64 results about "Knowledge classification" patented technology

The invention provides a building method of a knowledge map based on a vertical field. The method comprises the following steps of (1) extracting the word realization of classes of an on-line encyclopedia and the hyponymy between classes; (2) merging the field knowledge information, defining the data attribute and the relationship attribute of the field, and further setting the statute on the definition domain and the value domain of the attributes; (3) studying an entity layer, i.e., extracting an entity and filling the attribute value of the entity; performing mass processing on structurized and semi-structurized data by D2R or data collecting tools; and for non-structurized text data, defining the classes and the attributes of the upper layer body and the relationship between the classes and the attributes, and recognizing examples according to the relationship between the classes and the attributes. The method has the advantages that by using the method, the built knowledge classification of the vertical field knowledge map is clear; the self study and the automatic expansion of the knowledge map are realized; and the key effects are achieved on the information retrieval and semantic analysis of the vertical field.

The invention discloses a fusion knowledge base processing method and a device and a knowledge base management system. The fusion knowledge base processing method comprises the steps of extracting a knowledge classification catalogue from a common problem FAQ library and an ontology knowledge base, wherein the ontology knowledge base comprises a knowledge ontology and a ontology relation, based on the knowledge ontology, ontology relation and the knowledge classification catalogue, mount the contents of the FAQ library into the ontology knowledge base, and generating a fusion knowledge base. The fusion knowledge base processing method solves the problem that in the prior art the performance of the FAQ system can hardly be enhanced because the retrieval question answering system, the question answering system based on the knowledge base and the community cooperative question answering system have many defects in the practical application.

A numeric, hierarchical knowledge classification system, (e.g., Dewey Decimal Classification (DDC)) of concepts, is used for universal annotation. For example, “game” (DDC=799) relates to hunting and “game” (DDC=794.105), to chess. The universal annotation may be combined with human-aided machine translation, so individuals can tag their documents in the source language, simultaneously specifying meanings in the source and target languages, exercising control over what is said in the target language. Multilingual communication, by e-mails, forums, blogs, Wikipedia and the like may proceed, with each participant working in his tongue and tagging his writing with universal concepts. Moreover, a Universal System of Expression (USE) is introduced, based on the universal concepts. USE documents may be generated as byproducts of human-aided machine translations between any two natural languages that recognize USE, for fully automatic translation to other languages, on demand, and libraries of USE documents may be formed.

The invention belongs to the technical field of computer application and relates to a knowledge management oriented user-defined knowledge classification method, aiming to provide the word-matching based user-defined knowledge classification method for solving the problems about realization of user-defined knowledge classification in the absence of text collection labeling in a knowledge management system. The problem of training the labeling of a text collection is bypassed, user-defined keywords can be expanded by the aid of 'synonym word stock', the keywords can be rapidly matched by a Wu-Manber multiple-pattern matching algorithm, and classification categories can be judged through the hit rate of the keywords in a document. The knowledge management oriented user-defined knowledge classification method has the advantages of adapting to the conditions lack of text labeling, and speed is high.

The invention discloses an SOLR-based knowledge management system, and relates to the technical field of software. The SOLR-based knowledge management system is a BS-structured knowledge base system,and deployed at the server side; a client side accesses the knowledge management system at the server side through a browser; the knowledge management system comprises four modules including a knowledge creating module, a knowledge presentation and application module, a knowledge maintenance and management module and a knowledge statistical function module; the knowledge creating module includes knowledge input, knowledge check and knowledge publishing; the knowledge presentation and application module is responsible for knowledge check, knowledge searching and knowledge circle; the knowledgemaintenance and management module is responsible for knowledge classification maintenance, knowledge publishing rule maintenance, knowledge maintenance and knowledge filing; and the knowledge statistical function module is responsible for performing knowledge statistics, knowledge classification, knowledge statistics of tax authorities, and generating related reports. According to the SOLR-based knowledge management system disclosed by the invention, the overall management level is increased; enterprise knowledge reserve is increased; organizational memory loss is avoided; a knowledge sharingchannel is provided for staffs; knowledge circulation in an organization is promoted; and the efficiency of obtaining knowledge by members is increased.

