63 results about "Relational knowledge" patented technology

A method and systems for information security threat identification, management, and analysis, including identifying and managing threats posed by senders of unsolicited e-mail, pirates, hackers, and virus-spreaders. Methods are provided for identifying and facilitating legal action against a sender of unsolicited e-mail. A secure evidence repository can be used for storing copies of and information regarding unsolicited e-mails in a forensically sound manner. A relational knowledge database can be used for storing copies of and information regarding unsolicited e-mails such that the information can be queried, manipulated, or analyzed.

The invention relates to a character relationship graph construction method based on integration of an ontology and multiple neural networks. The method comprises the following steps: crawling data related to a character in a certain domain in the Internet; establishing a domain character ontology; extracting data from a structured data table which contains multiple types of entities and has repeated entities to construct a standardized entity table; matching the two class names of the character ontology model with the two entity table names through a semantic mapping algorithm, automaticallyobtaining all entity relationships, and storing the entity relationships in a Neo4j database in a graph structure; for the text data in the structured table, carrying out character entity recognitionand relationship extraction by using a sliding window, entity position characteristics and a bidirectional gating recurrent neural network; and updating the current graph structure of the newly addedrelationship to form a domain character relationship knowledge graph. The character relationship advanced features can be extracted from the original relational data and the text data, manual design is not needed, the recognition effect is improved, and the efficiency of constructing the character relationship graph by the complex webpage text is improved.

The invention relates to a rapid learning method for a large-scale knowledge base. The rapid learning method comprises the following steps: S1, carrying out the knowledge extraction: carrying out the extraction of knowledge data from heterogeneous data, and enabling the knowledge data to be a fact triple which comprises an entity and a relationship; S2, knowledge representation: constructing a TransH model, and embedding all entities and relationships in the knowledge data into a low-dimensional vector space by utilizing the TransH model so as to realize knowledge representation of the knowledge data; S3, knowledge fusion: fusing the knowledge data represented by the knowledge in the step S2 through an RLCD model, and storing the fused knowledge data into a knowledge map; and S4, knowledge reasoning: extracting the knowledge data fused in the step S3 from the knowledge map for reasoning, and storing the knowledge data after reasoning into the knowledge map; and obtaining updated knowledge and conclusions. According to the invention, rapid learning of the knowledge base can be realized.

The invention discloses a method utilizing Chinese online resources for supervising extraction of character relations remotely. According to the method, at first, an online encyclopedia website, formed through a semi-manual mode, on a web is utilized for automatically constructing a knowledge base so as to obtain accurate relation types comprehensive as much as possible, and examples of the character relations; then co-occurring names and context features are extracted from a text corpus, and the names and the relation examples in the knowledge base are matched to obtain name pair sets of the marked relations and name pair sets of unmarked relations; finally, a label propagation algorithm is introduced to achieve relation match of unmarked name pairs, so that extraction of the character relations is achieved. According to the method, the knowledge base of the character relations can be automatically constructed, and the richer and more accurate relation types are included; based on the knowledge base, the label propagation algorithm is introduced to supervise extraction of the character relations remotely, and therefore accuracy of results of the extracted relations can be ensured.

The invention discloses a power grid fault intelligent processing system based on a knowledge graph. A fault processing plan analysis module of the system is used for obtaining a fault processing entity and relation knowledge; the graph generation module is used for storing the identified entities and entity relationships in a triple form to obtain an equipment entity knowledge graph, an accident plan knowledge graph and a disposal flow knowledge graph; the fault sensing module is used for intelligently sensing and collecting information related to power grid faults in a power grid system; the fault risk assessment module is used for carrying out risk identification; and the fault intelligent processing module is used for reasoning a processing measure suitable for the current fault based on the accident plan knowledge graph and the processing flow knowledge graph in combination with the topological change, the power flow and the alternating current power supply frequency before and after the power grid fault after the power grid has the line fault. According to the invention, comprehensive intelligent alarm information can be sensed in all directions, and scheduling personnel are guided to carry out intelligent disposal on power grid risks and power supply restoration based on the established knowledge graph.

