entity "bursting disc", the historical knowledge that is located in the knowledge correspondence and related to all associated entities "bursting disc" is pushed to user A.

[0077] It should be noted that step 51 and step 4 executed here are processed simultaneously and in parallel.

[0078] In order to meet the differences in personalized personnel information such as different roles, positions and posts of users in the organization, so as to push historical knowledge matching the user, the execution process of the intelligent push method of this embodiment may further include the following steps:

[0079] Get the user's personalized personnel information;

[0080] Through the correspondence between personalized information and knowledge, the historical knowledge that is in the knowledge correspondence and corresponds to the personalized personnel information of the user is pushed to the user. In the knowledge correspondence, each type of personalized personnel information has its corresponding historical knowledge; the personalized personnel information includes at least one of the user's role, position, job, and profession.

[0081] Considering that the user may change between different positions, different posts or different roles, and timely push the historical knowledge matching the latest personalized personnel information of the user, the intelligent push method of this embodiment also has the following improvement measures:

[0082] Step b1, obtaining the latest personalized personal information of the user at a preset interval; wherein the preset interval here can be adjusted as needed, for example, the preset interval is set to one month or one quarter;

[0083] Step b2: Push the historical knowledge that is in the knowledge correspondence and corresponds to the user's latest personalized personal information to the user. In this way, when the user's personalized personal information changes, the historical knowledge that matches it can be pushed to the user in a timely and accurate manner, so that the user can learn knowledge in a timely manner.

[0084] In addition, the user's current work content (i.e., work scene information) can also be obtained, and then, based on the current work content of the user, the historical knowledge related to the user's current work can be pushed to the user. For example, the user's current work content can be obtained through identification based on the user interface of the device used by the user, that is, if the user runs a certain software on the work interface of the computer, then the user's work content can be obtained based on the running interface of the software. The premise of this solution is that the user's authorization must be obtained before it can be carried out. Without the user's authorization, identification of work content based on the user interface of the device used by the user is not allowed.

[0085] In order to avoid pushing too much historical knowledge to users who need to obtain historical knowledge, the intelligent knowledge push method in this embodiment also pre-sets the push relevance value for historical knowledge, and searches for historical knowledge that is related to personalized feature information and meets the push relevance value in the engineering design knowledge database; finally, the historical knowledge that meets the push relevance value is pushed to the corresponding user.

[0086] This embodiment also provides an intelligent management system, which applies the above-mentioned engineering design knowledge intelligent push method. Figure 2 As shown, the intelligent management system of this embodiment includes a system management module 1, a knowledge management module 2, a knowledge search module 3, a personal center management module 4, an intelligent question and answer module 5, an intelligent push module 6, a knowledge forum module 7 and a knowledge graph module 8. The intelligent push module 6 is responsible for pushing engineering design knowledge. Among them:

[0087] See also Figure 3 As shown, the system management module 1 in this embodiment includes an organization management submodule 11, a position management submodule 12, a menu management submodule 13, a role management submodule 14, a label system management submodule 15, a log management submodule 16, a shielded keyword management submodule 17, a crawler configuration submodule 18, a knowledge feedback processing submodule 19 and an expert library management submodule 110. Among them:

[0088] The organization management submodule 11 is used to synchronize with the internal collaborative office platform of the enterprise and manage the units where the system users are located; for example, the organization management submodule 1 is used to manage different units within the same petrochemical enterprise;

[0089] The position management submodule 12 is used to synchronize the position management with the internal integrated project management system of the enterprise, and manage the positions of the system users; specifically, for example, it can manage the positions of the system users (such as design, verification, review, approval, etc.);

[0090] The menu management submodule 13 manages the system menu of the engineering design knowledge auxiliary management system; for example, the main interface and sub-interface of the engineering design knowledge auxiliary management system can be adjusted as necessary according to the needs of the system user and based on the authority of the system user;

