CN117112809B - Knowledge tracking method and system - Google Patents

Knowledge tracking method and system Download PDF

Info

Publication number
CN117112809B
CN117112809B CN202311385302.0A CN202311385302A CN117112809B CN 117112809 B CN117112809 B CN 117112809B CN 202311385302 A CN202311385302 A CN 202311385302A CN 117112809 B CN117112809 B CN 117112809B
Authority
CN
China
Prior art keywords
knowledge
entity
text
text data
association
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202311385302.0A
Other languages
Chinese (zh)
Other versions
CN117112809A (en
Inventor
屠静
王亚
赵策
张玥
雷媛媛
孙岩
潘亮亮
刘岩
刘莎
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhuoshi Future Beijing technology Co ltd
Original Assignee
Zhuoshi Future Beijing technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhuoshi Future Beijing technology Co ltd filed Critical Zhuoshi Future Beijing technology Co ltd
Priority to CN202311385302.0A priority Critical patent/CN117112809B/en
Publication of CN117112809A publication Critical patent/CN117112809A/en
Application granted granted Critical
Publication of CN117112809B publication Critical patent/CN117112809B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G06F16/18—File system types
    • G06F16/182—Distributed file systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/31—Indexing; Data structures therefor; Storage structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/3332—Query translation
    • G06F16/3335—Syntactic pre-processing, e.g. stopword elimination, stemming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3344—Query execution using natural language analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

The invention provides a knowledge tracking method and a knowledge tracking system, and belongs to the technical field of text information processing. The method comprises the following steps: extracting knowledge text data and caching the knowledge text data to a distributed file system; automatically identifying entity elements in the knowledge text data and association relations T between the entity elements through a pre-deployed AI entity identification model, and mapping and storing the association relations T to a database; the method comprises the steps that a knowledge base construction module is informed at fixed time, the association relation T is extracted, a knowledge map construction activity G is executed on the knowledge base construction module based on the association relation T, a knowledge map is generated by fusing text abstracts, emotion/content trends and attitudes of entity elements, and the knowledge map is recommended to a knowledge management platform in real time; and managing the knowledge graph through the knowledge management platform, and enabling a user to access the knowledge graph through an application end to develop the retrieval application of the knowledge graph. By adopting the method and the device, intelligent management, retrieval and tracking of knowledge can be realized.