The invention discloses an oracle bone knowledge graph construction method and system based on multi-modal data. The construction method comprises the following steps of S1, performing entity extraction and relationship extraction on oracle bone literature, an oracle bone database, an oracle bone text, an oracle bone text image and an oracle bone text ontology; s2, performing entity alignment andrelationship fusion to form a multi-modal fusion map; s3, mining and obtaining an implicit semantic relationship by utilizing ontology reasoning and rule reasoning; and S4, expanding the implicit semantic relationship into the multi-modal fusion map, and finally forming the oracle bone knowledge map. According to the method, various forms of oracle literature, description, database, corpus, video,image, 3D model and the like reflect multi-modal characteristics, knowledge fusion is performed on entities and relationships, and oracle knowledge maps are constructed by mining implicit semantics,so that potential entities and relationships hidden behind explicit entities and relationships are discovered, and the knowledge association is strong.

The invention discloses a method for constructing an articulation relationship knowledge graph, which is executed in computing equipment and comprises the following steps of obtaining financial statements of a plurality of companies, and extracting subject vectors of each company in different periods from the financial statements, the subject vectors comprising numerical values of a plurality of subjects, taking each subject as a dependent variable, taking other subjects as independent variables, and taking the plurality of subject vectors as training samples to train and generate a regressionmodel of each subject, and calculating an error of each regression model, and when the error is smaller than a preset error threshold, storing the corresponding regression model as an articulation relationship formula to generate an articulation relationship knowledge graph, the generated articulation relationship knowledge graph being used for checking an articulation relationship between financial statement subjects. The invention further discloses a corresponding financial statement checking method and computing equipment.

Disclosed within is a spatial knowledge extractor and a method of extracting spatial knowledge. The spatial knowledge extractor constructs an R-tree index, which is a minimum bounding rectangle (MBR)-based tree data structure, for geometric data about a plurality of spatial objects and extracts topological relation knowledge and directional relation knowledge about the spatial objects using MBRs constructed by the R-tree index and central points of the MBRs to extract spatial knowledge from geometric data about spaces.

The invention discloses a knowledge management method based on access control and intelligent retrieval. A database is used for realizing the storage of knowledge and files contained in the knowledge, classification is carried out according to knowledge content, meanwhile, an access range or specific browsing users authorized by the knowledge are set according to the target browsing user group and the security level of the knowledge, and finally, knowledge release is carried out after auditing is carried out; and when the knowledge browsing user logs in, the knowledge browsing user inputs a keyword to retrieve relational knowledge in virtue of a search engine and then applies an application algorithm routine to carry out precise matching filtering according to an authorization range of the retrieved knowledge and an organization to which the browsing user belongs, and finally, an authorized knowledge retrieval result is returned to the user so as to realize the storage, the retrieval and the access control of the knowledge information and guarantee knowledge accumulation and safe access. The knowledge management method based on access control and intelligent retrieval provides efficient and reliable knowledge storage, classification, retrieval and access control.

The invention discloses a network social relationship knowledge graph generation method and system based on artificial intelligence. The method comprises an acquisition step, an extraction step, an intersection step, an intersection information acquisition step, a relationship generation step, a knowledge graph generation step and an entity acquisition step. According to the method and the system,through a network social relation knowledge graph generation technology based on artificial intelligence and big data, the intelligence and high efficiency of network social relation knowledge graphgeneration are improved.

The invention discloses a rolling bearing fault diagnosis method and system based on relational knowledge distillation, and belongs to the technical field of fault diagnosis. After the original vibration signals of the bearing are collected, a time-frequency diagram is constructed for each processing sample to serve as a fault sample, the fault sample serves as input of a fault diagnosis system, and due to the fact that the time-frequency diagram contains complete time-frequency information of the vibration signals, the real-time response efficiency and accuracy of fault diagnosis are improved. A student model is adopted to simultaneously learn a multivariate relationship between the output soft label of Softmax of a teacher model and output of a plurality of samples in the last pooling layer, namely, a student network learns from two aspects of a teacher structure and output of a single sample in the teacher network; and the classification performance of the fault diagnosis system is effectively improved under the condition that the memory and the training time are not increased. According to the invention, bearing fault diagnosis is realized by using a relational knowledge distillation transfer learning method, and the calculation complexity is effectively reduced through the idea of replacing a large model with a small model.

