61 results about "Web of knowledge" patented technology

A data construct called a semcard is a semantic (meaning-based) software object including semantic meta-tags and meta-data that describes a target object or thing. A target object can be any type of digital or physical entity or identifier, or it can be tacit knowledge, such as ideas, concepts, processes or other data existing in a user's mind, provided that the user represents this knowledge in the semcard. A semcard embodies information about its own structure-rules, history, state, policies and goals regarding automation, display, access permissions, sharing and other operations of the semcard and any optional target object. It can also represent a semantic link between two semcards, or a semantically typed link or a standard Web hyperlink between a semcard and its referent target. A collection of semcards represents a knowledge network; single semcards, and knowledge networks, can be browsed, shared, searched, disseminated, manipulated, displayed, organized, and stored.

Among other disclosure, a knowledge network and semcards enabling intelligent matching of offers and requests, involving all types of information and knowledge, including information such as classified ads, data about products and services, or knowledge, expertise, ideas, suggestions, opinions, and other forms of tacit knowledge is described.

The invention discloses a method and system for constructing an intelligent tourist customer service system. The method comprises the following steps of collecting the tourism industry data, classifying the tourism industry data according to ontology concept to construct tourism industry domain ontology knowledge base; through the intelligent response engine, analyzing the user terminal travel problems, and extracting keyword information; retrieving corresponding answers in the ontology knowledge base according to the extracted keyword information; if the matching answer is retrieved, feedingback the answer to the user terminal through the intelligent response engine; if no matching answers are retrieved, answering the answers by manual customer service, and saving the questions and the answers to the ontology knowledge base by a self-learning module; saving customer service data to the data management platform for the next quick response. Compared with the related technology, the construction method of the intelligent tourist customer service system intelligently understands the user's problem, and adopts knowledge network to establish knowledge base, which is easy to maintain.

The invention provides a practical effect evaluation and learning path recommendation method based on cognitive diagnosis. The practical effect evaluation and learning path recommendation method comprises an expansion feature preprocessing model, a deep knowledge tracking model, a knowledge network construction model and a path recommendation algorithm based on cognitive ability. The expansion feature preprocessing model performs primary evaluation of cognitive competence according to skill attributes in a learner test process, and introduces personalized difference information into the diagnosis model. And the deep knowledge tracking model predicts the knowledge mastery ability of the learner according to the test sequence and the heterogeneous features of the implicit coding, and the knowledge mastery ability serves as a basis of learning guidance. The exercises and knowledge network construction model provides a global map of scientific thinking, and the cognitive diagnosis is combined to recommend a learning path to a learner, so that the cognitive ability difference in the learning process is considered, and the logic of a knowledge structure is followed.

The invention discloses a method and a terminal for automatically generating ideas based on a knowledge network. The method comprises the following steps of firstly, building an invention knowledge network, extracting keywords capable of representing invention features from patent data, using Euclidean distance to calculate the similarity of keywords in a vector space model, clustering the keywords with high similarity by a layered clustering type, and establishing an idea keyword set with layered relationship. The method has the advantages that by customizing the invention goal, the keyword set of the invention ideas is extracted to form a word cloud according to the customized goal, the correlation degree of the keywords in the word cloud, and the layer compactness of the set in the word cloud; the required idea information is marked, the invention idea is automatically generated by a trained language model, and the idea guide process is completed.

The embodiment of the invention provides an abnormal data processing method and system applied to a business system, and the method can achieve the mining of abnormal behaviors in an abnormal behavior mining model based on abnormal operation activity data, and acquires a candidate abnormal behavior cluster. Therefore, a frequent behavior item knowledge network of the abnormal operation activity data is constructed based on the frequent behavior item data in the abnormal behavior item data in the candidate abnormal behavior cluster, and the behavior contact data corresponding to the frequent behavior item knowledge network is mined, and by considering the behavior contact data of the frequent behavior item knowledge network, the reliability of the distribution process can be improved when abnormal knowledge node distribution is performed on the candidate abnormal behavior cluster based on the behavior contact data.