Description

Knowledge tracking method and system
Technical Field
The invention relates to the technical field of text information processing, in particular to a knowledge tracking method and system.
Background
Knowledge management is becoming vital in the age of information explosion today. In the knowledge learning process, such as the student learning process and the enterprise production learning process, a large amount of texts, images or other types of knowledge materials can be derived, and the knowledge materials contain abundant precious knowledge which does not appear in the existing resources, so that the knowledge learning process has a large knowledge value, in particular technical knowledge, technical experience and the like. If the knowledge materials are better subjected to knowledge management and extracted and utilized, more value can be brought to users or enterprises.
Knowledge tracking plays a major role in knowledge management. Aims at tracking the technical materials generated in the learning and production processes, constructing a corresponding knowledge base, and putting the knowledge base into subsequent learning or production operation.
The existing knowledge base manages knowledge assets of enterprises or users by employing knowledge base management tools/assistants. The prior general knowledge base is mainly a database for knowledge management and is used for collecting, rearranging and extracting knowledge in related application fields. Knowledge in the knowledge base is derived from experience and training of an expert or a professional, and although the knowledge base is simple in construction, some problems exist:
The knowledge data managed in the prior knowledge base mainly comprises the collection of knowledge in the related application field and the formal arrangement and application, so that the knowledge base lacks entity identification of knowledge elements, lacks association relation among the knowledge elements, has fuzzy association information, cannot intuitively provide visualized association data for enterprises or users, and is inconvenient for management, retrieval and tracking of the knowledge information;
knowledge data in the prior knowledge base is a knowledge base document provided for a user, so that the knowledge base document is relatively direct, the user is directly provided with the arranged experience and training, the key information extraction of the knowledge data and the analysis of knowledge elements are lacked, and the deep knowledge analysis application is lacked.
Disclosure of Invention
The embodiment of the invention provides a knowledge tracking method and a knowledge tracking system, which can carry out orderly and visual knowledge data management and application by relying on the association relation of knowledge entities, can enable a user to quickly know the corresponding knowledge point positions and the association by fusing the generated knowledge maps, provide the association of the knowledge points, emotion/content tendency and attitude analysis of knowledge texts, provide the user with better knowledge content understanding, deep analysis and knowledge map application, and maximize the utilization of knowledge value, thereby realizing intelligent management, retrieval and knowledge tracking. The technical scheme is as follows:
In one aspect, a knowledge tracking method is provided, and the method is applied to an electronic device, and includes:
s1, extracting knowledge text data and caching the knowledge text data to a distributed file system;
s2, automatically identifying entity elements and association relations T among the entity elements in the knowledge text data cached in the distributed file system through a pre-deployed AI entity identification model, and mapping and storing the association relations T to a database;
s3, extracting the association relation T by a timing notification knowledge base construction module, executing a knowledge graph construction activity G on the knowledge base construction module based on the association relation T to fuse text abstracts, emotion/content trends and attitudes of entity elements to generate a knowledge graph, and recommending the knowledge graph to a knowledge management platform in real time;
and S4, managing the knowledge graph through the knowledge management platform, and enabling a user to access through an application end to develop the retrieval application of the knowledge graph.
Further, the extracting knowledge text data and caching to the distributed file system includes:
s11, extracting knowledge text data, and preprocessing and cleaning the extracted knowledge text data in batches;
S12, classifying the knowledge text data after batch processing according to text types to obtain a data set M consisting of a plurality of knowledge text data blocks with different text types, wherein,
m= { knowledge text data block 1, knowledge text data block 2, knowledge text data block 3.
S13, orderly numbering each knowledge text data block in the data set M according to a preset knowledge text priority, and carrying out sequence priority rearrangement to obtain a knowledge text optimal ranking data set N;
s14, traversing all storage nodes of a distributed file system, checking available storage nodes, and distributing and storing all knowledge text data blocks in the knowledge text priority data set N in the storage nodes of the distributed file system according to a priority rearrangement order;
and S15, the storage addresses of the knowledge text data blocks are sent to a background server.
Further, the automatically identifying the entity elements and the association relations T between the entity elements in the knowledge text data cached in the distributed file system through the pre-deployed AI entity identification model, and storing the association relations T in a database includes:
S21, when the background server receives the storage address, notifying a pre-deployed AI entity identification model to call a knowledge text data block stored in the storage address;
s22, carrying out entity recognition on the knowledge text data block through the AI entity recognition model, automatically recognizing to obtain an entity element m in the knowledge text data block, and carrying out association recognition to obtain an association relation T between the entity elements according to the context text information of the entity element m: m1→m2; wherein m1 and m2 both represent physical elements;
s23, carrying out association binding on the entity element m and the association relation T, and mapping and storing the association relation T to a database;
s24, sequentially carrying out entity identification and association binding and storage on each knowledge text data block in the knowledge text priority data set N according to the steps S21-S23.
Further, the step of performing a knowledge graph construction activity G based on the association relation T to fuse the text abstract, emotion/content tendency and attitude of the entity element, and the step of generating the knowledge graph includes:
s31, distributing corresponding entity representative nodes for each entity element on a creation page of a knowledge management platform;
S32, associating the entity representative nodes with the relevance according to the association relation T among the entity elements;
s33, configuring the entity representing node, and binding text abstract information corresponding to the entity elements, the emotion/content tendency and attitudes to the entity representing node;
and S34, after binding is completed, carrying out metadata configuration on the entity representative node, and generating the corresponding knowledge graph on the created page.
Further, the activity G is expressed as:
G=∏X K L，
wherein,
x represents entity element association, and the entity elements with association are associated and bound;
k represents text abstract extraction, and text abstract information about the entity elements is extracted by using a text abstract algorithm;
