Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

117 results about "Data semantics" patented technology

Data semantics is the study of the meaning and use of specific pieces of data in computer programming and other areas that employ data. When studying a language, semantics refers to what individual words mean and what they mean when put together to form phrases or sentences. ... The creation of data semantics is similar to mapping out grammar and style rules that determine how words are used together to convey a specific meaning.

Multi-source heterogeneous data semantic integration model constructed based on domain ontology and method

The invention discloses a multi-source heterogeneous data semantic integrated model constructed based on domain ontology. The multi-source heterogeneous data semantic integration model comprises a local ontology constructing module, a domain ontology merging module and a semantic inquiring dynamic propagation and protocol module. A multi-source heterogeneous data semantic integrating method comprises the following steps: constructing the domain ontology through an ontology merging technique and establishing a semantic mapping relation between a data source and the local ontology as well as between the local ontology and the domain ontology; combining the complementary advantage of social label and ontology on knowledge representation and performing inquiring protocol and propagation for the semantic inquiring request of the user, thereby generating a formal semantic inquiring statement; respectively inquiring a plurality of data sources, performing duplicate removal and aggregation optimization on the inquiring result, and lastly feeding back to the user. The invention provides an aggregation heterogeneous data semantic integration model constructed based on domain ontology and the method through the construction and mapping of the domain ontology, semantic inquiring propagation and result aggregation optimization.
Owner:CHONGQING UNIVERSITY OF SCIENCE AND TECHNOLOGY

Multi-modal knowledge graph construction method

PendingCN112200317ARich knowledge typeThree-dimensional knowledge typeKnowledge representationSpecial data processing applicationsFeature extractionEngineering
The invention discloses a multi-modal knowledge graph construction method, and relates to the knowledge engineering technology in the field of big data. The method is realized through the following technical scheme: firstly, extracting multi-modal data semantic features based on a multi-modal data feature representation model, constructing a pre-training model-based data feature extraction model for texts, images, audios, videos and the like, and respectively finishing single-modal data semantic feature extraction; secondly, projecting different types of data into the same vector space for representation on the basis of unsupervised graph, attribute graph, heterogeneous graph embedding and other modes, so as to realize cross-modal multi-modal knowledge representation; on the basis of the above work, two maps needing to be fused and aligned are converted into vector representation forms respectively, then based on the obtained multi-modal knowledge representation, the mapping relation of entity pairs between knowledge maps is learned according to priori alignment data, multi-modal knowledge fusion disambiguation is completed, decoding and mapping to corresponding nodes in the knowledge maps are completed, and a fused new atlas, entities and attributes thereof are generated.
Owner:10TH RES INST OF CETC

Clinical-knowledge-graph construction method based on electronic medical records

The invention relates to a clinical-knowledge-graph construction method based on electronic medical records. The method is characterized by including the following steps: step 1, concept layer designing, step 2, instance layer learning, and step 3, knowledge graph updating. Compared with the prior art, the method is better in a construction effect. Construction of a graph is automatic, partial manpower is saved, for unique data features of medical data, professionalism, data semantics, connection, knowledge time sequences and the like of the medical data are maintained, updating of the clinical knowledge graph is supported, and subsequent use is facilitated.
Owner:DONGHUA UNIV

Data semantic processing method

The invention provides a data semantic processing method. The data semantic processing method can automatically obtain sematic classified information, form into dynamic analysis processing mode and achieve statement automatic classified processing which faces to a free text transformation format command and is based on semantic information. The data semantic processing method comprises dividing obtained texts to be statements and numbering obtained statements through dividing; constructing an LDA (Latent Dirichlet Allocation) model according to dividing; solving through a Gibbs algorithm to obtain theme distribution of statements and vocabulary distribution of themes in the LDA model; computing theme membership degree values of the statements; confirming categories to which the statements belong according to the theme membership degree values of the statements and outputting category results according to corresponding statement sequence numbers.
Owner:ACADEMY OF ARMORED FORCES ENG PLA