The invention discloses a knowledge system modeling method based on knowledge ontology. The modeling method includes the following steps that a knowledge acquisition module acquires knowledge in the technical field and sends the knowledge to a knowledge classification module by providing a knowledge template for a user according to the characteristics of each technical field; the knowledge of all the technical fields is uniformly classified by the knowledge classification module, and the classified knowledge is sent to a knowledge representation module; the knowledge of the same category is represented in the mode capable of being recognized by a computer and sent to a knowledge packaging module by the knowledge representation module; all the knowledge of all the technical fields is uniformly packaged by the knowledge packaging module based on the knowledge ontology according to uniform packaging specifications to obtain base elements of knowledge and knowledge components, and the base elements of knowledge and the knowledge components are sent to a knowledge file generation module; the base elements of knowledge and the knowledge components are used for generating knowledge files capable of being recognized by a multi-mode inference engine according to the owl specifications by the knowledge file generation module.

The invention discloses a 95598 knowledge base system. The 95598 knowledge base system comprises a knowledge sorting module, a knowledge consulting module, a knowledge consulting query module, a knowledge collecting module, a knowledge collecting query module, a knowledge maintaining module, a knowledge retrieving module and an index re-establishing module. The knowledge consulting module is used for generating work orders of a corresponding power supply company unit according to the power supply company unit selected by a user, issuing the work orders to the corresponding unit, and configuring a time limit. The knowledge sorting module maintains and displays knowledge in a tree catalog structure, and knowledge is divided into services of new addition, updating and the like of knowledge catalogs and catalog hierarchical relationships according to services. The knowledge collecting module is used for screening and perfecting knowledge consulting information and typing the information in the system according to knowledge types and the conditions of the applicable unit. A knowledge element neutral network is established, a knowledge structure professionally divided can be provided for related personnel, and the knowledge retrieval effective rate and accuracy are improved.

The invention discloses an induction and deduction knowledge unconsciousness seal-learning method based on the 6w rule and relates to the technical field of information. The induction and deduction knowledge unconsciousness seal-learning method based on the 6w rule comprises the steps that according to marks in knowledge, the knowledge is classified, the marks of the knowledge of the same type are classified, upward and downward learning is conducted according to existing knowledge in a complete knowledge structure through a sublevel mark deduction and primary-level mark induction method, so that new knowledge is obtained, the new knowledge is verified, and the correct knowledge is stored in a knowledge base. By the adoption of the induction and deduction knowledge unconsciousness seal-learning method based on the 6w rule, a knowledge system is effectively expanded, a knowledge base is established, and the practical application capacity of the knowledge is improved.

Embodiments of the present invention address deficiencies of the art in respect to expert modeling in a KM system and provide method, system and computer program product for organically ranked knowledge and categorization for a KM system. In one embodiment of the invention, a method for organically ranked knowledge and categorization in a KM system can be provided. The method can include bookmarking answer content for a first end user of the knowledge management system, suggesting a set of categories previously associated with the answer content by other end users of the knowledge management system, and categorizing the bookmarked answer content with a category selected from the set of categories.

The invention discloses a method for automatically classifying, obtaining and storing complex knowledge of a high-end device. The method comprises: an automatic complex knowledge classification method of performing induction and reorganization on knowledge resources from the following three dimensions of the high-end device: a life cycle dimension, a knowledge manifestation pattern dimension and a knowledge theme dimension, and automatically classifying the knowledge resources by using a naive Bayes classifier; a complex knowledge obtaining method of obtaining a template according to complex knowledge based on a meta-knowledge model and obtaining the complex knowledge resources through semi-automatic obtaining technology based on the obtained template; and an automatic complex knowledge storage method of dividing the complex knowledge resources from the physics through a series of automatic division rules, compressing key information and storing the same in different storage spaces in a distributed manner. The method disclosed by the invention covers the automatic complex knowledge classification method, the complex knowledge obtaining method and the automatic complex knowledge storage method, and provides foundation and support for the high-end device manufacturers to use the complex knowledge resources.