l represents text content analysis, text emotion/content analysis is carried out on text abstract information of the entity elements which are bound in an associated way by using a semantic analysis algorithm, and emotion/content tendency and attitude of the entity elements are extracted;
g represents the activity result of X, K and L, and a corresponding knowledge graph is constructed and generated.
Further, the recommending the knowledge graph to the knowledge management platform in real time includes:
Sending a warehousing notification to the knowledge management platform, and informing the knowledge management platform to audit the constructed knowledge graph in time according to the warehousing conditions of a knowledge base:
if the knowledge graph accords with the knowledge base warehousing condition, the constructed knowledge graph is saved to a database;
and if the knowledge graph does not accord with the warehouse-in conditions of the knowledge base, outputting a corresponding warehouse-in failure result, issuing a corresponding warehouse-in warning notice to a background manager, and simultaneously transmitting a corresponding warehouse-in requirement to the background manager.
Further, the managing the knowledge graph through the knowledge management platform for the user to access through the application end, and developing the search application of the knowledge graph includes:
s41, the knowledge management platform receives and stores the knowledge graph to a database, and simultaneously informs an administrator, and the administrator sends a knowledge sharing notification to an application end where a user is located;
s42, accessing the knowledge management platform by a user through an application end, and carrying out knowledge authorization on the access by the knowledge management platform;
and S43, after authorization, entering a database of the knowledge management platform, retrieving metadata, retrieving and accessing the knowledge graph.
In one aspect, a knowledge tracking system is provided, comprising:
the knowledge extraction module is used for extracting knowledge text data and caching the knowledge text data to the distributed file system;
the background service module is used for automatically identifying entity elements and association relations T among the entity elements in the knowledge text data cached in the distributed file system through a pre-deployed AI entity identification model, mapping and storing the association relations T to a database, and informing a knowledge base construction module to extract the association relations T at fixed time;
the knowledge base construction module is used for executing a knowledge graph construction activity G based on the association relation T so as to fuse the text abstract, emotion/content tendency and attitude of the entity elements, generate a knowledge graph and recommend the knowledge graph to the knowledge management platform in real time;
and the knowledge management platform is used for managing the knowledge graph and allowing a user to access the knowledge graph through an application end to develop the retrieval application of the knowledge graph.
The invention also provides an electronic device comprising a memory and a processor, the memory storing a computer program, the processor implementing the knowledge tracking method as described in any one of the above when executing the computer program.
The invention also provides a computer storage medium storing a computer program which when executed by a processor implements a knowledge tracking method as described in any of the above.
The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:
in the embodiment of the invention, through an AI entity recognition model, knowledge text data stored in a distribution manner are recognized, the entity of a knowledge element and the association relation between the entities are recognized, a knowledge graph is executed based on the association relation to construct an activity G, so that a text abstract, emotion/content tendency and attitude of the entity element are fused, a knowledge graph is generated, and then the knowledge graph is recommended to a knowledge management platform in real time for a user to access through an application end, and search application of the knowledge graph is developed; therefore, orderly and visual knowledge data management and application can be carried out by relying on the association relation of the knowledge entities, and the user can quickly know the corresponding knowledge point positions and the association by fusing the generated knowledge maps, and provide the association of the knowledge points, emotion/content tendency and attitude analysis of the knowledge text, so that the user can better know the content of the knowledge, deeply analyze and apply the knowledge maps, and the knowledge value is utilized to the maximum extent, thereby realizing intelligent management, retrieval and knowledge tracking.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a knowledge tracking method according to an embodiment of the present invention;
FIG. 2 is a detailed flowchart of a knowledge tracking method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a distributed storage flow according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a training flow of an emotion dictionary according to an embodiment of the present invention;
fig. 5 is a schematic diagram of page display of a knowledge graph according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by a person skilled in the art without creative efforts, based on the described embodiments of the present invention fall within the protection scope of the present invention.
As shown in fig. 1 and fig. 2, an embodiment of the present invention provides a knowledge tracking method, which may be implemented by an electronic device, which may be a terminal or a server, and the method includes:
s1, extracting knowledge text data and caching the knowledge text data to a distributed file system (Hadoop Distributed Filesystem, HDFS);
as an alternative embodiment, as shown in FIG. 3, the steps of extracting knowledge text data and caching the knowledge text data in a distributed file system may specifically include the following steps:
s11, extracting knowledge text data, and preprocessing and cleaning the extracted knowledge text data in batches;
in this embodiment, text extraction is performed first, and knowledge text data that needs knowledge tracking at this time is automatically extracted from a technical document, a web page or other technical text sources of an enterprise by using a text extraction manner from a data text source such as an enterprise database or a log database; for example, NLP technology is used to automatically extract key information, facts, and concepts from text documents, web pages, and other text sources.
S12, classifying the knowledge text data after batch processing according to text types to obtain a data set M consisting of a plurality of knowledge text data blocks with different text types, wherein,
M= { knowledge text data block 1, knowledge text data block 2, knowledge text data block 3.
In this embodiment, in order to improve the construction efficiency of the knowledge graph, after the knowledge text data is classified, distributed storage is performed, and specifically: data classification is performed according to text types (such as text, picture-text mixture, image or mark, and the like), knowledge text data blocks composed of the text, the picture-text mixture, the image or the mark are obtained, and then a data set M is composed of the knowledge text data blocks.
S13, orderly numbering each knowledge text data block in the data set M according to a preset knowledge text priority, and carrying out sequence priority rearrangement to obtain a knowledge text optimal ranking data set N;
in this embodiment, in order to facilitate orderly management of each text data in the data set, priority of each knowledge text data block needs to be set. The priority of the knowledge text is set according to the importance of the text, for example, the data set M contains text, image-text mixture, images and marks (such as equipment marks and symbols), and then the knowledge text data blocks of different text types in the data set M are orderly rearranged according to the priority of the images, the image-text mixture and other later priority principles to obtain the knowledge text priority data set N. The entity identification operation is then performed in this reordered order to preferentially process the prior data.