Cross-modal retrieval method based on multilayer semantic alignment

The invention discloses a cross-modal retrieval method based on multilayer semantic alignment, which comprises the following steps of: acquiring a remarkable fine-grained region by utilizing a self-attention mechanism, promoting the alignment of entities and relationships among modal data, providing an image text matching strategy based on semantic consistency, extracting semantic tags from a given text data set, and performing global semantic constraint through multi-label prediction, so that more accurate cross-modal association is obtained. Therefore, the problem of cross-modal data semantic gap is solved.
Owner:BEIFANG UNIV OF NATITIES

Systems, methods, and computer-readable media for using immutable and copy-on-write data semantics to optimize record and replay frameworks

Systems, methods, and computer-readable media are disclosed for using managed runtime environment semantics to optimize record and replay frameworks. One method includes: executing, by the computing system, a managed runtime component; interacting, by the computing system, with another system during the execution of the managed runtime component; determining, by the computing system, whether a non-deterministic event is to be logged in event logs during the execution of the managed runtime component; determining, by the computing system when the non-deterministic event is to be logged, whether semantics of the non-deterministic event to be logged includes a predetermined semantic, wherein the predetermined semantic includes one or more of optimistic zero copy actions; and logging, by the computing system when the semantics of the non-deterministic event includes the predetermined semantic, a copy of contents of the non-deterministic event in event logs during the execution of the managed runtime component.
Owner:MICROSOFT TECH LICENSING LLC

Social media large data subject discovering method based on knowledge map

The invention relates to a social media large data subject discovering method based on a knowledge map. The social media large data subject discovering is realized by four steps of knowledge map probability processing, social media large data semantic similarity calculating, subject fuzzy density clustering and weak information subject filtering. The knowledge map probability processing step is to complete confidence assessment and true probability value generation of a triad fact in the knowledge map. The social media large data semantic similarity calculating step is to achieve semantic mapping structure of each pair of documents and semantic similarity assessment among the documents based on an approximate map matching strategy. The subject fuzzy density clustering step is to obtain different subjects of the social media large data and automatically determine the number of the subjects. The weak information subject filtering step is to delete the subjects with less semantic strength and return an optimal subject list to a user. Compared with the prior art, the method has the advantages of being high in extendibility degree, strong in self-adapting capability, high in accuracy and the like, and being effectively used in the fields of social public safety, public health care, Internet depth information service, e-commerce and the like.
Owner:TONGJI UNIV

3D point cloud data semantic segmentation method based on deep learning and self-attention

ActiveCN110245709AAvoid the impact of segmentation accuracyHigh precisionImage enhancementImage analysisPoint cloudSelf attention
The invention provides a 3D point cloud data semantic segmentation method based on deep learning and a self-attention mechanism, which is used for solving the technical problem of low segmentation precision in the prior art, and comprises the following implementation steps of: (1) obtaining a training set and a verification set; (2) constructing a 3D point cloud data semantic segmentation network of deep learning and a self-attention mechanism; (3) setting a loss function required by the 3D point cloud data semantic segmentation network for training deep learning and a self-attention mechanism; (4) carrying out supervised training on the 3D point cloud data semantic segmentation network of deep learning and a self-attention mechanism; and (5) obtaining a semantic segmentation result of the 3D point cloud data test set. According to the method, the self-attention module is added to the deep learning network, the deep features containing the relations between the feature channels can be better extracted, and therefore the segmentation precision is improved.
Owner:XIDIAN UNIV

Integration method of dynamic heterogeneous space information plotting data

The invention provides an integration method of dynamic heterogeneous space information plotting data. The method mainly includes the steps of semantic ontology base building, data filtering, data conversion, fault tolerance management, space topology integration and plotting data integration base forming. Firstly, a dynamic heterogeneous plotting data semantic ontology base is built, a mapping rule in ontology is declared, precise meanings of concepts are determined, and the semantic heterogeneity problem is solved; whether metadata meet a required standard or not is judged, and the data are integrated and checked; then logical operation is conducted on data items, and derived data items are generated; according to a built fault tolerance rule base, incorrect data are saved and removed; meanwhile topological relationship processing is conducted on dynamic heterogeneous data elements, and eventually a dynamic heterogeneous plotting data integration base is formed. According to the method, complex data are made to be standardized, and workloads of data maintenance and interaction can be effectively reduced. Data content is defined and converted through metadata formats, and display of plotting data has expandability and flexibility.
Owner:BEIJING AEROSPACE TITAN TECH CO LTD