[0091] The role management submodule 14 manages the role permissions of the system user in the engineering design project; the role of the system user in the engineering design project here can be, for example, a common user, a data manager, or a system manager, or a role that is added according to the needs of the engineering design project, and the role management submodule 14 can grant different data processing permissions based on the different roles of the user. For example, after the system user logs in to the system, the system first determines the role of the system user according to the role management submodule 14. If the current role of the system user is a common user, the system user is only granted the permission to search for knowledge and participate in the knowledge forum, but not other permissions. Of course, according to actual needs, the personal center module, intelligent question and answer, intelligent push and knowledge graph module can also be opened to common users. With the increase of the permissions corresponding to the role of the system user, the system user can obtain more operation permissions. Of course, the role management submodule 14 can also grant different permissions to the roles assumed by the system user at different stages of the project as needed. That is, for the same engineering design project, the permissions of the system user at different stages in the same project are different. That is to say, in the same engineering design project, users in the previous stage do not have the operating permissions that users in the next stage have, which can also achieve the purpose of keeping the engineering project confidential.

[0092] The label system management submodule 15 is used to label the knowledge files in the engineering design knowledge auxiliary management system. It should be noted that the label system management submodule 15 here can label various knowledge files in the engineering design project (such as general contracts, sub-contracts, rules and regulations, laws and regulations, and other commonly used files) with corresponding labels, so as to facilitate the use of the labels to distinguish and identify different knowledge files. Through such labeling operations, the search efficiency and accuracy of system users for knowledge files can be improved.

[0093] For example, in this embodiment, the tag system management submodule 15 can tag different knowledge files in the following steps a1 to a3:

[0094] Step a1: perform document structuring on the knowledge document to be labeled and extract entity words;

[0095] Step a2, counting the occurrence frequency of each extracted entity word in the corresponding knowledge document;

[0096] Step a3: make a judgment based on the statistical results of the frequency of occurrence of each entity word:

[0097] When the frequency of occurrence of any entity word is higher than the preset frequency value, the entity word is listed as the subject of the knowledge document to be annotated; otherwise, the entity word with the highest frequency of occurrence is listed as the subject of the knowledge document to be annotated; wherein the subject here is the label of the knowledge document to be annotated.

[0098] The log management submodule 16 manages various log information generated by the system; wherein the various log information includes system logs, security logs, audit logs and network logs; the system administrator can detect the cause of the system error or find the trace information of the network attacker left when the system is attacked by the network by operating the log management submodule 16;

[0099] The shielding keyword management submodule 17 performs shielding management operations on preset sensitive words; wherein the management operations include adding operations, deleting operations, and modifying operations; through the shielding processing performed by the shielding keyword management submodule 17 on the sensitive words set in advance, it can be ensured that the system users will not find any sensitive words in the search results of the system of this embodiment;

[0100] The crawler configuration submodule 18 edits and configures the crawler operation to realize knowledge crawling of data from different network resources. By setting up the crawler configuration submodule 18, this embodiment can integrate authoritative industry external website resources such as standards and specifications, industry consultations, etc. through crawling according to the work needs of designers, thereby greatly improving work efficiency and quality.

[0101] The knowledge feedback processing submodule 19 receives feedback information from system users and processes the feedback information. Specifically, system users can provide feedback on the accuracy and relevance of the knowledge or search results in the system to optimize the accuracy of system knowledge and search results.

[0102] The expert database management submodule 110 is responsible for maintaining and managing the enterprise's internal expert information, and providing answers to knowledge questions raised by system users through the knowledge forum from enterprise internal experts in the field corresponding to the knowledge questions.

[0103] See also Figure 4 As shown, the knowledge management module 2 in this embodiment comprises a knowledge release management submodule 21, a knowledge release query submodule 22 and a knowledge version management submodule 23. Among them:

[0104] The knowledge release management submodule 21 releases the knowledge tagged by the tag system management submodule; the knowledge release management submodule 21 can be based on automatic release of the system or manual release of the user. Automatic release can be the system using the knowledge release management submodule 21 to release knowledge in batches or individually, and the tag system management submodule 15 can perform tagging operations according to the content of the knowledge to be released, so that each piece of knowledge issued by the knowledge release management submodule 21 can have a tag for identification;

[0105] The knowledge publishing query submodule 22 queries the knowledge that has been published by the knowledge publishing management module according to the query operation of the user;

[0106] The knowledge version management submodule 23 manages different versions of the same knowledge file.