S14, traversing all storage nodes of the distributed file system HDFS, checking available storage nodes, and distributing and storing all knowledge text data blocks in the knowledge text priority data set N in the storage nodes of the distributed file system according to a priority rearrangement order;
in this embodiment, after sorting, the distributed file system HDFS is used to store the knowledge text data blocks of different text types in a distributed manner.
In this embodiment, the text extraction data includes knowledge text data of different types, such as information, text, and flags. Therefore, in order to better orderly manage the text data of each type, a distributed storage technology is adopted to orderly manage the knowledge text data of each type, so that the efficiency of subsequently executing the knowledge graph construction activity G is improved, and data errors are avoided.
And S15, the storage addresses of the knowledge text data blocks are sent to a background server.
In this embodiment, the background server records the storage addresses of the text data blocks of each knowledge, so that the pre-deployed AI entity identification model on the background server can track and process the data stored under each storage node according to the addresses.
S2, automatically identifying entity elements and association relations T among the entity elements in the knowledge text data cached in the distributed file system through a pre-deployed AI entity identification model, and mapping and storing the association relations T to a database;
as an optional embodiment, the automatically identifying, by using a pre-deployed AI entity identification model, the entity elements and the association relations T between the entity elements in the knowledge text data cached in the distributed file system, and mapping and storing the association relations T to a database (for example, a Nosql database) may specifically include the following steps:
s21, when the background server receives the storage address, notifying a pre-deployed AI entity identification model to call a knowledge text data block stored in the storage address;
s22, carrying out entity recognition on the knowledge text data block through the AI entity recognition model, automatically recognizing to obtain an entity element m in the knowledge text data block, and carrying out association recognition to obtain an association relation T between the entity elements according to the context text information of the entity element m: m1→m2; wherein m1 and m2 both represent physical elements;
In this embodiment, when the storage is found, on the background server, an AI entity recognition model pre-deployed on the background server, for example, a model tool configured based on tools such as PaddleUIE, may be invoked to perform tasks such as text recognition or image recognition in each knowledge text data block, entity extraction, relationship extraction, event extraction, etc., so as to implement recognition of entity elements and recognition of association relationships between entity elements on the stored knowledge text data; for example, entity elements such as named entities (e.g., person names, place names) in the knowledge text data can be automatically identified, extracted and marked; meanwhile, the relation among the entity elements in the text can be extracted by utilizing an entity text recognition algorithm and the like, so that the association relation T among the entity elements can be obtained. When the knowledge point search is carried out later, the association relation T can be relied on to carry out association check on the two entity elements m with the association.
In this embodiment, a semantic analysis algorithm may also be used to analyze text information between contexts where entities are located and between expressed entity elements; computer vision techniques may also be used to identify and extract information, text, and logos in the image.
S23, carrying out association binding on the entity element m and the association relation T, and mapping and storing the association relation T to a Nosql database;
in this embodiment, the Nosql database is a data read-write module with high performance and uses multiple data models, so that various types of text knowledge elements (text information associated with entity elements) on the knowledge graph can be processed efficiently, and dynamic information can be provided, so that the Nosql database is selected for storage.
S24, sequentially carrying out entity identification and association binding and storage on each knowledge text data block in the knowledge text priority data set N according to the steps S21-S23.
S3, extracting the association relation T by a timing notification knowledge base construction module, executing a knowledge graph construction activity G on the knowledge base construction module based on the association relation T to fuse text abstracts, emotion/content trends and attitudes of entity elements to generate a knowledge graph, and recommending the knowledge graph to a knowledge management platform in real time;
in this embodiment, as shown in fig. 2, when a knowledge graph is constructed, activities G and G need to be constructed, several text processing results need to be fused at the same time, a knowledge graph of a knowledge base is comprehensively constructed, and the knowledge graph can be displayed on a knowledge management platform;
Wherein the activity G is represented as:
G=∏X K L，
wherein,
x represents entity element association, and the entity elements with association are associated and bound;
k represents text abstract extraction, and text abstract information about the entity elements is extracted by using a text abstract algorithm;
l represents text content analysis, text emotion/content analysis is carried out on text abstract information of the entity elements which are bound in an associated way by using a semantic analysis algorithm, and emotion/content tendency and attitude of the entity elements are extracted;
g represents the activity result of X, K and L, and a corresponding knowledge graph is constructed and generated.
In this embodiment, the extraction of the X relationship aims at identifying the relationship between entities in the text, and constructing a knowledge graph to represent the relevance of the knowledge.
In this embodiment, when extracting the text abstract: and automatically generating a text abstract by using a text abstract algorithm, and refining key information of the document. And performing abstract extraction on text content where the entity elements are located by a text abstract algorithm such as TextRank to extract text abstract information representing the entity elements, such as keywords, key abstract sentence segments and the like which can represent and adjective the entity elements.
In this embodiment, when text emotion/content analysis is performed: by analyzing the emotion of the text, the emotion tendency and attitude of the text are helped to be understood. And analyzing text emotion, namely performing text emotion/content analysis on the text abstract information of the entity element by adopting an emotion analysis method based on an emotion dictionary, an emotion analysis method based on traditional machine learning, an emotion analysis method based on deep learning and the like, and extracting emotion/content tendency and attitude of the entity element. Through emotion/content trend and attitude of the entity elements, the text trend represented by the entity elements corresponding to the current knowledge point can be simply and quickly known.