The invention belongs to the technical field of computers and information services, and relates to a knowledge graph construction method and system for closed switch equipment. The method comprises the following steps: identifying text data from a scanned image of a closed switch equipment quality inspection report; performing text labeling on the recognized text information described by the equipment, and labeling a triple relationship (entity, relationship and entity attribute) describing equipment attribute in each sentence; constructing a relation triple extraction model based on BERT; based on the trained triple extraction model, carrying out the relation extraction on the test sample, obtaining a triple relation describing equipment attributes, constructing a relation knowledge graph, comparing the relation knowledge graph of the test sample with a standard equipment relation knowledge graph, and achieving equipment inspection. The method can be widely applied to the fields of character extraction, naming recognition, relation extraction, knowledge graph construction and the like of scanned images.

The invention provides an information recommendation method and device based on user investment and financing preferences. The method comprises the steps of generating investment and financing preference data of a user according to historical investment and financing data and behavior data of the user and current position data; generating an article label according to the investment and financingpreference data by using a domain knowledge graph method and a natural language processing method; generating a matching retrieval rule according to the article label; and recommending information tothe user according to the matching retrieval rule. According to the method, the product relationship knowledge graph is introduced, so that information recommendation can more deeply understand investment and financing related products and related information thereof.

The invention provides a poison-target literature knowledge mining method and system based on network crawling. The poison-target literature knowledge mining method comprises the steps of: acquiring poison and target data information, and processing the poison and target data information, so as to establish a comprehensive data set; developing a web crawler tool; crawling poison and target literature text information by utilizing the web crawler tool based on the comprehensive data set, and processing the poison and target literature text information to establish a literature text database; based on the literature text database, determining a poison-target potential action relationship by utilizing a natural language processing technology to form a poison-target relationship knowledge base; and performing poison-target literature knowledge mining by utilizing the literature text database and the poison-target relationship knowledge base. The poison-target literature knowledge mining method and the system based on network crawling are high in efficiency, good in accuracy and high in intelligent degree.

When performing semantic search for related clinical concepts based on their semantic meaning, a semantic search engine framework employs a client web interface (12) (e.g., a browser) that receives keyword search information (14), and a web server (16) that communicates with a semantic analysis engine (18). A natural language processing (NLP) engine (server) (22) receives and stores clinical notes and information (24), extracts clinical concepts from the clinical notes and stores them in a NoSQL database (26). The NLP engine converts unstructured free text notes into structured actionable data. The semantic analysis engine analyses user-entered keywords, maps them to a UMLS concept, and identifies related concepts based on one or more relational knowledge sources. The web server searches for the related concept IDs in the NoSQL database and returns a list (28) of relevant notes and concepts for display to, and selection by, the user.

Predefined relation items on paths traversing predefined entities of a knowledge base are collected and feature sets are assembled from the collected relation items. A classifier is computed for the feature sets and a relation score of a query pair of the entities is computed using the classifier.

The embodiment of the invention discloses a data query method and device, equipment and a medium. The method comprises the steps of obtaining a target query formula input by a user, and determining at least one query object and an associated query condition in the target query formula, obtaining at least one original query data table matched with the query object, and searching whether an associated data table meeting the associated query condition with each original query data table exists in a data table association relationship knowledge base, according to the search result, generating at least one data table association structure pair, and querying in the data table structure knowledge base to obtain a joint query field corresponding to each data table association structure pair, and according to each joint query field, generating a target query rule corresponding to the target query formula, and querying to obtain query result data matched with the target query rule. According to the technical scheme, the personalized data query rule is formulated, and the data query efficiency is improved.

The embodiment of the invention provides a text processing method and device. The method comprises the steps of receiving a to-be-processed text, inputting the to-be-processed text into an entity recognition model, and obtaining a candidate text with an entity recognition tag, inputting the candidate text with the entity recognition tag into an entity discrimination model to obtain a candidate entity of the candidate text, and determining an alternative text based on the candidate entity, constructing a relational knowledge graph based on the candidate entities and the alternative texts, calculating the similarity between each relational knowledge graph node based on the relational knowledge graph, and determining a target entity relationship, and determining a target entity text in a preset knowledge base based on the target entity relationship.