[0107] The knowledge search module 3 has a historical project engineering data module 31, and the knowledge search module 3 searches for the target knowledge according to the target knowledge input by the user; wherein, the historical project engineering data module 31 stores the historical engineering project files managed by at least one unit in the enterprise; specifically, in this embodiment, the knowledge search module 3 includes a knowledge search main interface, which includes sub-modules such as a search box, a system hot search, and hot questions; according to actual business needs, modules such as standards and regulations, the latest information, industry encyclopedias, professional knowledge, bidding and procurement, skill improvement, and historical project engineering data can also be set to facilitate users to search for business needs according to their own needs; wherein, for the results searched by the knowledge search module 3, the ranking of indicators such as download volume, reading volume, and release time can be provided, and the search results can be streamlined according to different label options, so as to facilitate the display of the searched results from different dimensions and in a more targeted manner, thereby improving search efficiency and accuracy. In addition, different permissions (browse, preview, and download, etc.) can also be provided according to the permissions of different system users. If there is no corresponding permission, the user can apply for permission to obtain the relevant knowledge files in the search results.

[0108] The personal center management module 4 is used to manage at least the modules related to the system user operation in the engineering design knowledge auxiliary management system. For example, the personal center management module 4 provides a personal center homepage, which includes relevant sub-item columns such as my knowledge release, knowledge collection, knowledge question, knowledge question and answer, knowledge comment, knowledge feedback, knowledge sharing and knowledge approval.

[0109] The intelligent question-answering module 5 is used to provide answers to the knowledge questions based on the user's knowledge questions, and provide knowledge files related to the answers for detailed explanations of the answers. For example, the system user can ask non-specified professional questions, and then the system can use the intelligent question-answering module 5 to answer these non-specified professional questions raised by the user, and can also make necessary professional explanations as needed.

[0110] The intelligent push module 6 is used to push knowledge corresponding to the personalized personnel information of the user according to the personalized personnel information of the user. The personalized personnel information here can be the user's position, role, and search habits, etc. In this way, the intelligent push module 6 can accurately provide the user with professional knowledge or other knowledge information matching the user according to the obtained personalized personnel information of the user.

[0111] The knowledge forum module 7 is used to share the knowledge questions released by the system users through the knowledge release management submodule and provide them to professional field personnel or non-professional field personnel directly or indirectly to answer the questions released by the system users. For example, the professional field personnel here can be experts or ordinary personnel in the field corresponding to the knowledge question. In the specific implementation process, the knowledge forum module 7 includes sub-columns such as knowledge questions, knowledge questions and answers, knowledge search and experience sharing. Users can use the knowledge forum module to publish knowledge questions, and invite experts to answer their own knowledge questions by prompting professional field experts (such as experts in their own companies). The knowledge forum module 7 can also summarize and organize the excellent questions raised and the excellent answers to the questions, and then organize them into more formal knowledge documents to facilitate other users to query related knowledge questions and knowledge answers in the future. The organized knowledge questions and answers can be further upgraded to system knowledge files for publication as needed for users to query and download.

[0112] The knowledge graph module 8 is used to draw the knowledge related to the target knowledge searched by the system user in the form of a graph, and display the drawn graph to the system user. Specifically, the knowledge graph module 8 can draw the knowledge related to the target knowledge searched by the user in the form of a graph through data mining, information processing, knowledge measurement and graph drawing, and then display these graphs to the system user, thereby greatly improving the relevance and accuracy of the search results.

[0113] It should be noted here that, for the intelligent management system in this embodiment, by setting a knowledge search module, and in the knowledge search module, a historical project engineering data module for storing historical engineering project files managed by at least one unit in the enterprise is set, and by setting the historical project engineering data module, it is convenient for system users to query previous historical engineering project files, thereby improving the reuse rate and work efficiency of historical engineering project files. In addition, through the set system management module, it is possible to share historical engineering project files between different units in the enterprise, thereby improving the sharing efficiency and reuse rate of historical engineering project files between different units in the enterprise, improving the project management efficiency and work efficiency of the entire enterprise, and also avoiding the disadvantages of different files under the same traditional engineering design project being independently and scatteredly distributed among different units and different technicians in the enterprise.

