47 results about "Implicit knowledge" patented technology

Methods and systems that can detect GNSS spoofing attacks and that do not require explicit or implicit knowledge of exact position or attitude and that provide hypothesis test statistics, threshold values, and probabilities of false alarm and missed detection.

The invention provides a BIM and GIS integration method based on semantic fusion. The BIM and GIS integration method comprises the steps: S1, building a unified semantic framework of BIM and GIS integration, and forming a shared domain ontology; S2, obtaining instance information from the BIM data model to fill the shared domain ontology; S3, semantizing the GIS data and filling the GIS data intoa shared domain ontology in an ontology alignment mode; and S4, realizing information query and reasoning based on the fused domain knowledge model. The BIM and GIS integration method researches a method for converting IFC and CityGML into an RDF graph model, so as to enable the heterogeneous data to be expressed by a semantic-based knowledge model, thus solving the fusion problem of heterogeneousinformation of BIM and GIS, classifies, maps and hierarchically organizes classes and relations in the field to form concept integration under a unified framework, well processes and displays representation of knowledge, and infers implicit knowledge through establishment of logic rules.

The invention discloses an equipment resource configuration optimization method based on knowledge graph driving. The method comprises the steps that a knowledge graph of manufacturing resources is organized and generated, and machining process knowledge modeling and workshop machining equipment modeling are involved; the formed knowledge graph is subjected to a distributed representation learningand resource prediction matching method, and the relation between the knowledge graph and mining manufacturing resource implicit knowledge is further represented; when a processing order is issued, acandidate equipment set is constructed in combination with a fuzzy hierarchy method based on knowledge graph driving, and further the candidate equipment set is evaluated and screened through a community load model to complete reconstruction of the manufacturing unit; and finally, configuration optimization of an order task is completed by utilizing an optimization algorithm of the resource configuration mathematical model, an available processing equipment set link is formed, and a production process is guided. According to the method provided by the invention, the data processing capabilityis enhanced to a great extent, the equipment utilization rate and the equipment processing flexibility are improved, and meanwhile, the optimal configuration cost of manufacturing resources requiredby a processing task is also reduced through integrated knowledge reuse.

The invention discloses a building energy efficiency evaluation method and a system based on a hybrid semantic reasoning technology. The method comprises the steps that a cloud server obtains multi-source heterogeneous data (including semantic reasoning rules, semantic ontology data and semantic instance data) related to the building energy efficiency, and multi-level reasoning is conducted on themulti-source data related to the building energy efficiency through the hybrid semantic reasoning technology. A reasoning module comprises a Jena reasoning machine and a Pellet reasoning machine; theJena inference engine comprises a term axiom set assembly and an assertion fact set assembly. If yes, the cloud server performs reasoning on different data sources by adopting a term axiom set component, an assertion fact set component and a semantic reasoning rule component; the reasoning process relates to sensor network information, weather information, schedule information and the like, the output implicit knowledge / association forms a building energy efficiency knowledge base, the knowledge base can meet the data flow requirements of building energy efficiency evaluation indexes, automatic calculation of the building energy efficiency evaluation indexes is completed, and deep and all-around evaluation of the building energy efficiency is achieved.

The invention relates to an ANN and ES integration-based substation alarm information analysis and decision-making system and method. The system includes a database, a reasoning module, an explanation module, an implicit knowledge extraction module, a knowledge base and a man-machine interface module, wherein the database is used for storing real-time alarm information and equipment static configuration data which are received by the system, the reasoning module coordinates and controls the reasoning process of the alarm information, the explanation module explains the reasoning result of the reasoning module, the implicit knowledge extraction module is used for explicitly expressing knowledge expressed by numerical values, the knowledge base is used for storing explicit knowledge rules and implicit knowledge contained in connection weights, and the man-machine interface module is used for converting information outputted by the explanation module and the implicit knowledge extraction module into a form which can be understood by a user and outputting the form. According to the system of the invention, numerical value operation-based reasoning in an ANN is introduced into an ES system, so that single symbol-based expression of a traditional ES can be replaced, and therefore, the implementation efficiency of the ES can be improved; different analysis strategies are taken according to different faults; and therefore, the maintainability and fault-tolerance of a whole system can be improved, and the diagnosis range of the alarm information analysis and decision-making system can be also extended.