The invention provides a pension subsidy policy matching method and system based on a knowledge graph. The method comprises the steps that policy information is crawled and a policy library is dynamically and incrementally updated; in combination with the previously collected and accumulated pension related policies and the newly crawled pension related policies, a data set of pension field text classification is constructed, and a text classification model is trained; based on the pension related policies screened out by the trained text classification model, policy text paragraph-level structuring and vocabulary-level structuring of policy subsidy objects in a quintuple are completed, and a term vocabulary relation knowledge graph in the pension field is constructed by utilizing vocabulary data; and the user carries out matching in the term vocabulary relationship knowledge graph policy library. According to the method, the policies related to the pension service are screened out, then the pension service policies are subjected to structured processing based on the knowledge graph technology, important information in the policies is extracted, and the user can conveniently inquire and push the subsidy policies which can be enjoyed by the user in a personalized mode.

The invention provides a correlation degree determination method and device. The method comprises the steps: constructing a fund flow direction relation knowledge graph based on financial products andcorresponding fund flow; and determining the association degree corresponding to each financial product according to a fund flow direction relationship knowledge graph. According to the application,the association degree of financial products can be determined by using the fund flow relationship graph and the graph calculation technology based on fund flow.

The invention discloses an electromagnetic target classification method based on knowledge vector embedding, and the method comprises the steps: building a graph structure of an electromagnetic target through the data of known electromagnetic target information, and carrying out the embedded vector representation of a graph node corresponding to each electromagnetic target type based on a graph neural network; the method comprises the following steps: acquiring an electromagnetic target signal, performing short-time Fourier transform on electromagnetic target data to obtain time-frequency data of the electromagnetic target data, and preprocessing the time-frequency data to serve as a sample for training a convolutional neural network; constructing a convolutional neural network, training the convolutional neural network based on a result represented by the embedded vector of the graph node corresponding to the electromagnetic target category, and finally obtaining a reference vector for subsequently classifying and identifying the acquired electromagnetic target signal; and the acquired electromagnetic target signal is classified and identified by using the obtained reference vector. The method is high in applicability, the category relationship knowledge is combined into network training, and the defect that a traditional classification network is only suitable for recognizing categories appearing in a training set is overcome.

The invention discloses a relation extraction method suitable for small samples. The relation extraction method comprises the following steps: (1) obtaining training data; (2) training a general domain relation knowledge model; and (3) training a specific domain relation extraction model. Common knowledge contained in various relationships is obtained by utilizing a general domain relationship knowledge module, and samples are automatically generated based on remote supervision by utilizing an open-source knowledge graph and combining with unsupervised noise reduction data to train relationship knowledge models of general and specific domains; a general domain relation knowledge module is adopted to learn general knowledge contained in various relations; training samples are automaticallygenerated on the basis of remote supervision, and noise data is reduced in combination with unsupervised data, so that manual annotation data is reduced; when the relation knowledge model is generated, a large amount of manual marking data does not need to be obtained, time and money consumption caused by a large amount of manual marking is avoided, and a relation extraction task in a specific field can be completed through a small amount of marking data in the specific field.

The invention discloses a multi-stage unsupervised domain adaptive causal relationship identification method. The method comprises the following steps of (1) dividing a data set, (2) carrying out pre-training by utilizing self-adaptive contrast learning, (3) performing adversarial learning in combination with knowledge distillation, (4) filtering multi-stage data to obtain a seed set, (5) filtering single-stage data to obtain a pseudo label set, (6) obtaining a subclass prototype by using a k-means clustering method, (7) enhancing introduction consistency by adopting feature-level data, and (8) carrying out self-training by using the filtered pseudo labels. According to the method, rich causal relationship knowledge in a source domain is obtained through self-adaptive comparative learning, then the source domain knowledge is migrated to a target domain in a knowledge distillation and adversarial learning mode, and multi-stage subclass prototypes and false labels of unlabeled samples are obtained through data filtering. Feature-level data enhancement is conducted by using a prototype so as to introduce consistency loss, and self-training is conducted in a target domain by using a pseudo label.

The invention relates to a character relation visual query method and system based on a knowledge graph. The method comprises the following steps: crawling webpage information related to a current character from the Internet by using a Scrapy framework; analyzing the webpage information, and extracting character information and relation information related to the current character; normalizing thecharacter information and the relation information, and constructing quintuple data of all characters related to the current character; performing character relation backward reasoning on the quintuple data to obtain backward quintuple data; constructing a character relation knowledge graph according to the quintuple data, and storing the character relation knowledge graph in a graph database; and obtaining a character relation knowledge graph corresponding to the to-be-queried character from the graph database and carrying out visual display. The character relation knowledge graph is storedin the graph database, so that the query speed is increased. In addition, knowledge in the graph is displayed through a visualization technology, a sense of visual impact is generated, and the character relation can be better understood.