As shown in fig. 4, for example, a emotion dictionary method can be adopted, emotion polarity division under different granularity can be realized according to emotion polarities of emotion words provided by different pre-trained emotion dictionaries, and content attitudes of knowledge points corresponding to entity elements can be easily analyzed and understood by reflecting unstructured features of texts, so that emotion classification effects are more accurate.
In this embodiment, when training the emotion dictionary, knowledge text is first input and preprocessed (e.g., word segmentation), and then the emotion dictionary is trained based on the loaded emotion words such as positive words, negative words, degree adverbs, negative words, positive words, exclamation words, and the like, and emotion is output according to a preset judgment rule.
As an optional embodiment, the performing a knowledge graph construction activity G based on the association relation T to fuse the text abstract, emotion/content tendency and attitude of the entity element to generate a knowledge graph may specifically include the following steps:
s31, distributing corresponding entity representative nodes for each entity element on a creation page of a knowledge management platform;
s32, associating the entity representative nodes with the relevance according to the association relation T among the entity elements;
s33, configuring the entity representing node, and binding text abstract information corresponding to the entity elements, the emotion/content tendency and attitudes to the entity representing node;
and S34, after binding is completed, carrying out metadata configuration on the entity representative node, and generating the corresponding knowledge graph on the created page.
In this embodiment, knowledge base fusion construction is performed through multi-azimuth entity identification, relation extraction and semantic analysis of text content to obtain a knowledge graph, and the correlation among entity elements and key information and content trend of each entity element are displayed through the knowledge graph, so that a user intuitively knows the correlation of each entity element and the key information and knowledge content trend of the knowledge point where the entity element is located according to the knowledge point, and a subsequent enterprise user can log in a knowledge management platform through an application end to access a background server to search and use the knowledge graph and check information contained in the entity element where the required knowledge point is located. Through the device. The knowledge graph is searched, the relevance of the entity elements where the corresponding knowledge points are located and the corresponding key information can be queried, and the core content and attitude of the knowledge points can be rapidly mastered.
In this embodiment, after the knowledge base construction module constructs the corresponding knowledge spectrum, the knowledge spectrum may be recommended to the knowledge management platform in real time, and specifically may include the following steps:
sending a warehousing notification to the knowledge management platform, and informing the knowledge management platform to audit the constructed knowledge graph in time according to the warehousing conditions of a knowledge base:
if the knowledge graph accords with the knowledge base warehousing condition, the constructed knowledge graph is saved to a Nosql database;
and if the knowledge graph does not accord with the warehouse-in conditions of the knowledge base, outputting a corresponding warehouse-in failure notification, and simultaneously sending the reason of the warehouse-in failure and the corresponding warehouse-in requirement to a background manager.
In this embodiment, the knowledge management platform may be set up by a server of a user such as each enterprise, and may use knowledge management software adopted on a background server to manage and operate a knowledge graph.
In this embodiment, the knowledge base storage conditions configured on the knowledge management platform include requirements of enterprises on different knowledge points or knowledge elements, and are specifically set by a background administrator. For example, some knowledge entity elements need to be discarded, when the knowledge map is found to have the entity elements, the knowledge map is informed that the knowledge base storage conditions are not met, a corresponding storage failure notification is output, and meanwhile, the reason of the storage failure and the corresponding storage requirement are sent to a background manager for later rectification.
And S4, managing the knowledge graph through the knowledge management platform, and enabling a user to access through an application end to develop the retrieval application of the knowledge graph.
In this embodiment, the knowledge graph is managed by the knowledge management platform, and is accessed by a user through an application end, so as to develop a search application for the knowledge graph, which specifically may include the following steps:
s41, the knowledge management platform receives and stores the knowledge graph to a database, and simultaneously informs an administrator, and the administrator sends a knowledge sharing notification to an application end where a user is located;
s42, accessing the knowledge management platform by a user through an application end, and carrying out knowledge authorization on the access by the knowledge management platform;
and S43, after authorization, entering a database of the knowledge management platform, retrieving metadata, retrieving and accessing the knowledge graph.
In this embodiment, a knowledge graph tool, such as INCEpTION, neo J, may be installed on the knowledge management platform of the enterprise, a page may be created on the knowledge management platform, a knowledge graph between the entity elements may be constructed according to the association relationship T between the entity elements, and finally a page of a knowledge graph as shown in fig. 5 may be obtained on the created page.
In this embodiment, entity representative nodes may be dragged and created on a creation page through a small tool, each entity representative node represents an application node on a knowledge graph, the application node will represent an entity element, and the application nodes represented by the entity elements with relevance may be connected in an associated manner through an association relationship between the entity elements, meanwhile, attribute configuration is performed for text abstract information and content tendency of the entity elements represented by the entity representative nodes, node attribute parameter configuration is performed for each application node, the text abstract information, content tendency and attitude are configured and bound on the application node, so as to construct a current knowledge graph, and the entity identification may be referred to a knowledge graph of a social network shown in fig. 5, so as to construct node links, and generate a knowledge graph of the social network.
In this embodiment, the enterprise user may log into the knowledge management platform through an APP of an application end, such as a smart phone. After authorized access of the knowledge management platform is obtained, the knowledge graph on the knowledge management platform is accessed, and data retrieval and knowledge graph retrieval application are performed.
In this embodiment, for the authority, authority verification and management of the knowledge management platform on the access user, the authority configuration and authority verification scheme of the existing application system on the access user may be referred to.
In this embodiment, for application of the knowledge graph, the extracted knowledge is represented as entities and relationships, so as to facilitate intelligent retrieval and association.
In this embodiment, a metadata management function may be added: metadata is added to the knowledge entity elements, including tags, keywords, time stamps, etc., to better organize and retrieve the knowledge.
