36 results about "Knowledge space" patented technology

A context knowledge modeling method is provided. The context knowledge modeling method includes the steps of: a) defining a context knowledge space as a two-dimensional space based on an abstract level and an application domain of knowledge; b) locating a share ontology as a highest level of the abstract level for defining a common ontology concept at a plurality of applications and services performed in various environment and domains; c) locating at least one of domain ontologies as a lower abstract level than the share ontology by taking over the ontology concept defined at the share ontology and defining a class and an attribute specialized at a corresponding domain and a developing application; and d) locating one or more instance bases expressing knowledge about real objects to have a lower abstract level than the domain ontologies.

The invention discloses a knowledge visualization interface system and method based on the virtual three-dimensional space. The system comprises a knowledge space construction module, a knowledge hotspot editing module, a knowledge space roaming module and the like, wherein the knowledge space construction module is used for introducing a space model and inserting the knowledge hotspot into the space model to generate and publish the knowledge space, the knowledge hotspot editing module is used for editing the hotspot entity, the knowledge content and a display mode, including selecting the display mode, replacing the hotspot entity, editing or uploading the knowledge content, adjusting the spatial attitude of the knowledge hotspot, etc., and the knowledge space roaming module is used forentering the knowledge space for three-dimensional roaming and triggering the knowledge hotspot to display the knowledge content. The system is advantaged in that individuals and organizations are facilitated to better utilize the visual and spatial cognitive ability of a human brain, reduce the cognitive burden in the learning process, improve learning efficiency of the individuals and the organizations, strengthen effective management of the knowledge flow and provide sufficient technical conditions for construction of learning type society and innovative organizations.

The invention discloses a load balance resource optimization method under a safety perceptive cloud radio access network. The method comprises the steps of collecting resource information allocated by a system; obtaining a convex optimization problem by use of regional balance and transmission balance strategies; optimizing a power under load balance condition by use of a self-adaption dynamic particle swarm optimization when a load balance threshold value is exceeded; initializing the parameters of the dynamic particle swarm optimization, initializing particle swarm space, and initializing a population; calculating the fitness of each unit in the population according to a fitness function; updating the optimum position of the current unit and the optimum position of the population according to a particle swarm optimization formula; obtaining the position and speed of each particle after displacement; carrying out intersection and variation on knowledge space; judging iteration convergence indexes; outputting optimum units if convergence occurs; and repeating the steps until the preset iteration times is finished if the convergence does not occur. The method is the load balance and power optimization method under safety perception through an optimization algorithm without changing network architecture under the condition that the system loads are imbalanced.

The invention discloses a secure resource optimization method under congestion control in a heterogeneous cloud wireless access network. The method comprises the steps of collecting system resource information in the heterogeneous cloud wireless access network; carrying out congestion control optimization system modeling on the system, and converting a model into three sub-processes; carrying out optimization operation on three sub-problems; initializing parameters of dynamic particle swarm optimization; initializing a particle swarm space, and initializing a particle swarm; calculating fitness of each individual in the swarm according to a fitness function; updating the optimum location of the current individual and the optimum location of the swarm according to a particle swarm optimization formula; carrying out intersection and variation operations on an knowledge space, thereby obtaining new individual locations and speeds after variation; and judging an iteration convergence index, outputting the optimum individual if convergence is carried out, and repeatedly carrying out iteration until preset iteration times is finished if the convergence is not carried out. According to the optimization method provided by the invention, the resource secure distribution of the heterogeneous cloud wireless access network can be improved under the congestion control, the energy efficiency resource is utilized fully, and the congestion control can be carried out dynamically.

The invention discloses an adaptive learning method and system based on a knowledge model, and a storage medium. The method comprises the steps: collecting initial learning data, storing the initial learning data in a learning resource database, carrying out the ontology term extraction and marking, and building the knowledge model through an ontology editor; initializing a knowledge space of thelearner according to a pre-measurement result of the learner in the learning system; updating the learning process state data input into the learner in real time, performing DKT model modeling of multiple knowledge points on the learning process state data through a DKT algorithm, feeding back the mastering state of each knowledge point of the learner, and dynamically updating the knowledge spaceof the learner; and the adaptive engine compares the knowledge space of the learner with the knowledge model, matches a learning path corresponding to the learner, and pushes a learning path link to the learner. According to the method, the DKT is tracked through deep learning based on the knowledge model, multi-knowledge-point modeling is effectively carried out, the latest knowledge state of a learner is accurately mastered and updated, and a corresponding learning path is matched.