The invention discloses a time sequence knowledge graph completion method based on a space-time architecture, and the method is characterized in that the method comprises the following steps: dividinga to-be-completed time sequence knowledge graph into a plurality of static knowledge sets according to the time marks of knowledge of the to-be-completed time sequence knowledge graph, respectively constructing a plurality of knowledge networks through the knowledge in each set, and obtaining a plurality of snapshots; constructing a polyhedral graph attention network, inputting the snapshots intothe polyhedral graph attention network, and obtaining static embedded representation of the entity under each snapshot; constructing a self-adaptive time sequence attention mechanism, and obtaining afinal embedded representation of the entity according to the static embedded representation of the entity and by using the self-adaptive time sequence attention mechanism; and calculating confidenceof knowledge in the to-be-completed time sequence knowledge graph through final embedding representation of the entity, and predicting missing content in the to-be-completed time sequence knowledge graph through the confidence. The invention has higher expansibility and flexibility, and can be combined with any static completion method to complete the timing sequence knowledge graph.

The invention discloses an intelligent decision-making method and device based on a knowledge network, a terminal and a storage medium, and the method mainly comprises the steps: carrying out the knowledge expression of source knowledge data based on the knowledge federation of a cognitive layer, and enabling the source knowledge data to be the knowledge data provided by a participant participating in the construction of the knowledge network; forming knowledge graph libraries of different fields according to the knowledge expressions corresponding to the participants, wherein knowledge graphsin the knowledge graph libraries form a knowledge network; and when a decision request initiated by a target participant is detected, knowledge expression extraction is performed on the to-be-decideddata, and a final decision result is determined in the knowledge network according to a knowledge expression result. By adopting the method and the system, a more comprehensive decision basis can beprovided for intelligent decision, and the difference between'perception intelligence 'and'cognitive intelligence' is reduced.

The embodiment of the invention relates to the technical field of information processing, and discloses a brief report generation method and device, electronic equipment and a computer readable storage medium, and the brief report generation method comprises the steps: determining text features of a to-be-generated brief report text, and determining each concept word of the text through a conceptword classification system according to the text features, determining an association relationship among the conceptual words of the text through a conceptual word relationship classification system;then, a tree-shaped knowledge network model being generated according to the knowledge network and the text structure of the text, and the knowledge network being generated according to the concept words and the incidence relation between the concept words; then, based on a tripartite deep network representation TriDNR algorithm, a Node2Vec algorithm and a TransE model, generating a network representation learning model according to the tree-shaped knowledge network model; and then generating a brief report according to the network representation learning model and the tree knowledge network model.

A similar case recommendation method based on network representation learning is mainly used for finding out ten legal documents most similar to legal documents in a huge legal document database underthe condition that the legal documents are given. According to the invention, the method includes taking legal instruments on the Internet as starting points, using a crawler technology for crawlinga large number of legal instruments, and using regular expressions for carrying out preliminary processing on the legal instruments; crawling a large number of legal entities by utilizing a crawler technology, calculating the similarity between the legal entities through bert, constructing a similar entity relationship between the legal entities reaching a threshold, constructing a legal knowledgenetwork by utilizing the legal entities and legal documents, inputting the constructed network into a network representation learning model to obtain vector values of the legal documents, and obtaining ten documents most similar to the legal documents through cosine similarity calculation.

The invention relates to a path learning planning system based on a knowledge graph and a target ontology. The path learning planning system comprises a subject data acquisition module, a student information acquisition module, an evaluation demand acquisition module, a knowledge graph construction module, a data link module, a data operation module and a data operation module. On one hand, corresponding learning plans and learning plans can be made for students in a targeted manner according to the respective requirements and knowledge reserve conditions of the students; therefore, the requirements of learning ability and learning habits of students are met; on the other hand, the learning contents are effectively combined; the knowledge points are reasonably combined to construct an efficient knowledge network, the distribution rationality of the knowledge points in the knowledge network and the comprehensiveness of the knowledge network coverage content are improved to meet different learning requirements and flexibly adjust the learning requirements, and finally, the convenience and the flexibility of students during learning are improved.