Moving path hybrid forecasting method oriented to data sparse environment

The invention provides a moving path hybrid forecasting method oriented to a data sparse environment. The moving path hybrid forecasting method comprises the steps of: acquiring mobile position data information; processing data , wherein data preprocessing and data semantic analysis are carried out on the data; constructing a semantic knowledge base, wherein original trajectory data is subjected to rich semantic transformation and fusion processing, so as to construct the semantic knowledge base; constructing a hybrid online prediction model which is based on the semantic knowledge base and established on the basis of forward pattern similarity degree matching calculation and a high-order Markov model; and outputting a predicted path, wherein a trajectory fragment to be predicted is input into the hybrid online prediction model for prediction, and the predicted path is output. The moving path hybrid forecasting method effectively overcomes the problem of pattern matching failure caused by data sparse condition, significantly improves the accuracy of path prediction, and satisfies the requirements on real-time performance, high efficiency, predictability and the like of mobile service application.
Owner:XIAN UNIV OF SCI & TECH

Method and device for sorting coding and integrated exchange and management of heterogeneous data of enterprise

The invention provides a method and device for sorting coding and integrated exchange and management of heterogeneous data of an enterprise. Due to the fact that the expressive forms of standard formats of heterogeneous data are unified and heterogeneous data acquiring middleware is programmed, remote and local heterogeneous data of the enterprise are acquired; the acquired heterogeneous data are clustered according to different modes and different topic attribute units, topic attribute units and modes are formed through clustering and can be used for sorting, mode regrouping is performed on the topic attribute units which can be used for sorting and formed through clustering according to sorting restraint, topic data formed by clustering are sorted by selecting one mode with restraint, and a sorting rule of intelligent heterogeneous data is derived; on the basis of an original information system of the enterprise, original codes are converted into new codes through an interface converting mode, and a new code system is gradually achieved through transition; a sharing and exchanging mechanism of the heterogeneous data is constructed based on a uniform service interface. The method and device solve the problems that data are not consistent, data semantic conflict is caused, and a large number of data are in redundancy.
Owner:JIANGXI UNIV OF SCI & TECH

Data integration using conceptual modeling framework

A computer implemented system and method that combines data from multiple sources to provide a unified view of data using multi-layered conceptual modeling framework. The invention provides means to model conceptual data models at multiple levels of abstraction and a means to map them with each other and with physical models. A query rewriting approach adopted in the present invention is capable of translating queries down multiple levels of conceptual models into physical models and Conceptual modeling framework provide a means to capture and process semantic intent of data more accurately thereby reducing data integration errors significantly.
Owner:TATA CONSULTANCY SERVICES LTD

Cross-modal video retrieval method and system based on multi-head self-attention mechanism and storage medium

The invention provides a cross-modal video retrieval method and system based on a multi-head self-attention mechanism and a storage medium, and the cross-modal video retrieval method comprises a videocoding step, a text coding step and a joint embedding step. Semantic information in the training multi-modal data is fully utilized for training, a multi-head self-attention mechanism is introduced,fine interaction in videos and texts is captured, key information of the multi-modal data is selectively concerned to enhance the characterization capability of the model, data semantics are better mined, and the invention has the advantages of being high in practicability and easy to popularize. Consistency of the distances of the data in the original space and the shared subspace is ensured. Theinvention has the beneficial effects that experiments prove that the similarity of the data in the original space can be effectively maintained, and the retrieval accuracy can be improved.
Owner:HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL

Implementation method for integrated exchange platform

The invention discloses an implementation method for an integrated exchange platform. The implementation method includes the following steps that the integrated exchange platform independent from service applications is established; standard data formats inside the platform are defined through the integrated exchange platform; different information formats and grammars are reconverted into data formats and semantics capable of being understood by a target application system through data format conversion and data semantic conversion; all service requests are sent to the integrated exchange platform in a centralized mode, then the requests are recognized, a provider and a request forwarding route of services are determined through a route scheduling module arranged in the integrated exchange platform, and the service requests are forwarded to the actual service provider. Compared with the prior art, by means of the implementation method for the integrated exchange platform, multiple independent technologies are integrated to achieve the function of maintaining stable and efficient operation of a whole system, good compatibility is achieved, and the method is suitable for automatically acquiring data of multiple data sources, high in practicality and easy to popularize.
Owner:INSPUR COMMON SOFTWARE

Data mining method based on multi-source heterogeneous patent data semantic integration

The invention discloses a data mining method based on multi-source heterogeneous patent data semantic integration. The data mining method sequentially comprises the following steps that (1) global patent data and inter-translation dictionaries are taken as data sources, and ontology study is conducted, and a patent global ontology is constructed; (2) according to corresponding attributes, a function used for judging the individual information similarity degree in each data source is constructed; (3) according to the constructed similarity degree functions, similarity information of individuals in different data sources is calculated under the guidance of the patent global ontology; (4) data mining is conducted according to the similarity information. According to the data mining method based on multi-source heterogeneous patent data semantic integration, the extracting precision of concepts and concept relations is greatly improved, patent global ontology study is achieved, and data mining efficiency and accuracy degree are greatly improved.
Owner:肖冬梅

Method for transmitting transactional commands and data between computer networks

The present invention relates to a method for transmitting transactional commands and data between computer networks, and in particular a method for securely transmitting transactional commands and data between two computer networks by encoding the transaction based upon the ontology and semantics of the transaction commands. The invention establishes the process in which transactional commands are decomposed into context less text, based upon a local representation of the transaction command ontology and semantics. The invention also establishes the process for authenticating the transaction using deep knowledge of the user and business rules for the transaction. The decomposition of the transaction command and data is achieved using command semantic knowledge objects, data semantic knowledge objects and a decision cube populated as a result of an examination of the transaction command set ontology and semantics. The resultant transmission consists of a set of indexes or tags and context less data fields.
Owner:BOWEN DAVID

Method and device for integrating extracted web table data

The invention discloses a method and device for integrating extracted web table data. The method and device for integrating extracted web table data can improve accuracy of a database formed by integrating the web table data. The method comprises the steps of obtaining a web table corpus; obtaining candidate semantics of each table; calculating inconsistency between the candidate semantics of each table and data semantics; if the inconsistency is larger than a first predetermined threshold, indicating that the candidate semantics of the table is incorrect, utilizing crowdsourcing to confirm the candidate semantics of the table, re-calculating the candidate semantics of the table according to semantic likelihood of the table and crowdsourcing feedback results, and re-calculating the inconsistency; if the inconsistency is larger than a second predetermined threshold and smaller than the first predetermined threshold, utilizing a knowledge base and a crowdsourcing mode to confirm the correctness of data in the table, and re-calculating the inconsistency; if the inconsistency is smaller than the second predetermined threshold, indicating that the candidate semantics of the table is correct, and labeling data in the table; and when it is determined that candidate semantics of all tables is correct, performing mode mapping and data cleaning.
Owner:江苏省现代企业信息化应用支撑软件工程技术研发中心

Print Data Semantic Analyzer

A print data semantic analysis system may comprise a print data handler and a database. The print data handler may include a semantic analyzer and a computer. The semantic analyzer may include a semantic identifier and a semantic operator. Semantic pattern(s), rule(s), and / or semantic category or categories may be stored in mutually associated fashion at the database. The print data handler may be a printer driver having a page description language (PDL) generator. The print data handler may be a raster image processor (RIP), a print server, or a printer having a PDL interpreter. The database may take the form of source code, function(s), lookup or other such table(s), and / or any other suitable format(s). The database may be a function table containing function(s) incorporating the semantic pattern(s), rule(s), and / or semantic category or categories.
Owner:KYOCERA DOCUMENT SOLUTIONS INC