The invention relates to the technical field of computer networks, and discloses a visual collection system and method based on network resources, and the system comprises a configuration module, a collection module, a classification indexing module, and a data storage module. And the configuration module comprises a configuration program which can call a browser, so that the browser renders and presents a page to be configured, nodes of the page to be configured can be conveniently clicked, page data extraction and processing rules are generated, and a visual operation environment is realized. The collection module comprises a collection program, is used for accessing a to-be-collected page and extracting data according to a page data extraction and processing rule; the classification indexing module comprises a classification unit; wherein the data storage module can establish a knowledge classification system of a page to be collected according to content characteristics of the pageto be collected, classify collected data according to the knowledge classification system and index keywords according to texts in the collected data, and the data storage module comprises a storagedatabase which can store the collected data.

The invention relates to the technical field of cloud computing and especially relates to a method for managing an operation and maintenance knowledge base of a cloud platform based on role control and external interaction. The method provided by the invention comprises the following steps: establishing role management for an operation and maintenance knowledge base system; dividing the roles of user operation class and cloud component class, thereby being convenient for right control and operation and maintenance knowledge classification; and supplying an API service based on the user role control, including establishing, deleting, updating and inquiring a knowledge base data item containing the component class. According to the invention, the cloud platform can be excellently accessed through the API service based on the user role control; the operation and maintenance knowledge item for the cloud component classification can be automatically and semi-automatically established; the cloud platform can be assisted in automatically and semi-automatically solving the problem in the operation and maintenance process; the cloud platform can be endowed with the capacity of sharing and accumulating operation and maintenance knowledge; and the efficiency and quality of operation and maintenance can be indirectly promoted.

Embodiments of the present invention address deficiencies of the art in respect to expert modeling in a KM system and provide method, system and computer program product for organically ranked knowledge and categorization for a KM system. In one embodiment of the invention, a method for organically ranked knowledge and categorization in a KM system can be provided. The method can include bookmarking answer content for a first end user of the knowledge management system, suggesting a set of categories previously associated with the answer content by other end users of the knowledge management system, and categorizing the bookmarked answer content with a category selected from the set of categories.

The invention relates to a knowledge graph dynamic maintenance method, terminal equipment and a storage medium. The method comprises the steps of S1, collecting service data; S2, loading a thesaurus file, and extracting knowledge from the business data according to the thesaurus file; S3, converting the thesaurus into concepts in the ontology and relationships among the concepts; S4, classifying the extracted knowledge into specific concepts in a thesaurus; S5: according to each knowledge and each concept associated with the knowledge in the thesaurus, constructing a new knowledge ontology anda relationship between the knowledge ontologies; S6, combining the new knowledge ontology and relationship with the existing knowledge ontology and relationship; S7, updating the knowledge graph andthe thesaurus according to the combined knowledge ontology; and S8, for the newly added concepts, according to the classification principle of the thesaurus, according to semantic analysis, discovering upper thesaurus of the newly added concepts and corresponding concept categories, and adding new concept entities. According to the invention, the problems of hysteresis, inflexible expansion and low efficiency in updating the knowledge graph are solved.

The invention discloses a scientific and technological resource personalized active pushing method, and the method comprises the steps: building knowledge classification information, and building a knowledge-expert database based on the knowledge classification information; obtaining questioning information of the user; analyzing the questioning information to obtain category information of the questioning information; acquiring expert information corresponding to the category information based on the knowledge-expert database; and pushing the questioning information to an expert account corresponding to the expert information. First, expert libraries of various technical fields are established. The questions are pushed to professional answering experts to be timely and professionally answered in an active pushing mode according to the questions proposed by the user, so that the answering efficiency of the questions is greatly improved, the accuracy of answers to the questions is effectively improved, and the user experience is improved.