The invention discloses a subject knowledge expression and description method. Starting from fusion of multiple subjects, expressions and descriptions of knowledge from complicated relationship to simplified relationship are searched with the basic theory of psychological cognition acting as guidance, and the phenomenon of influence of complicated relationship on cognition under mind mapping and other modes, especially cognition of subject knowledge of primary and secondary schools, can be solved. Factors of hierarchical cognition, association relation cognition, type cognition and knowledge information content are comprehensively considered, the expressions of the four factors are combined and the complete expressions of the knowledge can be obtained from different aspects. The relationship of knowledge cognition and geographical cognition is put forward firstly, and an automatic conversion and mapping method from a knowledge space to a map space is considered so that the application blank is filled.

The invention relates to the field of GIS space-time data model building, in particular to a pan-knowledge space-time object expression database building method. In order to solve the problems that an existing space-time data model has scarce capacity of space-time integrating expression and four-dimensional space-time data expression, the spatial or temporal continuity phenomenon cannot effectively expressed, and the geometrical shape and the gesture of a space-time object cannot be expressed, the pan-knowledge space-time object expression database building method comprises the steps that a space-time object expression data model is built, wherein the types of the space-time object expression data model comprise a rigid motion model type, a fluid motion model type, a field model type, a quasi-dynamic model type and a static model type; space-time data is collected; according to the type of the space-time data, the space-time data is modeled by using the space-time object expression data model, and the modeled space-time data is stored in a database. The method is suitable for space-time data modeling.

The invention provides a knowledge-space-based behavior result evaluation method. The method is applied based on the analysis result of a user offline prediction model provided by an operator, an airline or any other service providers, and the behavior result of a user that might be off the line is compared and evaluated finally. Through training an offline prediction model, the data of the user is subjected to offline prediction and probability prediction. Meanwhile, through evaluating the influence of marketing or promotion means on the offline prediction probability change, a most efficient marketing means is selected. Therefore, the retention is conducted for users that are mostly likely to get off the line. Meanwhile, the maximum revenue is realized on the premise that the cost is limited to a certain degree.

The invention discloses a knowledge graph visualization method based on a semantic attention model, and provides a knowledge visualization solution by combining multi-level multi-angle knowledge semantic similarity distance measurement with a user semantic attention model so as to support a visualization method of a global+local structure. According to the method, a semantic attention model between knowledge objects in a knowledge space is established, and the semantic attention between a focus object and the knowledge objects in the knowledge space is calculated; and visually displaying the knowledge graph according to the semantic attention. According to the invention, complex battlefield situation big data is analyzed and processed; the information objects with relatively short semanticdistances are highlighted and actively presented to the user; the big data knowledge graph retrieval method and system are beneficial to intelligently assisting a user in retrieving complex big dataknowledge, so that the user can see local details of an interested object and keep the overall impression of surrounding information, and good visual support is provided for big data knowledge graph retrieval application.

The invention provides a dynamic knowledge graph representation learning method and system based on anchor points, and the method comprises the steps: firstly finding key entities which play a role insupporting global information in an existing knowledge graph, and building a base coordinate system through the vectors of the entities; secondly, performing semantic alignment, including entity alignment and relationship fusion, on the newly added knowledge and the existing knowledge graph; finally, carrying out representation learning under a base coordinate system, so that only newly-added knowledge and related local knowledge of an existing knowledge graph need to be combined for training, a new knowledge entity is placed at a proper position in a knowledge space, and self-adaptive growthof the dynamic knowledge graph is achieved. The method has the beneficial effects that text information of entities and relationships is used as a semantic basis, and an information basis of knowledge fusion is provided, so that entity alignment and relationship fusion are more comprehensive and sufficient; a word2vec vector generation model is utilized to convert text information of entities andrelations into a vector form, so that the text information is used for mathematical operation.