Semantic-based multi-modal spatio-temporal data association method

ActiveCN110647662AEffective associationSolve the problem that it is difficult to understand the semantic content information of dataOther databases indexingSpecial data processing applicationsSpatial information systemsData link
The invention discloses a semantic-based multi-modal spatio-temporal data association method, and belongs to the technical field of geographic space information systems. According to the method, the constructed ontology model is used for carrying out feature-semantic mapping and conceptualized semantic expression on the data, complex semantic information of the multi-modal spatio-temporal data iseffectively described in a unified format, and the problem that a computer is difficult to understand semantic content information of the data is solved; meanwhile, based on ontology semantic expression of the spatio-temporal data, time association, space association and content object semantic association between the spatio-temporal data are established through three factors of time, space and objects contained in data content, and association between the multi-modal spatio-temporal data is effectively carried out. The method is mainly used for cross-modal retrieval of multi-modal spatio-temporal data, an accurate and comprehensive data link association processing result is provided for a user, and then the cross-modal retrieval performance is improved.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Video natural language text retrieval method based on space time sequence characteristics

The invention relates to a video text retrieval method based on spatial time sequence characteristics, which comprises the following steps of: performing hierarchical fine-grained comprehensive video unified representation on spatial time sequence semantic information of a video by utilizing three different types of neural networks; and constructing a video text public semantic embedding network to fit a semantic gap of cross-modal data, and training the network by using a comparison sorting loss function. The method can be used for mutual retrieval of video natural language texts, the layered feature extraction method fully excavates more discriminative complex space-time semantic information of video modal data, and the video text public semantic embedding network effectively learns public space feature representation with different modal heterogeneous data semantic feature identical distribution. The semantic association between the high-order features of the video and the natural language text is accurately measured through public space feature representation, and the retrieval precision of the video natural language text is improved.
Owner:XIDIAN UNIV

Method for information recommendation in could environment based on data semantics

The invention relates to a method for information recommendation in a could environment based on data semantics. The method achieves the information recommendation through a basic data and user preference information semantization module, a could environment mass semantic information indexing module and a semantic calculation based information recommendation module to achieve the information recommendation, wherein the basic data and user preference information semantization module obtains basic data and user preference information through a cloud platform and conducts semantization description on the basic data and the user preference information to form a basic data and user preference information body library, the could environment mass semantic information indexing module establishes an indexing structure for semantic information and performs index division and regrouping when indexing nodes are overload, the semantic calculation based information recommendation module conducts semantic calculation on the basic data and user preference information body to obtain an information recommendation result. Compared with the prior art, the method for information recommendation in the could environment based on data semantics has the advantages of being good in real-timeliness, high in robustness and recommendation quality and the like.
Owner:TONGJI UNIV

Data virtualization and query method and device

The invention discloses a data virtualization and query method and device, and the method comprises the steps: collecting multiple kinds of data from a database, and building a data dictionary according to the data and attributes of the data; Obtaining vocabulary information in the common word library dictionary and the industry term lexicon; And constructing a data semantic layer according to thevocabulary information, the data in the data dictionary and the mapping relationship thereof. According to the invention, The data dispersed in the enterprise is virtualized, unified data access based on virtualization definition is provided for a user, and the user does not need to pay attention to where the data exist, format requirements and the like, so that the data query difficulty is reduced, and operation such as statement analysis and query can be conveniently carried out when a subsequent user carries out statement query.
Owner:湖南亚信软件有限公司

Method for analyzing sparse semantic relationship by combining BiLSTM-CRF algorithm and R-BERT algorithm

The invention provides a method for analyzing a sparse semantic relationship by combining a BiLSTM-CRF algorithm and an R-BERT algorithm, which comprises the following steps: acquiring text data of emerging industries through a web crawler, and performing semi-supervised annotation on the text data; preprocessing the labeled text data, and constructing a training data set and a verification data set; training a BiLSTM-CRF algorithm model and an R-BERT algorithm model according to the training data set and the verification data set; extracting entities contained in the text data to be predictedthrough the trained BiLSTM-CRF algorithm model; predicting the relationship between the text data to be predicted and the entities through the trained RBERT algorithm model, and establishing relationship connection between the related entities; and extracting the triad pair of the semantic relationship of the text data to be tested according to the established relationship connection, and completing semantic analysis of the text data to be tested. The invention provides a high-precision semantic relation extraction method for information extraction of an unstructured text, a BiLSTM-CRF algorithm model is used for extracting required entities in the text, and the relation between the text and the extracted entities is predicted through an R-BERT model.
Owner:江苏网谱数据科技有限公司