[0114] When executing the intelligent knowledge push, the intelligent management system of this embodiment pre-builds a knowledge database, and establishes a knowledge correspondence relationship containing personalized feature information and corresponding historical knowledge based on the knowledge database, and then pushes the historical knowledge corresponding to the personalized feature information and located in the knowledge correspondence relationship to the user corresponding to the personalized feature information based on the acquired personalized feature information of the user, thereby realizing the intelligent push of knowledge to users who need to learn, and reducing the time cost of users to acquire historical knowledge. Secondly, by acquiring the latest personalized personnel information of the user, after the personalized personnel information of the user changes, the intelligent management system can automatically and intelligently push the knowledge matching the changed personalized personnel information to the user, so as to facilitate the user to obtain the knowledge required for the position in a timely manner. At the same time, according to the user's work scenario, such as work content, the knowledge related to the current work scenario can be pushed to the user, thereby providing reference knowledge for the user's current work.

[0115] It should be noted that in actual projects, the engineering design knowledge in the above-mentioned knowledge intelligent push method and intelligent management system can be replaced with other knowledge as needed to meet the needs of intelligent push of knowledge in different business fields.

[0116] Although the preferred embodiments of the present invention are described in detail above, it should be clearly understood that various modifications and variations are possible for those skilled in the art. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A knowledge publishing method, characterized in that: The steps include: Step S1, obtaining the electronic document of the knowledge file of the knowledge to be released; Step S2, converting the electronic document of the knowledge file to be published into an electronic document in a target format; Step S3, extracting key information from the target format electronic document; Step S4, extracting entity relationships in the target format electronic document, and constructing a knowledge graph using the extracted entity relationships; Step S5, extracting summary information from the target format electronic document; Step S6, generating a landing page containing all key information based on the extracted key information; Step S7, generating a knowledge cloud graph according to the target format electronic document, the extracted key information and entity relationships; Step S8, according to the type and source of the knowledge to be released, the target format electronic document, landing page and knowledge cloud map are associated with the department to which they belong as a whole, and the knowledge attribution association is established; Step S9, publishing the entire knowledge after establishing knowledge attribution association as knowledge.

2. The knowledge publishing method according to claim 1, characterized in that: In step S3, the key information extraction process includes the following steps: Step S31, determining the type of the target format electronic document: When the target electronic document is a graphic document or a table document, the process proceeds to step S32; otherwise, the process proceeds to step S4; Step S32, converting the target format electronic document into a target document of a preset type; Step S33, regionally locating and dividing the key information in the target document; wherein, when the preset type is a graphic document, the key information is a graphic label; when the preset style is a table document, the key information is a table header; Step S34, defining a key information area in the target document, and setting a coordinate range of the key information area in a plane rectangular coordinate system according to a drawing size corresponding to the target document; Step S35, dividing the key information area into a target document type area, a key information identification area and a key value area; wherein each area after the key information area is divided corresponds to a different coordinate range in a plane rectangular coordinate system, and each area after the key information area is divided has one-to-one corresponding key information; Step S36, extracting information from the target document type area, the key information identification area and the key value area to obtain corresponding key information.

3. The knowledge publishing method according to claim 1, characterized in that: In step S3, the process also includes: tagging the target format electronic document according to the extracted key information to obtain a tagged target format electronic document with tags.

4. The knowledge publishing method according to claim 3, characterized in that: The key information in step S3 includes at least one of a project name, a unit name, a design phase, a profession, and a knowledge type, and the label is information related to the project.

5. The knowledge publishing method according to claim 1, characterized in that: In step S1, the knowledge to be published is obtained through a crawler or a system interface.

6. The knowledge publishing method according to claim 1, characterized in that: It also includes the process of creating version information for published knowledge.

7. The knowledge publishing method according to any one of claims 1 to 6, characterized in that: The target electronic document format is a PDF document.

Images