A learning system includes a non-transitory memory, and one or more hardware processors configured or programmed to read instructions from the non-transitory memory to cause the learning system to perform operations including generating a user knowledge mesh, wherein the operation of generating the user knowledge mesh includes generating a plurality of topic nodes, each of the plurality of topic nodes corresponding to a topic included in the user knowledge mesh, and generating a plurality of concept nodes, each of the plurality of concept nodes corresponding to a key learnable concept, wherein each of the plurality of topic nodes is connected to another one of the plurality of topic nodes, each of the plurality of concept nodes is connected to one of the plurality of topic nodes, and each of the plurality of key learnable concepts includes one or more interactions related to the key learnable concept.

The invention discloses a freight supervision abnormal information collection method and device, a server and a storage medium. The method comprises the steps that triples are extracted based on freight supervision data to generate a knowledge network graph; obtaining a plurality of corresponding sub-graph query results through a plurality of preset sub-graph query models based on the knowledge network graph; determining the support degree of each sub-graph query result; determining the sub-graph query result with the support degree greater than a support degree threshold as a target sub-graphquery result; and determining a confidence coefficient of a preset association rule according to the target sub-graph query result, and determining abnormal information of the freight supervision process based on the confidence coefficient and the preset association rule. Compared with a traditional method, the method pays more attention to the association relationship between entities, each preset association rule has practical significance in a specific behavior mode, two entities without the practical association relationship cannot be mined, the efficiency and the accuracy are higher, andmeanwhile, the result is displayed more visually by utilizing a calculation mode of a graph structure.

A method for processing data for discovering a new drug candidate substance by a data processing apparatus, includes receiving a predetermined search word, extracting at least one biological entity related to the predetermined search word from a big data database (DB), extracting a degree of mutual association between the predetermined search word and the at least one biological entity, generating a first knowledge network in which a plurality of nodes including the predetermined search word and the at least one biological entity are connected according to the degree of mutual association, computing a graph theory index of the first knowledge network, and generating a second knowledge network using some nodes of the plurality of nodes of which the graph theory index is equal to or greater than a threshold value.

The invention relates to the technical field of information systems, in particular to an enterprise-based policy and regulation acquisition information system, which comprises a policy and regulation knowledge graph model, a knowledge management subsystem and a policy and enterprise automatic matching platform. According to the system, the defects in the prior art are overcome, the knowledge management subsystem and the knowledge sharing module are arranged, and information retrieval, browsing, publishing and evaluation are completed through the knowledge security management module and the dynamic diversified browsing component; and the knowledge management module completes information auditing, induction and arrangement through knowledge automatic collection, classification and correlation analysis and portable knowledge network construction, the purpose of more emphasizing management of the knowledge forming process besides knowledge document management is achieved, and the knowledge management system can more emphasize knowledge visual expression. The problem of enterprise information isolated island is solved by using a new mature informatization technology of the Internet.

The invention discloses a knowledge recommendation method and system based on improved position social contact and a storage medium. Knowledge points needing to be accessed by a user in the next stepare predicted by utilizing existing access data of the user and other data information in a knowledge network; the prediction result is recommended to the user to serve as a learning recommendation list of the user, the current situation that a common commodity-oriented recommendation model and algorithm are not suitable for recommendation on a knowledge learning system can be solved, and the knowledge obtaining efficiency of the user on the knowledge system and the learning efficiency of the user are improved; meanwhile, the defect that at present, common recommendation system models and algorithms are mainly used for recommending commercial articles is overcome; according to the method, the path set of the knowledge points accessed by the user is constructed, knowledge recommendation inthe knowledge social network is realized through an improved position social recommendation algorithm, and the efficiency of learning and understanding certain knowledge by the user is improved. The method can be applied to the field of knowledge recommendation systems.