The invention belongs to the technical field of die design knowledge management and relates to a die design implicit knowledge acquiring and searching method. The die design implicit knowledge acquiring and searching method comprises the following steps: (1) expression of die design implicit knowledge; (2) acquisition of the die design implicit knowledge; (3) similarity calculation of the die design implicit knowledge; (4) searching of the die design implicit knowledge; (5) service workflow of the die design implicit knowledge. The die design implicit knowledge acquiring and searching method uses a designer knowledge structure model as a core, combines a body implicit knowledge base to express the die design implicit knowledge, organically combines the implicit knowledge with a knowledge owner, provides a comprehensive die design implicit knowledge life cycle processing method and means and helps an enterprise to promote value of knowledge resources. By statistics and analysis on data, the participation degree and the contribution degree of a user can be known; by virtue of a corresponding accumulated point counting and ranking mechanism, activity and the sense of honor of the user can be greatly stimulated.

The invention discloses a shield tunneling machine fault prediction method based on LSTM. Starting from actual shield construction practice, for the problem that multiple subsystems have faults at thesame time in shield construction, an LSTM-based shield system fault multi-label prediction model is established, and construction time sequence data is used for predicting the faults of the shield system. The model can automatically identify the implicit knowledge in the construction time sequence data and mine the rule that the shield system breaks down, manual analysis of interrelation betweenthe construction data is not needed, the fault prediction process is greatly simplified, constructors are helped to early warn system faults in time, and shield tunneling operation is reasonably carried out.

The invention provides a novel electromagnetic signal identification method and device based on a graph convolution network and transfer learning, and the method comprises the steps: constructing a graph structure based on the implicit knowledge of an electromagnetic signal; constructing a graph convolutional neural network, obtaining a category weight vector to which the novel electromagnetic signal belongs, and constructing an updated electromagnetic signal classification weight matrix; extracting a deep feature vector of the novel electromagnetic signal to be identified; and according to the updated electromagnetic signal classification weight matrix and the deep feature vector, completing transfer learning of the novel electromagnetic signal, and generating a perception identificationresult of the novel electromagnetic signal. The method can recognize the novel electromagnetic signal based on the graph convolution network and transfer learning, and effectively guarantees the recognition precision of the novel electromagnetic signal for a target, the robustness for the conversion of a scene and sensing equipment, the recognition response speed, and the adaptive capability whena new target appears.

The invention discloses a short text classification method based on multiple weak supervision integration, and the method comprises the steps: obtaining an original data set and a knowledge base, andcarrying out the data preprocessing; carrying out knowledge extraction on the preprocessed data; representing the extracted knowledge as an annotation function, and using the annotation function for data annotation; carrying out label integration through a conditional independent model; training a classification model based on a full-connection neural network; evaluating and optimizing the classification model to obtain an optimal model; and performing short text classification by utilizing the optimal model. According to the short text classification method based on multiple weak supervisionintegration, explicit knowledge and implicit knowledge are completely expressed in a mode of combining keyword matching, regular expression and remote supervision clustering; by means of probability labels generated by a label integration mechanism, automatic labeling of label-free data is achieved, the problem of data sparsity of short texts is relieved, and the problem of unbalanced classification of the short texts is effectively solved.

The invention discloses a knowledge distillation standing field detection method based on similarity maintenance, and belongs to the technical field of text emotion detection. The method aims to solve the problems that an existing vertical field detection method is large in computing resource and computing time expenditure and the recognition accuracy needs to be improved in the existing method. According to the method, the Text-CNN network model is utilized to carry out vertical field detection on a to-be-detected vertical field text; a Text-CNN network is determined based on a knowledge distillation mode, namely, a BERT model is used as a teacher model, and implicit knowledge learned from the teacher model is introduced into a student model text; and a loss function based on similarity maintenance is provided on the basis of a traditional knowledge distillation mode, so that the performance of the student model obtained on the basis of the knowledge distillation mode is further improved. The method is mainly used for text standing field detection.