The invention discloses a multi-modal knowledge representation method fusing entity image information and entity category information. A model carries out united joint training on structure-based knowledge representation, image-based knowledge representation and category-based knowledge representation of entities to complete multi-modal knowledge graph representation. The method includes an entity image information embedding method which is responsible for extraction of entity image feature information and conversion from an image space to a knowledge space; according to the entity category information embedding method, modeling is carried out on a semantic relation between an entity category and a corresponding triple relation, and representation of an entity based on the category is constructed; and the multi-modal knowledge representation model fusing the entity image information and the entity category information is responsible for multi-modal knowledge representation learning fusing the entity image information and the entity category information. By constructing the multi-modal knowledge representation method, a new thought is provided for solving the problem of data sparseness in a knowledge graph reasoning technology and efficiently calculating semantic relation, fusion and reasoning performance of entities and relations.

The invention relates to a student knowledge space system based on student growth archives and a management method thereof. The system comprises a growth archive data acquisition port and a data management chip. The growth archive data acquisition port is connected with an archive data processor through a transmission module; and the archive data processor is connected with the data management chip. The data management chip is provided with a plurality of bus ports, and is connected with an image record management storage module, a text record management storage module, a development state management storage module, a homework state management storage module, a physical condition management storage module, a reward and punishment situation management storage module and a book management storage module through the bus ports. The archive system is mainly used for realizing maintenance and management of growth and development of each student, promotes comprehensive development of the students, and plays a vital role for the students. The system is mainly used by students, parents and teachers. The system can stereoscopically present the processes from young to mature and from learning to life of each student.

The invention discloses a pedestrian detection method for a crossroad scene, and the method comprises the following steps: S1, preparing and arranging a training sample set of a crossroad scene, the training sample set comprising a positive sample set and a negative sample set; S2, establishing a background model of the crossroad scene by using an average background model algorithm; S3, training by using the training sample to obtain a classifier of each part; S4, combining the classifiers of each part by using a behavior knowledge space algorithm to form a final classifier; and S5, inputtinga to-be-detected picture into the model, and completing the detection and recognition of the pedestrian in the to-be-detected picture. According to the method, through fusion of the algorithm based onthe overall characteristics and the algorithm based on multiple parts, and combining the preprocessing of background modeling by using the similarity of scenes, a rapid, feasible and good-effect scheme is provided for pedestrian detection in a crossroad scene.

The invention discloses a symbol prediction method and system based on bridging domain transfer learning. According to the prediction method, the bridging domain is used, so that the network without intersection generates a public knowledge space, and knowledge of the symbolic network is efficiently migrated to the target network without symbols. According to the prediction method, the bridging domain selection algorithm based on the status theory is high in universality, and the target domain can be predicted without any bridging domain symbol information. According to the prediction method,interference samples can be effectively removed, so that effective migration of reliable knowledge is ensured, the prediction error is small, and symbol information in the target network can be accurately predicted.

The invention relates to a knowledge point evaluation method and device which can be realized through a cloud server. The method comprises the following steps: acquiring a test question set corresponding to a to-be-tested knowledge point set and determining a corresponding question bank knowledge space based on knowledge point subsets respectively corresponding to each test question in the test question set; determining a mastering result of an evaluated knowledge point according to a historical answer result of a target object, and screening candidate knowledge point subsets from the question bank knowledge space according to the mastering result of the evaluated knowledge point; determining a target knowledge point subset from the candidate knowledge point subsets based on the historical answer result and pushing target test questions corresponding to the target knowledge point subset in the test question set; stop pushing until test stop conditions are met; and obtaining a knowledge point evaluation result based on answer results obtained by answering the pushed target test questions by the target object. The knowledge point evaluation efficiency can be improved through the method, and the corresponding content is pushed to the target object according to the knowledge point evaluation result.

The invention provides an entity disambiguation method and device based on a UCL knowledge space, and the method comprises the steps: firstly constructing a basic knowledge base, and completing the construction of the UCL knowledge space; then obtaining a candidate entity set related to the entity to be disambiguated from the UCL knowledge space, and generating an embedded representation of the candidate entity and the entity to be disambiguated by using a word vector representation method; then extracting concept features of an entity to be disambiguated and contexts thereof, and extracting features of contexts of candidate entities; finally, the four vector representations generated previously are used as input, a self-attention matching network based on a deep structured semantic matching model DSSM is adopted, and the matching degree is obtained; and obtaining a final disambiguation result according to the sorting of matching results, and completing entity linking between the entities in the text and the entities in the UCL knowledge space. According to the method and the device, the problem of less entity related information in the short text can be solved, and the entity disambiguation accuracy can be improved.