The invention provides a learning trajectory-based learning ability evaluation and extended knowledge point set recommendation method. The method comprises steps: a knowledge network and the syllabusare built; according to the syllabus, a required learning path and an initial recommended elective extended knowledge point set are built; according to each syllabus requirement and answers, a test set is built; according to the learning path and the learning state achieved by the learner himself or herself, the learner selects new knowledge points for learning, the learning condition of the knowledge points is recorded, an initial learning trajectory is built, the mastering condition of each knowledge point is evaluated, a dynamic learning trajectory for the mastering condition is built, thecomprehensive mastering ability score, the mastering breadth score and the comprehensive learning efficiency score of the learner are evaluated, and whether to achieve the current course mastering requirements is evaluated; the mastering breadth level is evaluated for the learner which already achieves the mastering requirements; and the comprehensive learning ability level of the learner is evaluated, a proper elective knowledge point set is selected, and individualized elective extended knowledge point set recommendation is carried out.

The invention provides a data preprocessing method and device, equipment and a medium. The data preprocessing method comprises the steps of obtaining a user demand; matching a knowledge network adaptive to the user demand in a knowledge network library; if the knowledge network adapted to the user demand is matched, classifying and combining the data sources according to the knowledge network, andestablishing a data set in which knowledge points among the data are mutually associated in the data preprocessing process; and if the knowledge network adapted to the user demand cannot be matched,preprocessing related data according to the user demand. In the data preprocessing phase, a knowledge network adaptive to user requirements is combined, and data sources are classified and combined, and a data set with knowledge points associated with one another among the data is established in the data preprocessing process instead of consuming a large amount of time to repeatedly filter uselessdata when the data is required each time, so that the calculation process is simplified, and the data preprocessing efficiency is improved, and the data preprocessing operation cost is reduced.

The invention discloses an air traffic control equipment application system and method based on a knowledge graph, and belongs to the field of air traffic management. The problems that in the prior art, an equipment monitoring system is relatively closed, the integration level is low, and interconnection and intercommunication are affected are solved. The invention provides an air traffic controlequipment knowledge graph application system. The air traffic control equipment knowledge graph application system comprises a data source layer, an atlas layer, an algorithm layer and an applicationlayer, the data source layer is used for importing air traffic control equipment data; the atlas layer is used for extracting, storing and mining the data of the data source layer to generate an equipment knowledge atlas; the algorithm layer is constructed on the map layer and used for connecting the map layer and the application layer, and algorithms are flexibly selected according to applicationrequirements; the application layer analyzes and processes the knowledge graph by using a corresponding algorithm to realize an application function; The method is applied to a system. According to the method and the system, an equipment link is routed, global data consistency is maintained, equipment reliability and safety are further improved, and an air traffic control equipment knowledge network is formed.

A method and system comprising receiving genetic and non-genetic data of an individual, calculating a trait predisposition score for the individual, organizing a knowledge base repository using dynamic network analysis of a plurality of genetic variants and of a plurality of phenotypic traits into a heterogeneous knowledge network model, assessing regulatory, catalytic or inhibitory utility of genetic factors by determining existence of said genetic factors within biological pathways, calibrating the phenotypic trait predisposition score for the individual using a machine learning analysis that relates the plurality of genetic variations to the plurality of phenotypic traits, calibrating the heterogeneous knowledge network model using the genetic data and the non-genetic data.

The invention provides a constructive neural network method based on knowledge cascading correlation, wherein the method relates to the field of artificial neural network, machine learning and artificial intelligence. Particularly the invention relates to cascading of learned knowledge with the neural network through relativity, and furthermore the neural network is constructively reconstructed based on cascading of the knowledge, thereby realizing a purpose of quickly and flexibly constructing the neural network. According to the method, a network structure and weight are adjusted according to training data in a frontward propagating neural network. The method mainly comprises two periods, namely an input period and an output period. In the output period, all weights which enter an output mode is optimized through a training algorithm so that the output error of a target network is reduced continuously. In the input period, a new hidden layer node or knowledge network is added into the target network. A candidate network which is finally selected for being added into the candidate network can best correlate an output error of the target network.