The invention provides a knowledge discovery method applicable to a power grid management system. The knowledge discovery method applicable to the power grid management system comprises the steps of selecting a decision event on which knowledge discovery is carried out, extracting data relevant to the decision event, converting the data relevant to the decision event into typical examples corresponding to a classification problem, training a decision tree classification model, and extracting knowledge corresponding to the decision event. According to the knowledge discovery method applicable to the power grid management system, implicit knowledge contained in the power grid management system is converted into explicit knowledge which can be analyzed by the system, and therefore the operation efficiency of a power system is improved and the integral intellectualization level of a power grid is improved.

A reservoir scheduling method based on optimal convolution two-dimensionalization. The method comprises a step 1 of performing convolution processing on the input data based on a multi-objective optimization algorithm; a step 2 of constructing a dynamic scheduling model based on a convolutional neural network; and a step 3 of generating an evaluation model and an adjustment scheme, including a training section and a real-time scheduling section. The method, in combination with a deep neural network algorithm and a weight sharing technology, trains a deep artificial neural network to understandthe implicit knowledge in a scheduling scene through a large amount of scheduling historical data provided by a big data system, researches the time-space correlation of the input reservoir dynamic refinement comprehensive scheduling data, reduces the number of weights in each layer in the model construction by a neuron link method of sharing weight, increases the depth of the model so that the network fully recognizes the reservoir dynamic refinement comprehensive scheduling, finds out a deeper structure in the reservoir dynamic refinement comprehensive scheduling process, and finally completes a process of constructing a dynamic scheduling model with fast response and high accuracy.

The invention discloses a body based sub assembly identification method. A body knowledge framework is constructed for complex assembling by utilizing the body, implicit knowledge is inferred according to explicit knowledge, an assembling information knowledge base is developed, and a sub assembly is identified aimed at a rigid connection structure by combining a language rule base of semantic network rules. According to the invention, the consistent knowledge description framework in the assembling field is provided between heterogeneous systems, the assembling structure in the sub assembly identification level is structured, the scale of plan problems of the assembling sequence is reduced, and the assembling and manufacturing efficiency is improved.

The present invention discloses a semantic evaluation method based on the layered product design. A graphical method is used for visualizing the product design, dividing the product design into layers, and transforming implicit knowledge. In the visualized product design, the orthogonal design and the full-profile method are used for semantic evaluation of the product design; the regression analysis is used for calculating the importance of the product components in the semantic evaluation of the whole design; the degree of recessive utilization of each candidate shape of the product design is then evaluated, to acquire the semantic evaluation of the ultimate product design. The semantic evaluation method effectively realizes the expression of the implicit knowledge in the process of the product design, compensates the quantitative research on the graphical method, and improves the semantic research efficiency of the product design.

The invention discloses a knowledge base system applied to government affair hall operation. The knowledge base system comprises a knowledge layer, a knowledge processing layer and a knowledge application layer, wherein the knowledge layer is a database layer and is used for data support and knowledge support of knowledge base system operation and providing knowledge base extension service; the knowledge processing layer provides a cloud AI platform deep learning processing module and a big data distributed cooperative processing module, is used for mining, analyzing, discovering and storing collected knowledge or data, and is associated with knowledge base information; and the knowledge application layer is used for users to obtain knowledge, information, data and services. The inventionprovides a knowledge base system applied to government affair hall operation. The application field of a knowledge base system in government affairs is expanded, implicit knowledge is mined, a government affair knowledge database is converted into the knowledge base system with a service application system, knowledge application and knowledge innovation are achieved, and a high-level data analysisresult and a more targeted problem solution are provided for a government affair hall.

The invention discloses a knowledge graph construction method integrating teaching feedback and learned understanding. The method comprises the following steps: 1) data acquisition: converting behaviors of students in a classroom into a text for describing the understanding degree of the students on current knowledge; 2) named entity recognition and relation extraction: performing entity recognition and relation extraction on classroom text data at the same time to obtain a related entity triple; 3) embedding the entity and the relation, generating information of neighbor nodes by a weighted graph convolutional network, learning richer semantic expressions of the entity and the relation to form a final entity embedding expression; 4) scoring alternative tail entities in the knowledge graph triple through a multi-scale convolutional neural network, selecting the tail entity with the highest score as a reasoning result, further reasoning implicit knowledge, and updating the knowledge graph; wherein the alternative tail entities are all entities updated in the step 3). According to the invention, a specific knowledge graph of each student can be constructed, and learning and teaching evaluation feedback can be carried out.