The invention discloses a maintenance application automatic ticket forming method based on natural language analysis. The method comprises the following steps: step 100, reading a adjustment and control maintenance historical record as a sample, and carrying out the cleaning and filtering of sample data, so as to complete the correlation correspondence of a maintenance application record and a maintenance operation record; step 200, carrying out multi-dimensional analysis on the sample data by utilizing a natural language processing technology to form a adjustment and control domain corpus, and simplifying an analysis processing process of a subsequent maintenance application; step 300, learning related knowledge of the power system, classifying and completing structured storage and input,and establishing a adjustment language feature point model set; step 400, based on the adjustment and control language feature point model set, carrying out quick analysis processing on the maintenance application so as to realize packaging and grouping of the maintenance application; and step 500, checking the integrity of safety measures of maintenance work, deriving an operation scheme and anoperation command, and automatically generating an operation ticket. According to the invention, the operation order can be automatically generated according to the maintenance application.

The invention provides a knowledge representation method for a cross-media knowledge reasoning task, and the method comprises the steps: extracting the RDF triad information of a cross-media knowledgegraph, and enabling the RDF triad data of the cross-media knowledge graph to be represented as an initial low-dimensional vector; training vector representation between positive and negative triad samples by utilizing a maximum interval cost function, mining similarity (or difference) between the positive and negative triad samples, and adding the similarity (or the difference) into the maximum interval cost function, so that the capability of recognizing similar entities by model knowledge reasoning is improved. According to the method, knowledge representation and knowledge reasoning can becarried out on the cross-media knowledge graph triples constructed on the basis of the RDF, entity linking and knowledge classification are carried out by utilizing the knowledge reasoning model learned by the method, and the accuracy of link prediction and triad classification in the cross-media knowledge graph can be improved.

The invention discloses a knowledge classification system, wherein a transmitting unit of a template module is used for transmitting a knowledge file; a selecting module is used for receiving a selecting signal to select one classification item and a corresponding code thereof from the classification contents in the template module; a generating module is used for generating a classification reference interface corresponding to the knowledge file according to the correspondingly selected classification items and codes thereof, and the classification contents of the classification reference interface are respectively corresponding to the selected classification items, the codes thereof and a code cluster formed by the selected codes. The knowledge classification system has the advantages of enabling a file reader to clearly master the knowledge field of the archive file, and know the source of an archive file and the department of the personnel uploading the archive file beside enabling the file reader to clearly master the knowledge field of the archive file, and enabling the file reader to quickly find the necessary archive file according to other set classification items of the file even if the knowledge field of the archive file is unknown.

The invention provides a complex target asymptotic identification method and device based on knowledge classification. The method comprises the following steps: image preprocessing: dividing an original image data set I into data sets with multiple levels of resolutions to serve as a reference data set for asymptotic recognition of a complex target; inputting the images into a pre-trained VGG-16 network on an Image Net data set in batches for feature extraction; performing bilinear feature fusion calculation and trilinear feature fusion calculation on the extracted features with multiple resolutions; and predicting the category by using the fused features. According to the method, characteristics of trilinear pooling and bilinear pooling are fused, and a coarse-grained task and a fine-grained task of a complex target are planned in a unified framework. The feature reference provided by the coarse-grained task neglected in fine-grained identification in actual life is solved.