The invention relates to an adaptive learning method and system of a knowledge space directed graph. The system comprises four core devices, namely a knowledge recognition device, a capability recognition device, an adaptive device and a capability tracking device. The knowledge recognition device and the capability recognition device are connected with personalized data through a knowledge vector, and are connected with a knowledge point through a capability directed graph. The capability recognition device and the adaptive device are connected with an adaptive module through capability recognition data, and are connected with the capability directed graph through learning resources. The adaptive device and the capability tracking device are connected with knowledge transfer through the adaptive module, and are connected with the learning resources through capability transfer.

The invention relates to a subject matter knowledge and information management method which comprises the following steps: establishing a first-level subject matter knowledge space for each first-level subject according to the subject classification; establishing a second-level subject matter knowledge space for each second-level subject in the first-level subject corresponding to the first-level subject matter knowledge space in each first-level subject matter knowledge space according to the subject classification; establishing a third-level subject matter knowledge space for each third-level subject in the second-level subject corresponding to the second-level subject matter knowledge space in each second-level subject matter knowledge space according to the subject classification; establishing a key scholar folder, a key project folder, a key journal folder, a key association folder, a key web site folder and a key conference folder in each third-level subject matter knowledge space, and associating third-level subject matter knowledge and information associated with the third-level subject matter knowledge space with the corresponding folders according to the classification by key scholars, key projects, key journals, key associations, key web sites and key conferences.

The invention discloses a knowledge representation learning method in combination with entity images and a server. The method comprises the steps of obtaining representations of the entity images corresponding to entities in a knowledge space according to the obtained entity images; according to the representations of the entity images corresponding to the entities in the knowledge space and structure-based representations of the entities, constructing image-based representations of the entities; and according to a triple of a knowledge graph and a preset training model, jointly training the image-based representations of the entities and the structure-based representations of the entities to obtain knowledge representations of the entities. According to the knowledge representation learning method in combination with the entity images and the server, provided by the invention, entity image information can be applied to knowledge representation learning, so that the knowledge representation learning accuracy is improved.

The invention relates to a knowledge graph technology-based optical transport network knowledge test method which comprises the following steps: constructing a knowledge point graph in the field of optical transport network OTN, and designing a representation method of a knowledge space by combining knowledge points and association relationships thereof; further designing a rule for extracting a knowledge state set from the knowledge space, wherein each knowledge state element in the knowledge state set represents a knowledge state which a learner may have; finally, designing a calculation method of the current knowledge state of the learner, deducing the knowledge point mastering level of the learner according to the knowledge point test result on the basis of automatically designing theknowledge point test sequence, and determining the current knowledge state of the learner from the knowledge state set.

The invention discloses a network knowledge unified representation architecture in the Internet of Things and an implementation method thereof. The architecture comprises a network knowledge classification module, a network knowledge space construction module, a local network knowledge representation module, a collaborative network knowledge representation module, a network knowledge storage module, a network knowledge fusion module, a network knowledge verification module and a network knowledge reasoning module. The method comprises the following steps: setting a network knowledge construction layer, a network knowledge representation layer, a network knowledge management layer and a network knowledge service layer, wherein the network knowledge construction layer is responsible for determining the category of network knowledge in the multi-domain environment, the network knowledge representation layer is responsible for systematic identification of network knowledge structures, modes and contents, the network knowledge management layer is responsible for network knowledge storage, fusion and verification under the condition that network communication resources are limited, and the network knowledge service layer is responsible for network knowledge learning, reasoning, distribution and on-demand acquisition. The method can support network elements to complete routing addressing, management control, security and other functions by using network knowledge, thereby reasonably deploying network resources.