The invention discloses a natural environment effect knowledge reasoning method based on hybrid reasoning. Through the method, implicit knowledge mining can be performed on structured explicit knowledge (e.g., an ontology model), the natural environment effect and reliability design analysis form dynamic connection, and natural environment effect relevant knowledge can be applied to actual reliability design analysis work. The method comprises the steps that 1, case reasoning of the natural environment effect knowledge is constructed, including case library construction based on feature vectors and a case reasoning method based on property similarity measurement; 2, rule reasoning of the natural environment effect knowledge is constructed; and 3, hybrid reasoning rules of the natural environment effect knowledge are designed.

The invention belongs to the technical field of assembly tolerance specification design and computer-aided tolerance design (CAT), and particularly relates to an assembly tolerance specification intelligent design method based on Ontology. The method specifically comprises the following steps: (1) constructing a body of an assembly tolerance specification; (2) establishing an SWRL reasoning rule of an assembly tolerance specification; (3) extracting related tolerance information of the assembly tolerance specification; (4) constructing an assembly tolerance specification instantiation ontologymodel; and (5) combining the Jess inference machine with the SWRL rule, inferring the assembly tolerance specification, and outputting a result inference result. An ontology reasoning knowledge framework for the assembly tolerance specification domain knowledge is constructed by utilizing an ontology, and an assembly tolerance specification knowledge base is expanded according to dominant domainknowledge reasoning implicit knowledge. In combination with a semantic network rule language rule base, reasoning design parameters of assembly tolerance specifications for specific design of geometric products and using environmental conditions. The consistency knowledge description framework of the assembly tolerance specification can be provided, so that a computer can automatically select appropriate tolerance principles, tolerance types, size parameters and the like according to shape and size parameters of geometric products, and a quick and effective method is provided for intelligent design of the assembly tolerance specification.

The invention discloses a collaborative sensing method based on multi-sensor information fusion, which comprises the following steps: 1, simulating a signal, establishing a signal model, and transmitting the signal to each node at the same time; 2, classifying and identifying the signal by each sensor node, classifying the signal, and transmitting the classified signal to different processors; and 3, calculating the credibility of the nodes, and providing cooperative control and decision in a complex manufacturing environment. According to the invention, by setting self-adaptive perception oriented to an open environment, a hierarchical network structure with strong adaptability, a machine learning strategy capable of continuously learning and a general efficiency measurement method are developed for solving the problem that the performance of an intelligent system is sharply reduced easily due to application scene change, the difficulties of unsupervised learning, experience memory utilization, implicit knowledge discovery and guidance, attention selection and the like are broken through, and a new machine learning method based on cross-media human common knowledge formation is researched.

The invention provides a question information obtaining method and device, electronic equipment and a storage medium, and is applied to the field of computers. The method comprises the steps that to-be-processed first text information is obtained; extracting candidate question information matched with the first text information from a preset question and answer library; according to an implicit knowledge graph, implicit knowledge information in the first text information is filtered, and second text information is obtained; respectively obtaining a first semantic feature of the first text information, a second semantic feature in the second text information and a candidate semantic feature in the candidate question information; and taking the candidate question information of which the feature similarity between the candidate semantic feature and the first semantic feature and the feature similarity between the candidate semantic feature and the second semantic feature meet the similarity requirement as target question information. According to the method and the device, the interference of semantic redundancy of implicit knowledge information on question information matching can be reduced, and the accuracy of question information extraction is improved, so that the determined question better fits the actual question intention of the user.