The invention discloses a scientific research knowledge management method and a system. The method comprises the following steps: (1) acquiring scientific research knowledge; (2) classification of scientific research knowledge; (3) constructing knowledge model; (4) retrieval; (5) displaying results. This method captures and acquires the knowledge generated by researchers in the process of scientific research through an integrated knowledge management platform, the explicit and implicit scientific research knowledge is processed by the model, and represented by the unified interface, and the scientific research knowledge is reused by the intelligent matching and reasoning algorithm, so that the knowledge resources of the laboratory can be accumulated and inherited, the scientific research progress can be accelerated, and the core competitiveness of the laboratory can be formed.

The invention discloses a user data preprocessing system for link prediction relation recommendation. The data preprocessing system comprises a link prediction relation recommendation layer, a data layer, a data processing layer, a system management layer, a service layer, an application layer, a data access layer and a view layer. According to the invention, a data preprocessing overall framework of a multi-agent architecture is constructed by analyzing the characteristics of a specific preprocessing algorithm and obtaining a corresponding intelligent recommendation scheme. The framework integrates the functions of a preprocessing algorithm, user interaction, system scheduling and the like into the whole system; the link prediction relation layer enables the framework to have openness and expandability, and provides support for preprocessing tasks under different backgrounds; aiming at the intelligent problem of the data preprocessing system, the knowledge discovery model agent describes each part of the data preprocessing process; a scientific algorithm recommendation scheme is provided for a user in a knowledge base mode, and intelligent recommendation of an algorithm is achieved through the knowledge classification capacity of a rough set theory.

The invention discloses a multi-screen knowledge visualization method based on knowledge classification. The method comprises the steps that (1) a logic regression classifier established based on a logic regression classification method in machine learning is utilized to divide knowledge into concept knowledge, process knowledge and application knowledge; (2) the knowledge is visualized, wherein a corresponding visualization tool is utilized to perform visualization operation on the knowledge, and visualized images of different types are made; and (3) a multi-screen technology is used to display the visualized images and text information of the visualized knowledge at the same time in combination with corresponding text description. Based on the knowledge visualization method, the viewpoint of classifying the visualized knowledge is proposed, display is performed in combination with a multi-screen teaching system, the information is displayed in a visual form and a text form at the same time, a learner enhances understanding and memorizing of the knowledge by processing the information through "dual coding", and application of the knowledge visualization technology in teacher teaching, knowledge transmission, thought enlightenment and other aspects is promoted.

The invention provides a medical intelligent diagnosis system establishing method for nervous system diseases. The method comprises the following steps of (1), establishing an artificial neural network model; (2), establishing a knowledge database; (3), performing expert knowledge classification arrangement; (4), performing neural network calculation output; (5), performing matching reasoning andobtaining a diagnosis conclusion; and (6), if a definite conclusion cannot be obtained, suspecting whether the disease is the nervous system disease, and reminding a user of necessary accessory examinations in a related knowledge database; after an accessory examination result is obtained, adding the result into input information, and repeating the steps of (3)-(5), and finally obtaining a definite diagnosis result. According to the method of the invention, early-period screening and early-period intervene are performed on the disease through big data, thereby finding the disease in a body examination process in the early period, and performing diagnosis through pattern recognition according to a big database.

The invention discloses a policy knowledge recommendation method. The method comprises the following steps of inputting the policy consultation question statement into a policy knowledge classification probability model; classifying and identifying policy consultation question statements; obtaining at least one knowledge label corresponding to the policy consultation question statement and the matching probability of each knowledge label, wherein the policy knowledge classification probability model is obtained by performing classification learning on knowledge point statements in a policy related file by utilizing a deep learning algorithm, and the knowledge point statements in the policy related file are labeled in advance based on a knowledge label system; determining policy related files and / or knowledge point statements related to the policy consultation question statements according to the knowledge labels corresponding to the policy consultation question statements and the matching probability of each knowledge label; and pushing related policy-related files and / or knowledge point statements to the user. According to the invention, a natural language processing technology and a machine learning technology are utilized to provide more intelligent, accurate and careful policy knowledge services for users.