Disclosed are a concept platform apparatus and method using schematic diagrams. The concept platform apparatus using a schematic diagram according to one embodiment of the present invention may comprise: a CK canvas module for providing a CK canvas divided into a concept space and a knowledge space, and connecting concepts and knowledge on the C-K canvas through a linking process to assist in a solution search for solving a problem; and a case management module for storing and managing, as one case, the C-K canvas on which the solution search is completed including knowledge, concepts, and information about the interconnection relationship.

The present invention relates to a method and system for realizing entity alignment in a cross-language knowledge space based on link prediction, which combines cross-language knowledge space triples, knowledge space representation learning, predicts new aligned entity pairs, and adds new training data through self-learning. The first step is to generate predicted entity pairs. For a small amount of training corpus, a simple method based on link prediction is designed to predict new entity pairs, and cross-language knowledge space fusion is performed. On this basis, data fusion in the two knowledge spaces is improved. The efficiency of judging whether the entity pairs in the cross-lingual knowledge space are the same entity, the design of the method model is relatively light, and it saves the manpower of labeling.

The invention relates to a method and a system for realizing cross-language knowledge space entity alignment based on link prediction. According to the method, predicted entity pairs are generated through four steps of cross-language knowledge space triple fusion, knowledge space representation learning, prediction of new aligned entity pairs and addition of new training data through self-learning, a new entity pair predicted based on a link prediction method is designed for small amount of training corpus, cross-language knowledge space fusion is carried out, the efficiency of judging whetherthe entity pair of the cross-language knowledge space is the same entity or not in data fusion in the two knowledge spaces is improved on the basis, the design of the method model is light and labeling manpower is saved.

The invention provides a quantum lookout non-circular direction finding method in an impact noise environment. The method comprises the steps: building a mathematical model for an array to receive non-circular signals, constructing a low-order real value weighted covariance matrix, and constructing a maximum likelihood direction finding equation through the low-order real value weighted covariance matrix; initializing a quantum lookout group and a quantum belief space, calculating the fitness of quantum positions in the quantum lookout group, and obtaining the optimal quantum position of the whole quantum lookout group; updating quantum specification knowledge, and updating a quantum situation knowledge space according to a lookout mechanism; using a simulation quantum revolving door to achieve the optimization searching process of quantum individuals through a quantum belief space and a quantum lookout mechanism; judging whether the maximum iteration number G is reached or not, if not, enabling g to be equal to g + 1, and returning to the step 3; and otherwise, terminating the iterative loop, and outputting the mapping state of the optimal quantum position in the last generation as a direction finding result. According to the invention, the method has robustness in a low-snapshot and impact noise environment, and breaks through the limitation of the existing non-circular direction finding method.

A logistics full-link integrated management and control digital twin system relates to the technical field of logistics management and control, and comprises a client, a server, a data processing system and a visual multi-user dimension management system;the data processing system is used for processing logistics full-link data; the server is used for extracting the data in the data processing system, converting the data into a data format identified by the client, and pushing the data format to the client; the client comprises a browser and a user interface, and the browser and the user interface are used for data display and operation display to realize three-dimensional construction of a data space twin body and an operation data knowledge space of a logistics full link; an intra-area cargo vehicle whole-process monitoring network system is established on the visual multi-user dimension management system, and the system performs offline and real-time analysis on logistics full-link data, constructs a data space twin body and operation data knowledge space, and provides intelligent decision support in combination with artificial intelligence and an operation planning optimization algorithm.

The embodiment of the invention belongs to the technical field of information correlation verification, and relates to a method for determining knowledge point correlation values in a knowledge space.The method comprises the following steps: according to a preset important value of a knowledge point under a node and a correlation value among all knowledge points under the node, determining the node, determining related values of all knowledge points under the node to the node, wherein correlation values among part of the knowledge points are known, and the knowledge points with unknown correlation values are replaced by a value approximate to 0, and the correlation values are unidirectional effective values; updating the correlation values among the knowledge points under the nodes according to the important values of the knowledge points under one node and the correlation values of all the knowledge points under the nodes to the nodes; repeatedly determining the correlation value between the knowledge points under the node and the node and the correlation value between the knowledge points until the correlation value between the knowledge points under the node is stable. According to the method, only part of data is required, and the effect of obtaining the correlation values among all knowledge points through algorithm operation can be achieved.