The invention discloses a system for middle and primary school education resource application, which is an education cloud resource sharing platform and comprises a user management unit and a resourcemanagement unit, the user management unit comprises a teacher unit, a student unit, a parent unit and a background management unit. Wherein the teacher unit, the student unit and the parent unit acquire learning resources through the resource management unit, a teacher uploads courseware through the teacher unit, the courseware comprises a method for converting implicit knowledge into explicit knowledge, and an administrator performs auditing, recommendation, offline and deletion management on the learning resources of the cloud resource sharing platform through the background management unit; and a message leaving unit and an integral unit are arranged on the resource management unit. The method for converting the implicit knowledge of the teachers into the explicit knowledge has the advantages that the teachers can share high-quality teaching resources on the sharing platform conveniently, excellent teaching experience of the teachers can be popularized conveniently, and more teachers and students are benefited.

The invention discloses an electromagnetic signal identification method and device for constructing a graph convolutional network based on implicit knowledge. The electromagnetic signal identificationmethod comprises the steps: obtaining a plurality of electromagnetic signals, extracting the feature descriptions of the plurality of electromagnetic signals, generating a feature matrix according tothe feature descriptions of the plurality of electromagnetic signals, and enabling the feature matrix to serve as first-class input information; mining implicit knowledge of the plurality of electromagnetic signals, and constructing a graph structure based on the implicit knowledge of the electromagnetic signals as second-class input information according to the implicit knowledge; and building agraph convolutional neural network according to the first type of input information and the second type of input information, and performing loop iteration training on the graph convolutional neuralnetwork by using a weak supervised learning method to enable the graph convolutional neural network to output the types of the plurality of electromagnetic signals. According to the electromagnetic signal identification method, the features of each node can be utilized to mine the relation between the nodes, and the features of each electromagnetic signal category are deeply extracted, and a graphconvolutional neural network is constructed, and the electromagnetic signals are identified, and more information is utilized, and the generalization capability is higher.

The invention relates to the technical field of oil and gas pipeline repair, and discloses a knowledge graph construction method and system for oil and gas pipeline defect repair, which comprises thesteps of collecting related data of pipeline defect repair, performing arrangement and classification according to a pipeline defect repair process to form a preliminary pipeline defect repair data model framework, analyzing and disassembling the pipeline defect repair data model framework, determining a corresponding defect repair business data domain, determining a triple of pipeline defect repair knowledge, identifying the triad, completing knowledge extraction of pipeline defect repair, and importing the identified triad into a graph database to construct a knowledge graph. According to the method, existing pipeline repair implicit knowledge is explicit, pipeline repair business knowledge reference is provided for a user based on semantic search and semantic matching by means of visualgraph data and knowledge reasoning, a foundation is laid for accumulation, transmission and sharing of defect repair knowledge points, and the knowledge service is provided for the user in the defectrepair business activity, and meanwhile, the way of improving the user skills is expanded.

The invention provides a self-healing control method and device for a power distribution network based on a rough set dynamic programming model. The self-healing control method comprises the steps ofstep 1, establishing a multi-objective optimized power distribution network self-healing control model; 2, performing data knowledge reduction on power grid data by applying a rough set theory; 3, performing fault location through evaluation function calculation and non-fault section search; and step 4, rapidly obtaining an optimal self-healing scheme of the intelligent power distribution networkto control fault self-healing. The self-healing control method for the power distribution network based on the rough set dynamic programming model is provided for solving the problem of how to quicklyrecover the load power consumption of a non-fault area after a fault of the power distribution network occurs, the rough set theory is adopted to process various incomplete information such as inaccuracy, inconsistency and incompleteness in the fault self-healing process, implicit knowledge and potential rules are found by directly analyzing and reasoning data, and the self-healing recovery speedof the system is improved.

The invention discloses a catalog and subject service business collaboration method based on a pre-press catalog. The catalog and subject service business collaboration method based on the pre-press catalog includes steps that (1) using a primary classification center and a secondary classification center of a pre-press catalog organization to decompose subject service tasks and judge the subject service task characteristics; (2) using a pre-press catalog working process to correspondingly match sub-tasks with a secondary subject catalog center; (3) enabling a secondary subject cataloger to confirm subject service types (explicit knowledge service or implicit knowledge service) according to subject service characteristics; (4) enabling the secondary subject cataloger to use the subject speciality to synergistically provide explicit knowledge service or implicit knowledge service according to the subject service types, wherein the service related to interdisciplinary knowledge needs to build a virtual service team; (5) enabling a superior classification department to use the pre-press catalog organization to synthesize the secondary subject service content; (6) cooperating with workers of a subject service center organization to evaluate the service content and create a catalog database for the similar subject users for referencing. The catalog and subject service business collaboration method based on the pre-press catalog smartly uses the subject characteristics to cooperate the catalogs with the subject service business.