The invention provides a knowledge network management method and a device based on financial services, equipment and a medium, which are suitable for improving the flexibility of a user in using a knowledge network platform, and the method comprises the following steps: obtaining an input financial service demand; detecting whether a corresponding custom script exists or not according to the inputfinancial service demand; when it is detected that the custom script exists, editing the custom script to change the processing logic of the script to the data so as to match the data source meetingthe financial business requirement and output the data source. According to the method, whether a custom script exists is detected; when it is detected that a custom script exists; the custom script is re-edited according to an input financial service demand; the processing logic of the custom script to the data source is changed, the financial service requirements of the user are flexibly met, the data source adapting to the financial service requirements is matched and output, the financial service requirements of the user are met in a personalized mode, and the flexibility and compatibilityof the knowledge network platform are improved.

The invention discloses a method for making professional knowledge into software to help families choose the best food and belongs to medical treatment, body conditioning, nutrition, health care, climate, weather, regions, customs, occupations, ages, public preferences, purchase histories, body building, chef field, gastronomy, personal preferences, dishes, fruit knowledge, data, etc. With the method adopted, best dishes, fruits and snacks can be selected out according to the comprehensive requirements of a plurality of people with factors in a plurality of aspects considered. The method is characterized in that the professional knowledge of a plurality of industries and specific requirements set by users are integrated. According to the technical schemes of the invention, industry knowledge and data about food in the plurality of aspects are integrated into a knowledge network, and the knowledge network and the requirements set by the users are jointly adopted as a selection condition, a dish or fruit bank is screened, scored and sorted. The method of the invention is mainly used for helping a family to select the best food and prompting a customer to modify dishes according to his or her own requirements.

The invention discloses a method and system for constructing a semantic knowledge base based on a graph neural network. The method comprises the steps of first large-scale corpus information is obtained; an object extractor based on the NLP technology performs text object recognition and extraction from the first large-scale corpus information to obtain a first text object set; an object level network is constructed; semantic element analysis is performed on each object in the first text object set according to the object hierarchical network to construct a first semantic element set; the first semantic element set is networked by using a graph neural network to obtain a first semantic knowledge network; reinforcement learning is performed on the first semantic knowledge network to obtain a second semantic knowledge network; and the second semantic knowledge network is output as a semantic knowledge base. The technical problems that in the prior art, construction of a multi-level semantic knowledge base from a large-scale corpus is not perfect, and the automation level is low are solved.

The invention discloses an intelligent association reply method and device based on artificial intelligence, and computer equipment. The method comprises the steps of performing structured processingon mined data information according to a structured processing rule to generate an associated knowledge network, adding the associated knowledge network to a knowledge graph, and obtaining received questioning information and a target main body node matched with historical response information associated with the questioning information from the knowledge graph; and if the target main body node and the questioning information meet the preset answering condition, obtaining data of the target main body node as first answering information to be fed back to the user terminal, and if the preset answering condition is not met, generating node questioning information or obtaining bottom answering information to be fed back to the user terminal. The invention is based on the intelligent decision technology, belongs to the field of artificial intelligence, obtains the matched target main body node in the knowledge graph according to the questioning information and the historical reply information associated with the questioning information, carries out the intelligent association reply according to the data of the target main body node, and can improve the accuracy of the reply to the questioning information.

The invention discloses a knowledge network construction method and system, and the method comprises the following steps: S1, extracting knowledge, carrying out the classification, and distributing types; S2, mutually connecting knowledge classifications in parallel after distribution, and constructing a knowledge network; S3, encoding and positioning knowledge classifications in construction, andperforming hierarchical separation on the knowledge classifications; S4, after the construction is completed, inputting the addresses into a server, and sorting the addresses; and S5, searching the sorted knowledge network addresses in the server through the network. The system is scientific and reasonable in structure and safe and convenient to use, knowledge can be better associated, a knowledge network is better constructed, the comprehensive effect of the knowledge network is improved, self-inspection is facilitated through the construction system, the error rate of the system is reduced,and the system is suitable for popularization and application.