According to the knowledge tracking method provided by the embodiment of the invention, valuable insights can be extracted from a large amount of information through automatic information extraction, knowledge management, intelligent retrieval and application, and various application scenes are supported, including intelligent retrieval, personalized recommendation, decision support, innovation and research, education, training and the like.
In the implementation process, factors such as security, performance, privacy protection and the like need to be comprehensively considered so as to ensure the stability and usability of the system.
In this embodiment, a general application scenario is briefly described:
and (3) intelligent searching: providing an intelligent searching function by using a natural language processing technology, so that a user can query a knowledge base in natural language;
Personalized recommendation: providing personalized knowledge recommendations based on historical queries and interests of the user to help the user discover relevant information; or recommending proper learning materials according to the learning needs and interests of students;
knowledge management training: the knowledge management training is provided for enterprises, and staff is helped to better utilize knowledge resources.
In summary, the knowledge tracking method provided by the embodiment of the invention has at least the following beneficial effects:
in the embodiment of the invention, through an AI entity recognition model, knowledge text data stored in a distribution manner are recognized, the entity of a knowledge element and the association relation between the entities are recognized, a knowledge graph is executed based on the association relation to construct an activity G, so that a text abstract, emotion/content tendency and attitude of the entity element are fused, a knowledge graph is generated, and then the knowledge graph is recommended to a knowledge management platform in real time for a user to access through an application end, and search application of the knowledge graph is developed; therefore, orderly and visual knowledge data management and application can be carried out by relying on the association relation of the knowledge entities, and the user can quickly know the corresponding knowledge point positions and the association by fusing the generated knowledge maps, and provide the association of the knowledge points, emotion/content tendency and attitude analysis of the knowledge text, so that the user can better know the content of the knowledge, deeply analyze and apply the knowledge maps, and the knowledge value is utilized to the maximum extent, thereby realizing intelligent management, retrieval and knowledge tracking.
The present invention also provides a specific embodiment of a knowledge tracking system, and since the knowledge tracking system provided by the present invention corresponds to the specific embodiment of the knowledge tracking method, the knowledge tracking system can achieve the purpose of the present invention by executing the steps of the flow in the specific embodiment of the method, so the explanation in the specific embodiment of the knowledge tracking method is also applicable to the specific embodiment of the knowledge tracking system provided by the present invention, and will not be repeated in the following specific embodiment of the present invention.
The embodiment of the invention also provides a knowledge tracking system, which comprises:
the knowledge extraction module is used for extracting knowledge text data and caching the knowledge text data to the distributed file system;
the background service module is used for automatically identifying entity elements and association relations T among the entity elements in the knowledge text data cached in the distributed file system through a pre-deployed AI entity identification model, mapping and storing the association relations T to a database, and informing a knowledge base construction module to extract the association relations T at fixed time;
the knowledge base construction module is used for executing a knowledge graph construction activity G based on the association relation T so as to fuse the text abstract, emotion/content tendency and attitude of the entity elements, generate a knowledge graph and recommend the knowledge graph to the knowledge management platform in real time;
And the knowledge management platform is used for managing the knowledge graph and allowing a user to access the knowledge graph through an application end to develop the retrieval application of the knowledge graph.
According to the knowledge tracking system provided by the embodiment of the invention, through an AI entity identification model, the identification of the entity of the knowledge element and the association relation between the entities is carried out on the knowledge text data stored in the distribution, and the knowledge graph construction activity G is executed based on the association relation so as to fuse the text abstract, emotion/content tendency and attitude of the entity element, so that a knowledge graph is generated, and then the knowledge graph is recommended to a knowledge management platform in real time for a user to access through an application end, so that the search application of the knowledge graph is developed; therefore, orderly and visual knowledge data management and application can be carried out by relying on the association relation of the knowledge entities, and the user can quickly know the corresponding knowledge point positions and the association by fusing the generated knowledge maps, and provide the association of the knowledge points, emotion/content tendency and attitude analysis of the knowledge text, so that the user can better know the content of the knowledge, deeply analyze and apply the knowledge maps, and the knowledge value is utilized to the maximum extent, thereby realizing intelligent management, retrieval and knowledge tracking.
Fig. 6 is a schematic structural diagram of an electronic device 600 according to an embodiment of the present invention, where the electronic device 600 may have a relatively large difference due to different configurations or performances, and may include one or more processors (central processing units, CPU) 601 and one or more memories 602, where at least one instruction is stored in the memories 602, and the at least one instruction is loaded and executed by the processors 601 to implement the knowledge tracking method described above.
In an exemplary embodiment, a computer readable storage medium, such as a memory comprising instructions executable by a processor in a terminal to perform the above-described knowledge tracking method, is also provided. For example, the computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program for instructing relevant hardware, where the program may be stored in a computer readable storage medium, and the storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
References in the specification to "one embodiment," "an example embodiment," "some embodiments," etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the relevant art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.
The invention is intended to cover any alternatives, modifications, equivalents, and variations that fall within the spirit and scope of the invention. In the following description of preferred embodiments of the invention, specific details are set forth in order to provide a thorough understanding of the invention, and the invention will be fully understood to those skilled in the art without such details. In other instances, well-known methods, procedures, flows, components, circuits, and the like have not been described in detail so as not to unnecessarily obscure aspects of the present invention.
Those of ordinary skill in the art will appreciate that all or a portion of the steps in implementing the methods of the embodiments described above may be implemented by a program that instructs associated hardware, and the program may be stored on a computer readable storage medium, such as: ROM/RAM, magnetic disks, optical disks, etc.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.