The invention discloses a method for digitally evaluating difficulty of questions. The calculation method is based on the following two steps: S1, basic assumption; s2, identifying and calculating answer keywords, and measuring the difficulty of questions according to the number and attributes of the answer keywords. On the basis of S1 and S2, the calculation method is used for evaluating the difficulty of questions of courses of different subjects and comprises the following steps that S01, parameters Sigma P =-ACGP + HKP + CFP + IP + SP + RP + RDP + CP + SRDP are calculated, and ACGP is an answer guessing parameter, wherein HKP is an implicit knowledge point parameter, CFP is a complex parameter, IP is an interference parameter, SP is a static parameter, RP is a reaction parameter, RDP is a reaction dynamic parameter, CP is a change parameter, and SRDP is a synthesis design parameter; and S02, a difficulty parameter of the question as DP is defined, and DP = AKP + Sigma P + Sigma WP+ Sigma LP is calculated. According to the calculation method, the difficulty of the questions is evaluated in advance by a digital means, so that the difficulty of the questions of different subjectscan be transversely compared.

The invention belongs to the technical field of rolling bearing clearance design and computer-aided tolerance design (CAT), and particularly relates to a rolling bearing radial clearance intelligent design method based on a body. The method specifically comprises the following steps: (1) constructing a body representation model of the radial clearance of the rolling bearing; (2) establishing an SWRL reasoning rule of the radial clearance of the rolling bearing; (3) extracting tolerance information related to the radial clearance of the rolling bearing; (4) establishing an instantiated body model of the radial clearance of the rolling bearing; and (5) combining the Jess inference machine with the SWRL rule, inferring the radial clearance of the rolling bearing, and outputting a result inference result. According to the method, the body theory is introduced into the design of the radial internal clearance of the rolling bearing, an inference knowledge framework of the radial internal clearance of the rolling bearing based on body pairs is constructed, implicit knowledge is inferred according to explicit domain knowledge, and the knowledge base of the rolling bearing is expanded. Theinvention provides a consistency knowledge description framework in the field of the radial clearance of the rolling bearing, so that a computer can automatically select a reasonable radial clearancegroup and a reasonable clearance value according to the actual working condition environment of the rolling bearing, and a feasible method is provided for intellectualization of the radial clearance of the bearing.

The invention relates to a PEC course question and answer method based on a knowledge graph and oriented to autonomous learning. The method comprises the following steps: obtaining a knowledge data packet provided by a teacher end, the knowledge data packet comprising knowledge point data and problem processing data; performing data cleaning on the knowledge data packet; calling the Neo4j graph database and the elastic search database to construct a knowledge graph through a Neo4j technology; obtaining a keyword input by a user in a search engine; analyzing the keyword to obtain a current question, and extracting an answer from an elastic search database, wherein the matching degree between the answer and the current question is greater than a first threshold value; and displaying the answers and the knowledge maps corresponding to the answers according to a sequence of matching degrees from high to low. Multi-angle implicit knowledge can be automatically and intelligently found and reasoned from existing knowledge, so that students can conveniently check, supplement, digest and absorb the knowledge, and the time of teachers is greatly saved.

The invention discloses a case implicit knowledge mining method and device and electronic equipment, and the method comprises the steps: constructing an initial case knowledge graph of target case information, and the target case information comprises case text information and / or legal information; based on a pre-created ontology library, performing rule reasoning on the initial case knowledge graph to obtain a first implicit information set; based on a pre-generated graph neural network model, performing information prediction on the initial case knowledge graph to obtain a second implicit information set; and processing the first implicit information set and the second implicit information set to obtain target implicit information. The purpose of improving the calculation efficiency and the information mining accuracy is achieved.