The invention provides a complex design knowledge network construction method based on multi-source data mining, which comprises the following steps of: firstly, considering the design of a network top layer model, and aiming at the problems of poor consistency, weak availability, low sharing rate and the like of multi-source data of internal and external systems of a manufacturing enterprise, extracting entities and relationships from the multi-source data by using dependency syntactic analysis; and constructing a knowledge network top layer model. Before top nodes of a knowledge network are matched with design knowledge, the design knowledge needs to be organized and arranged, and the design knowledge is divided into different design knowledge items according to characteristics of multi-domain data to construct a design knowledge representation model. On the basis, the top layer model and the design knowledge are linked and matched, and a complex design knowledge network in a field is established. According to the method, a'unified expression-efficient organization-standardization 'system of design knowledge is designed, inspiration and help are provided for the design process by utilizing multi-domain data, design knowledge and clues are found for designers, a knowledge-driven design technology is realized, and the product innovation ability of manufacturing enterprises is improved.

The invention discloses a method for displaying related knowledge when a user browses or searches a product. A page for displaying the product also comprises one or more related knowledge display areas for supporting the user to know the related knowledge when the user browses or searches the product. The knowledge platform is integrated or connected into the e-commerce platform, the user is helped to better perform filtering and decision making, and the situation that the user in the platform is lost is reduced. At present, users often find related knowledge from other knowledge-based websites to assist filtering and decision making. In this case, the user may purchase or enter a purchase link directly at the knowledge website, so that the user is lost from the original platform. Due to the importance of related knowledge to users, a bidding mode is introduced in the invention. The administrator position of each knowledge point can open bidding. Product-related knowledge or purchase knowledge provided by a platform or provided by a third party independently is fairer and more comprehensive than knowledge provided by product merchants in product description.

The invention discloses a lightweight gesture recognition method based on deep learning, and belongs to the technical field of artificial intelligence. The method comprises the following steps: in a hand detection stage, providing an improved MobileNetv2-YOLOv3 network structure, and greatly reducing the parameter quantity and the calculation quantity of a model while ensuring the accuracy; in the hand key point detection stage, an attention-based selective gesture distillation method (Attention-based Selective HAND Distillation, ASHD) is provided, firstly, a knowledge network (Teaser Model, T) and a lightweight basic network (Student Model, S) which are high in expression ability and large in parameter quantity are designed, then structured knowledge of the knowledge network is selectively migrated through the attention-based gesture distillation method, and the hand key point is detected through the Attention-based Selective HAND Distillation method. Jointly training a basic network with a small number of parameters in combination with a real label; in the gesture classification stage, ResNet is adopted as a basic network, and a series of tricks are combined to improve the generalization ability of the model. The method is used for designing a lightweight model, the calculation amount is reduced while the accuracy is guaranteed, and the method can be deployed on embedded equipment with low calculation power requirements.

The invention provides a knowledge network construction method, device and equipment and a storage medium, and the method comprises the steps: dividing a text into three types of text corpora according to three levels of a first sentence length, a second sentence length and a paragraph and in combination with professional word list screening vocabularies; for any text corpus, listing co-appearing vocabulary pairs in the text corpus and carrying out column connection table hypothesis testing one by one, and then calculating a pvalue value of the hypothesis testing through chi-square testing to form a pvalue set; for any pvalue set, the pvalue threshold q is used for screening pvalue values meeting the requirements, and a correlation significant word pair set is formed; and according to the minimum word frequency, the vocabulary pair co-occurrence number threshold value and the minimum correlation, further screening the correlation significant vocabulary pair set. According to the method, from the aspects of paragraphs, long sentences, short sentences, segmented words and non-segmented words, hypothesis testing is used for judging correlation, knowledge networks under different dimensions are constructed, and the knowledge networks are evaluated according to various parameter indexes.