Method for establishing high-efficient semantic indexing for large-amount RDF (resource description framework) data

The invention discloses a method for establishing high-efficient semantic indexing for large-amount RDF (resource description framework) data. The method comprises the following steps: step 1, configuring an open-source distributive RDF database to be used as a duration database for storing the RDF data; step 2, distinguishing TBox data and ABox data in the RDF database; step 3, generating a child-parent semantic relation indexing among categories in TBox data; step 4, generating child-parent semantic relation indexing among attributes in TBox data; step 5, incorporating the generated semantic relations into the RDF data including the original TBox data and the ABox data to form novel RDF data; step 6, persisting the novel generated RDF data into the well-configured RDF database. For inquiring and reasoning a large amount of RDF data, the novel scheme for establishing the RDF data semantic relation indexing is finally provided, so that the inquiring efficiency is guaranteed, and meanwhile rich offline reasoning can be supported.
Owner:TIANJIN UNIV

Sensor system used for building-structure safety monitoring

InactiveCN103442078AChange sensor typeChange typeTransmissionSensing dataWeb service
The invention proposes a sensor semantic system including sensor nodes, a local-area network, a monitoring-center server, the Internet and a text-message alarm module. The sensor nodes are arranged on parts to be detected and connected with the monitoring-center server through the local-area network. A built-in web server of the monitoring-center server provides data services or sends forewarning information to remote users through the Internet and forewarning text messages can also be sent through the text-message alarm module. The system is characterized in that the sensor nodes include a sensor information semantic module, a detected-point information semantic module and a sensed-data semantic module and the three semantic modules are associated through attributes and add a semantic background for data output by the sensor. An RDF (Resource Description Framework) is used to describe the three semantic modules and sensor information, detected-point information and sensed data are defined as resources on networks through a URI (Universal Resource Identifier). The sensed-data semantic module carries out semantic annotation on data so that semantic data are formed and then stored. When a storage space exceeds a limit, data saved at earliest time are covered automatically and at the same time, the semantic data are submitted to the monitoring-center server through a server interface module.
Owner:BEIJING UNIV OF TECH +1

Configurable business process analyzing method based on data flow constraint

InactiveCN106169111ASolve the problem of data flow constraint relationshipAvoiding Unusual Behavior of Data SemanticsResourcesPersonalizationValidation methods
The invention discloses a configurable business process analysis method based on data flow constraints. First, according to a data flow model between related activities in a specific field, the execution relationship between the activities is determined, and a CPN-based business process model is designed; Then, the data flow model and the business process model are combined to form a business process model reflecting the data flow relationship, which is converted into a configurable business process model according to the needs and guidance of the field; and then the configurable business is analyzed and verified according to the personalized needs in the field. The correctness of the data flow constraint of the process; finally, the configurable business process model can be adjusted according to the verification result, so that the business process can maintain the correctness of the design during the configuration process. The method of the present invention proposes the data flow constraint of the configurable business process model The analysis and verification method effectively solves the data flow constraint relationship in business process variability management and avoids abnormal behavior of data semantics in the process of business process configuration.
Owner:TONGREN UNIV

Method for converting XBRL data into OWL data based on semantics

The invention discloses a method for converting XBRL data into OWL data based on semantics. The method for converting the XBRL data into the OWL data based on semantics includes the steps that based on the semantic net technology, an XBRL financial report and fact data semantic meta model is established, an XBRL classification standard is analyzed to establish an XML information set, an XBRL instance document is analyzed to establish an instance document information set, a conversion algorithm from the XBRL classification standard to OWL ontology is designed, and a conversion algorithm from the XBRL instance document to the OWL ontology is designed. Data formats can be converted on the semantic level, and the mapping problem of XBRL semantic heterogeneity is solved.
Owner:UNIVERSITY OF CHINESE ACADEMY OF SCIENCES
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products