The invention discloses a knowledge management system, which comprises a server and multiple clients, wherein the server is connected with the multiple clients; the server comprises an organization member database, a knowledge classification database, a knowledge database and a port module. The client is provided with an identity certificate, when the client is linked to the server, the port module authenticates the identity certificate to judge the authority of the client, then, the port module provides an operation port including at least one functional unit to the client according to the authority; and the operation port is provided for the client to set classifications of knowledge files according to a classification datum in the knowledge classification database and can provide an operation interface with which the client can store the knowledge files into the knowledge database.

The invention discloses a multi-source text knowledge representation, acquisition and fusion system, and particularly relates to the field of natural language knowledge classification in a knowledge base. The system solves defects that in the prior art, a full-intelligent system is relatively deficient in a knowledge base establishing mode which autonomously obtains knowledge, autonomous knowledgeretrieve, acquisition and combination technologies, the established knowledge base is relatively low in accuracy and precision, a full-manual operation mode would consume a large number of resources,and the error rate is high. According to the system, a balance point is found from resource consumption and knowledge base accuracy. The system specifically comprises a visual interface, control buttons and a knowledge merging interface. The control buttons comprise a keyword search button, a knowledge recognition button, a knowledge merging button and an exit button, and the system carries out text segmentation storage and keyword extraction from a known knowledge source, knowledge filling by using keywords, and knowledge disambiguation and verification, and then stores and generates a knowledge base.

The invention discloses a citrus knowledge modeling and large scale ontology generation method based on an agriculture eight-word policy. The method comprises the following steps: S1, based on the agriculture eight-word policy, knowledge in the citrus planting field is classified, wherein each word is corresponding to one type, and eight types are formed; S2, as for each type, concepts contained in the type are recognized, and the relationship among the concepts is built; S3, the eight types are used to form a citrus planting knowledge framework KF with the relationship among the concepts, wherein KF=(T, S), T is a citrus planting knowledge type, and S is a semantic relational layer; S4, the citrus planting knowledge framework is converted to eight sub ontologies; and S5, the eight sub ontologies are merged to form a large scale ontology. According to the method of integrating the citrus planting field knowledge based on the agriculture eight-word policy, the difficult problems of agriculture field knowledge integration and ontology collaborative development can be solved, and a complete scheme is provided for building and managing the large scale agriculture ontology.

The invention discloses an aircraft assembly process design rule knowledge management method and system, and relates to the technical field of digital assembly, and the method comprises the steps: firstly, obtaining assembly process design rule knowledge related to aircraft assembly instruction verification; secondly, performing generative framework representation and reasonable classification andattribute definition according to the characteristics of knowledge; and finally, designing a knowledge storage form, and associating the assembly process design rule knowledge object with the classification structure for knowledge classification and storage, thereby forming an assembly process design rule base. The system comprises a data layer, a logic layer and an application layer. And the data layer provides data related to the assembly process design rule to the logic layer. And the logic layer processes the rule data source and provides the processed rule data source for the applicationlayer in different data views. And the application layer formulates different application modes oriented to different roles for the data formed by the logic layer, so that the technicians are assisted in checking the assembly instruction.

The embodiment of the invention provides a model training method and device, a knowledge classification method and device, equipment and a medium, and belongs to the technical field of machine learning. The method comprises the following steps: acquiring original annotation data; the original annotation data comprises question stem data, option data and answer data; performing coding processing on the question stem data to obtain a question stem representation vector; encoding the option data and the answer data according to a preset knowledge graph to obtain an option attribute value and an answer attribute value; performing word segmentation and splicing processing on the option attribute value and the answer attribute value to obtain an option answer representation vector; performing vector splicing on the question stem representation vector and the option answer representation vector to obtain question data; a preset pre-training model is trained according to the question data, a knowledge classification model is obtained, the knowledge classification model is used for conducting knowledge classification on the target question, and the knowledge classification model obtained by the embodiment of the invention can improve the accuracy and efficiency of knowledge classification.