Claims (5)

1. A method of knowledge tracking, comprising:
s1, extracting knowledge text data and caching the knowledge text data to a distributed file system;
S2, automatically identifying entity elements and association relations T among the entity elements in the knowledge text data cached in the distributed file system through a pre-deployed AI entity identification model, and mapping and storing the association relations T to a database;
s3, extracting the association relation T by a timing notification knowledge base construction module, executing a knowledge graph construction activity G on the knowledge base construction module based on the association relation T to fuse text abstracts, emotion/content trends and attitudes of entity elements to generate a knowledge graph, and recommending the knowledge graph to a knowledge management platform in real time;
s4, managing the knowledge graph through the knowledge management platform, and enabling a user to access through an application end to develop retrieval application of the knowledge graph;
wherein the extracting knowledge text data and caching to the distributed file system comprises:
s11, extracting knowledge text data, and preprocessing and cleaning the extracted knowledge text data in batches;
s12, classifying the knowledge text data after batch processing according to text types to obtain a data set M consisting of a plurality of knowledge text data blocks with different text types, wherein,
M= { knowledge text data block 1, knowledge text data block 2, knowledge text data block 3.
S13, orderly numbering each knowledge text data block in the data set M according to a preset knowledge text priority, and carrying out sequence priority rearrangement to obtain a knowledge text optimal ranking data set N;
s14, traversing all storage nodes of a distributed file system, checking available storage nodes, and distributing and storing all knowledge text data blocks in the knowledge text priority data set N in the storage nodes of the distributed file system according to a priority rearrangement order;
s15, the storage addresses of the knowledge text data blocks are sent to a background server;
the step of executing the knowledge graph construction activity G based on the association relation T to fuse the text abstract, emotion/content trend and attitude of the entity element, and the step of generating the knowledge graph comprises the following steps:
s31, distributing corresponding entity representative nodes for each entity element on a creation page of a knowledge management platform;
s32, associating the entity representative nodes with the relevance according to the association relation T among the entity elements;
s33, configuring the entity representing node, and binding text abstract information corresponding to the entity elements, the emotion/content tendency and attitudes to the entity representing node;
S34, after binding is completed, carrying out metadata configuration on the entity representative node, and generating the corresponding knowledge graph on the created page;
wherein the activity G is represented as:
G=∏X K L，
wherein,
x represents entity element association, and the entity elements with association are associated and bound;
k represents text abstract extraction, and text abstract information about the entity elements is extracted by using a text abstract algorithm;
l represents text content analysis, text emotion/content analysis is carried out on text abstract information of the entity elements which are bound in an associated way by using a semantic analysis algorithm, and emotion/content tendency and attitude of the entity elements are extracted;
g represents the activity result of X, K and L, and a corresponding knowledge graph is constructed and generated.
2. The knowledge tracking method according to claim 1, wherein the automatically identifying entity elements and association relations T between entity elements in the knowledge text data cached in the distributed file system by a pre-deployed AI entity identification model, and storing the association relations T in a database includes:
s21, when the background server receives the storage address, notifying a pre-deployed AI entity identification model to call a knowledge text data block stored in the storage address;
S22, carrying out entity recognition on the knowledge text data block through the AI entity recognition model, automatically recognizing to obtain an entity element m in the knowledge text data block, and carrying out association recognition to obtain an association relation T between the entity elements according to the context text information of the entity element m: m1→m2; wherein m1 and m2 both represent physical elements;
s23, carrying out association binding on the entity element m and the association relation T, and mapping and storing the association relation T to a database;
s24, sequentially carrying out entity identification and association binding and storage on each knowledge text data block in the knowledge text priority data set N according to the steps S21-S23.
3. The knowledge tracking method of claim 1, wherein recommending the knowledge-graph to a knowledge management platform in real time comprises:
sending a warehousing notification to the knowledge management platform, and informing the knowledge management platform to audit the constructed knowledge graph in time according to the warehousing conditions of a knowledge base:
if the knowledge graph accords with the knowledge base warehousing condition, the constructed knowledge graph is saved to a database;
and if the knowledge graph does not accord with the warehouse-in conditions of the knowledge base, outputting a corresponding warehouse-in failure result, issuing a corresponding warehouse-in warning notice to a background manager, and simultaneously transmitting a corresponding warehouse-in requirement to the background manager.
4. The knowledge tracking method according to claim 1, wherein the managing the knowledge graph by the knowledge management platform for a user to access by an application end, and developing a search application for the knowledge graph comprises:
s41, the knowledge management platform receives and stores the knowledge graph to a database, and simultaneously informs an administrator, and the administrator sends a knowledge sharing notification to an application end where a user is located;
s42, accessing the knowledge management platform by a user through an application end, and carrying out knowledge authorization on the access by the knowledge management platform;
and S43, after authorization, entering a database of the knowledge management platform, retrieving metadata, retrieving and accessing the knowledge graph.
5. A knowledge tracking system, comprising:
the knowledge extraction module is used for extracting knowledge text data and caching the knowledge text data to the distributed file system;
the background service module is used for automatically identifying entity elements and association relations T among the entity elements in the knowledge text data cached in the distributed file system through a pre-deployed AI entity identification model, mapping and storing the association relations T to a database, and informing a knowledge base construction module to extract the association relations T at fixed time;
The knowledge base construction module is used for executing a knowledge graph construction activity G based on the association relation T so as to fuse the text abstract, emotion/content tendency and attitude of the entity elements, generate a knowledge graph and recommend the knowledge graph to the knowledge management platform in real time;
the knowledge management platform is used for managing the knowledge graph and allowing a user to access the knowledge graph through an application end to develop search application of the knowledge graph;
wherein the extracting knowledge text data and caching to the distributed file system comprises:
s11, extracting knowledge text data, and preprocessing and cleaning the extracted knowledge text data in batches;
s12, classifying the knowledge text data after batch processing according to text types to obtain a data set M consisting of a plurality of knowledge text data blocks with different text types, wherein,
m= { knowledge text data block 1, knowledge text data block 2, knowledge text data block 3.
S13, orderly numbering each knowledge text data block in the data set M according to a preset knowledge text priority, and carrying out sequence priority rearrangement to obtain a knowledge text optimal ranking data set N;
s14, traversing all storage nodes of a distributed file system, checking available storage nodes, and distributing and storing all knowledge text data blocks in the knowledge text priority data set N in the storage nodes of the distributed file system according to a priority rearrangement order;
S15, the storage addresses of the knowledge text data blocks are sent to a background server;
the step of executing the knowledge graph construction activity G based on the association relation T to fuse the text abstract, emotion/content trend and attitude of the entity element, and the step of generating the knowledge graph comprises the following steps:
s31, distributing corresponding entity representative nodes for each entity element on a creation page of a knowledge management platform;
s32, associating the entity representative nodes with the relevance according to the association relation T among the entity elements;
s33, configuring the entity representing node, and binding text abstract information corresponding to the entity elements, the emotion/content tendency and attitudes to the entity representing node;
s34, after binding is completed, carrying out metadata configuration on the entity representative node, and generating the corresponding knowledge graph on the created page;
wherein the activity G is represented as:
G=∏X K L，
wherein,
x represents entity element association, and the entity elements with association are associated and bound;
k represents text abstract extraction, and text abstract information about the entity elements is extracted by using a text abstract algorithm;
L represents text content analysis, text emotion/content analysis is carried out on text abstract information of the entity elements which are bound in an associated way by using a semantic analysis algorithm, and emotion/content tendency and attitude of the entity elements are extracted;
g represents the activity result of X, K and L, and a corresponding knowledge graph is constructed and generated.
CN202311385302.0A 2023-10-25 2023-10-25 Knowledge tracking method and system Active CN117112809B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311385302.0A CN117112809B (en) 2023-10-25 2023-10-25 Knowledge tracking method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311385302.0A CN117112809B (en) 2023-10-25 2023-10-25 Knowledge tracking method and system