The invention discloses a knowledge expression method, device and system oriented to a medical information system. The method comprises the following steps: acquiring multiple paths of data sources related to medical data; performing data core description on the multi-path data source to obtain each group of standard data elements of the multi-path data source; performing keyword recognition processing on the multi-path data source, and extracting a plurality of keywords from the multi-path data source; obtaining each domain factor corresponding to each keyword based on the standard data element; performing knowledge extraction by utilizing each keyword and each domain factor to obtain entity knowledge and relational knowledge; according to the relationship between the entity knowledge, performing knowledge combination on the entity knowledge and the relationship knowledge to generate a knowledge graph of the medical data; and performing knowledge reasoning on the knowledge graph to obtain a knowledge reasoning result. By adopting the embodiment of the invention, the data standard of the medical information system is standardized, and the unification of the standard of the medical information system and the association of the system can be ensured.

The invention relates to the field of artificial intelligence, and discloses a gene disease relationship knowledge base construction method and device and computer equipment, and the method comprisesthe steps: carrying out the dependency relationship analysis of a specified number of natural statements, and obtaining a dependency relationship; determining a path descriptor of the natural statement according to the dependency relationship; generating a rule template according to the path descriptor, and establishing a rule template library; and carrying out knowledge extraction on total medical literature by utilizing the rule template to obtain a gene disease relationship, and establishing a gene disease relationship knowledge base. According to the method, a large number of rule templates can be automatically learned, then the rule templates are used for automatically extracting the relationship knowledge of the gene diseases from the medical literature, high labor cost is not needed, the number of extracted knowledge is large, the extraction effect is good, good mobility is achieved, and the method can be used for extracting the relationship between more medical entities. The invention further relates to a blockchain technology. The rule template, the gene disease relationship knowledge base and the like are stored in a blockchain.

The invention provides an enterprise association relationship identification method and device, and relates to the technical field of big data. The method comprises the following steps: establishing each fund flow direction path according to enterprise associated data and fund flow directions, wherein each fund flow path comprises at least N objects, and N is an odd number greater than or equal to 3, wherein the enterprise associated data is obtained in advance, and the object is an enterprise or an individual; according to each fund flow direction path and an association relationship confirmation rule, obtaining each enterprise with an association relationship, and determining an association relationship type according to the enterprise association data; and drawing an enterprise association relationship knowledge graph based on each enterprise with the association relationship and the corresponding association relationship type. The device is used for executing the method. According to the enterprise association relationship identification method and device provided by the embodiment of the invention, the accuracy and comprehensiveness of the enterprise association relationship knowledge graph are improved.

The invention provides a target object determination method and device and a server. Based on the method, when a potential customer object needs to be searched for to push a target service product, a corresponding relational knowledge graph can be firstly obtained and constructed according to relational data of a first object; screening the first objects based on the relational knowledge graph, and finding out a second object meeting a preset first requirement; next, a preset potential object prediction model can be called to process the individual data of the second object so as to obtain a prediction result of the second object; According to a prediction result of the second object, screening the second object, and finding out a target object meeting a preset second requirement as a potential customer object so as to push a target service product. Therefore, by comprehensively utilizing the relation data and the individual data of the first object, the potential customer object with a relatively high probability of accepting the target service product can be efficiently and accurately screened out from the first object.

The invention discloses a privacy protection link prediction method and system based on mail data. The method comprises the following steps: constructing a character relationship knowledge graph by using the mail data; using the generative adversarial network to train a generative model for distribution of training data for learning; reconstructing the multivariate relation data so as to obfuscate sensitive and non-sensitive relation information implied in the data; the reconstructed multivariate relation data is used for complementing the relation between the entities, and the sensitive relation between the entities is protected while the non-sensitive relation between the entities is complemented. The invention further provides a privacy protection link prediction system based on the mail data, and the privacy protection link prediction method is implemented by the privacy protection link prediction system. According to the method, the relationship between the entities is complemented by using the reconstructed multivariate relationship data, so that the purpose of protecting the sensitive relationship between the entities while the non-sensitive relationship between the entities is complemented is achieved, and the technical problem that the social relationship of personnel in a mail system cannot be protected in the existing link prediction technology is solved.