Publications (2)

Publication Number Publication Date
CN117112809A CN117112809A (en) 2023-11-24
CN117112809B true CN117112809B (en) 2024-01-26

Family

ID=88795207

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311385302.0A Active CN117112809B (en) 2023-10-25 2023-10-25 Knowledge tracking method and system

Country Status (1)

Country Link
CN (1) CN117112809B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110162639A (en) * 2019-04-16 2019-08-23 深圳壹账通智能科技有限公司 Knowledge figure knows the method, apparatus, equipment and storage medium of meaning
CN114595686A (en) * 2022-03-11 2022-06-07 北京百度网讯科技有限公司 Knowledge extraction method, and training method and device of knowledge extraction model
CN114706948A (en) * 2022-03-28 2022-07-05 北京明略昭辉科技有限公司 News processing method and device, storage medium and electronic equipment
CN114741533A (en) * 2022-04-21 2022-07-12 北京明略昭辉科技有限公司 Method and device for constructing map, electronic equipment and storage medium
CN115169283A (en) * 2022-06-28 2022-10-11 中国人民解放军国防科技大学 Planning plan distributed rapid supervision method, system, equipment and storage medium
CN115544235A (en) * 2022-11-09 2022-12-30 国网甘肃省电力公司经济技术研究院 Power grid planning intelligent question-answering system based on text parsing
CN116881476A (en) * 2023-08-01 2023-10-13 空间视创(重庆)科技股份有限公司 Knowledge graph construction method, platform and computer storage medium

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110162639A (en) * 2019-04-16 2019-08-23 深圳壹账通智能科技有限公司 Knowledge figure knows the method, apparatus, equipment and storage medium of meaning
CN114595686A (en) * 2022-03-11 2022-06-07 北京百度网讯科技有限公司 Knowledge extraction method, and training method and device of knowledge extraction model
CN114706948A (en) * 2022-03-28 2022-07-05 北京明略昭辉科技有限公司 News processing method and device, storage medium and electronic equipment
CN114741533A (en) * 2022-04-21 2022-07-12 北京明略昭辉科技有限公司 Method and device for constructing map, electronic equipment and storage medium
CN115169283A (en) * 2022-06-28 2022-10-11 中国人民解放军国防科技大学 Planning plan distributed rapid supervision method, system, equipment and storage medium
CN115544235A (en) * 2022-11-09 2022-12-30 国网甘肃省电力公司经济技术研究院 Power grid planning intelligent question-answering system based on text parsing
CN116881476A (en) * 2023-08-01 2023-10-13 空间视创(重庆)科技股份有限公司 Knowledge graph construction method, platform and computer storage medium

Also Published As

Publication number Publication date
CN117112809A (en) 2023-11-24

Similar Documents

Publication Publication Date Title
US11645317B2 (en) Recommending topic clusters for unstructured text documents
US7912816B2 (en) Adaptive archive data management
CN112711937A (en) Template recommendation method, device, equipment and storage medium
JP2013541793A (en) Multi-mode search query input method
US20220107980A1 (en) Providing an object-based response to a natural language query
CN111723256A (en) Government affair user portrait construction method and system based on information resource library
US9886479B2 (en) Managing credibility for a question answering system
CN115687647A (en) Notarization document generation method and device, electronic equipment and storage medium
CN110188291B (en) Document processing based on proxy log
CN115374781A (en) Text data information mining method, device and equipment
CN117149988B (en) Data management processing method and system based on education digitization
CN111383072A (en) User credit scoring method, storage medium and server
CN117112809B (en) Knowledge tracking method and system
Fu et al. Enhancing Semantic Search of Crowdsourcing IT Services using Knowledge Graph.
CN113961811B (en) Event map-based conversation recommendation method, device, equipment and medium
US20230142351A1 (en) Methods and systems for searching and retrieving information
Hong et al. An efficient tag recommendation method using topic modeling approaches
Kamath et al. Semantic Similarity Based Context-Aware Web Service Discovery Using NLP Techniques.
CN112418260A (en) Model training method, information prompting method, device, equipment and medium
WO2019246252A1 (en) Systems and methods for identifying and linking events in structured proceedings
CN111324711B (en) Post timeliness judging method and device and storage medium
CN111949781B (en) Intelligent interaction method and device based on natural sentence syntactic analysis
US20240112103A1 (en) System and method for recommending next best action to create new employment pathways for job applicants
Rodrigues et al. Using textual CBR for e-learning content categorization and retrieval
Zhao et al. Cluster analysis and its application in teaching resources of university curriculum: A personalized method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant