53 results about "Rdf graph" patented technology

Disclosed are a method, information processing system, and computer program storage product for managing Resource Description Framework (“RDF”) graphs in a distributed system. At least one modification request for modifying an RDF graph in a shared data store is received. The modification request includes a query set, an expected result set, and a change set associated. The query set is a set of preconditions for guaranteeing a state of the RDF graph before executing the change set. The expected result set is results expected by the query. The change set is at least one statement that changes data elements in the RDF graph. The query is applied to the RDF graph at the shared data store. A result of the query is analyzed. A transaction request, which includes the change set, associated with the modification request is sent to a server information processing system communicatively coupled to the shared data storage.

A method and system of machine-based extraction of information from a text document employ the steps of performing semantic / syntactic analysis for sentences of the document to build semantic-syntactic structures of the sentences; applying production rules to the semantic-syntactic structures to generate a set of logical conclusions about objects of information comprised in the document, wherein the production rules are based on linguistic characteristics and lexical-morphological properties of the semantic-syntactic structures and ontologies of subject matters of the sentences; and using the set of logical conclusions about objects of information comprised in the document to build an ontology-based RDF graph.

A method, software and / or computer system for representing a set of SPARQL queries. The set of SPARQL queries are converting into an RDF representation of the set of SPARQL queries (that is, a SPARQL RDF graph). The SPARQL RDF graph is applied to a target RDF graph to yield a transformed target representing the responses to the set of queries, as an ontology or as text.

The invention discloses a partitioning and parallel distribution processing method of super-large scale RDF graph data. The method comprises the following steps: preprocessing original RDF graph data, generating a corresponding Hash dictionary file and reshaped three-column list data, and converting the reshaped three-column list data into an associated matrix M; creating a hypergraph model of the associated matrix M, wherein the subject, the predicate and the object of the M in the hypergraph model are hyperedges, and data associated with the hyperedges is hyperedge data; judging whether the RDF graph data is a connected graph or an unconnected graph; if it is the unconnected graph, partitioning the unconnected graph into a plurality of connected graphs; on the basis of the hypergraph model and concurrent breadth traversal, equidistantly placing the hyperedge data on a path, classifying and ranking the hyperedge data, uniformly partitioning the hyperedge data into K portions to be put into K slave nodes; and establishing mapping relationships among the hyperedge data and the slave nodes. The partitioning and parallel distribution processing method of the super-large scale RDF graph data is quick in partitioning speed, high in partitioning quality, balance in data and task loads, high in parallelism of query processing, and fast in query processing speed.

The invention discloses a knowledge map interactive visual query method based on RDF, which classifies user query into three types: entity-based query, pattern-based matching query, and regular path-based query, and visualizes query results. The invention designs a knowledge map interactive visualization method according to the characteristics of RDF and SPARQL standards and the modern interactivedesign concept, which can help users to query entities and relationships from RDF graph database, and accomplish complex queries by using the characteristics of pattern matching query and regular path query. The design is implemented based on front-end technology and React framework.

Technique to determine relative relevance of pairs of nodes in a linked data graph, such as an RDF graph, including calculation of raw relevance scores (which are based upon adjacency relations between nodes after edge filtering, the incoming and outgoing degrees of nodes in the graph after the edge filtering, and the distance (or weight) values of the edge(s)) and / or “refined relevance scores,” which take into account “complementary nodes.”

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

Systems and methods for visualizing relational data as RDF graphs in order to explore connections between data in the relational schema. The relational data is first converted into an initial RDF graph. Referential constraints between tables in the relational data, including unasserted referential constraints and pseudo-referential constraints, are automatically detected and used to augment the RDF graph. In addition, datatype properties in the RDF graph may be folded into annotation objects for better visualization. The resulting graph may be an edge-node graph, with edges corresponding to referential constraints between nodes corresponding to relational table rows.

This invention discloses an issuing / subscribing method suitable for isomerized event based on RDF event, which comprises the following steps: 1) judging whether the event issued is RDF type, if not, to step 2), if is, to step 3); 2) shifting the non-RDF event into RDF type, then issuing to net platform; 3) if event issued matching the condition of subscriber, sending corresponding event to he. Map event is shifted by following steps: ordering the MAP event data in turn of property name, whether the property must appear, whether the property can appear multi-times, logic sequence among the properties, explaining type of the property value, and processing the data to create RDF graph nodes and arcs, forming the RDF event; XML event is shifted by these: adopting XML schema to express XML event structure, making out its schema, using XSLT to shift XML event into RDF event.

The invention discloses an event publishing and subscribing method supporting fuzzy matching based on ontology, and the method allows a publisher and a subscriber to build an RDF event and an SPARQL statement with key word description through a browser on the basis of the domain ontology. The domain ontology stipulates the relationship between a concept in a certain field and a concept and is expressed by adopting the OWL; an event model stipulates the event expression means and is expressed by an RDF graph; a subscription request stipulates the subscription expression means and is expressed by the SPARQL statement and key words. In matching, firstly, events which meet the subscriber event type requirements are obtained by utilizing the upper and lower reasoning of the ontology; secondly, the relevancy score between the key words and the events is obtained through calculation by utilizing the relevancy measurement means in the information theory, and sequencing is performed on the events which meet the subscription requirements by utilizing the relevancy score; lastly, the sequenced event collection is returned to a user, and therefore fuzzy matching of event publishing and subscribing can be achieved.

The present specification relates to Ontology modeling, and, more specifically, to systems and methods for populating a triple store (RDF Graph) data structure from a parse tree diagram and producing a measurable increased degree of confidence in the reliability of the inferences based on the matched axioms derived from the ontology model. The steps of populating and producing can be performed automatically.

The invention provides a test case reduction method based on RDF reasoning, which comprises a preprocessing step, an OWL modeling, a preliminary reasoning, a further reasoning and a test case reduction. This method extracts the RDF information from the HTML source code containing RDFa and generates the initial RDF graph. Make use of the tool to build the owl model, and finish the preliminary reasoning work; Analyzing the relationship between elements in owl model, perfecting the attributes of elements, designing reasoning rules, deducing the hidden relationship between page elements by using reasoning tools, generating perfect RDF diagram; Combined with the principle of generating test cases from RDF diagrams, all test paths are marked, and the generated test paths are reduced according tothe test paths, and finally a simplified test suite is generated. This method can generate a wide range of small number of reduced test suite, and improve the test efficiency, reduce the test cost.

The invention discloses an RDF graph and attribute graph unified storage method based on a relation model. According to definitions and characteristics of an RDF graph model and an attribute graph model of two data models of a knowledge graph and a storage concept of a relation table, two logic models of an RDF graph and an attribute graph are stored in a physical model of the relation table at abottom layer, including bottom-layer storage of the RDF graph and bottom-layer storage of the attribute graph. The method comprises the steps of storing all semantic information of an RDF graph and anattribute graph in a bottom layer in the form of a relation table, and maintaining a system table for an RDF three-super graph structure to realize edge point double-copy storage for managing certainedges in the RDF graph. According to the method, the problem of difference of semantic expression capacities of the RDF graph and the attribute graph can be solved, large-scale knowledge graph data storage and management are achieved, data redundancy is effectively reduced, efficient query is achieved, and the method has high application value.

The invention discloses a data restoration method and apparatus, and belongs to the technical field of information. The method comprises: according to data in a resource description framework (RDF) knowledge base, obtaining RDF graph data and a neighbor step number; according to the RDF graph data and the neighbor step number, determining at least one frequent graph structure; mining a semantic dependency relationship in each frequent graph structure; and according to the semantic dependency relationship, detecting abnormal data in the RFD graph data and restoring the abnormal data. According to the data restoration method and apparatus, the at least one frequent graph structure is determined according to the RDF graph data and the neighbor step number, and the abnormal data in the RDF graph data is detected according to the mined semantic dependency relationship in each frequent graph structure, so that the number of error types of the detected abnormal data is relatively large and the data restoration range is expanded by restoring the detected abnormal data.

The invention discloses a large-scale knowledge graph storage scheme based on a distributed key value library. A "KG over SQL" unified storage scheme or a "KG over KV" unified storage scheme is adopted; the unified storage scheme of "KG over SQL" comprises the following steps: reading knowledge graph data; extracting metadata in the mapping knowledge domain, and generating a corresponding SQL statement; extracting data in the knowledge graph, storing the data in a data structure, and generating a corresponding SQL statement; executing an SQL statement through an external SQL interface of the Cockroach DB database, and establishing a relation table; mapping the data in the relation table into a key value pair form by using an original data coding mapping scheme in a Cockroach DB database, and storing the data in a Cockroach DB bottom layer; By means of the "KG over KV" unified storage scheme, knowledge graph data are read, a knowledge graph is converted into a graph containing index nodes, index node graph data are stored in a data structure, and the data are mapped into a key value pair form through an optimized data coding mapping scheme and stored in a Cockroach DB bottom layer. According to the method, efficient data storage compatible with the RDF graph model and the attribute graph model can be realized.

The invention relates to an urban security data retrieval method based on a knowledge graph. The characteristics of a Neo4j graph database are utilized to map the urban security knowledge map RDF graph data and continuously flowing knowledge fragments into a graph database; in a distributed environment, when multiple keywords are spread over multiple mutually associated label nodes, the multiple keywords are associated with the multiple mutually associated label nodes. In a label map of interaction, according to the proportion of the instance nodes anchored by the keywords to the key relationship, obtaining the key relationship, According to the method, the label with the highest value is quickly calculated in advance according to the key node and the proportion between the labels, the initial instance node can be determined at the fastest speed from the label, meanwhile, the time for reaching the next hop of key node is shortened, and the database retrieval efficiency of the dynamic change knowledge map is improved.

The invention belongs to the field of natural language processing and discloses a vector space model based relative mapping method. The method includes steps: firstly, according to a corresponding relation between relatives and entity pairs, constructing a characteristic vector of each vector by statistics of entity pair occurrence frequency and specificity; secondly, adopting a vector space modelto calculate similarity of the relatives to corresponding predicates, and selecting high-similarity predicates as candidates of the relatives; finally, sorting all candidates of the relatives, and selecting the predicate candidates highest in reliability to construct a mapping dictionary. By adoption of the relative predicate mapping method, an effective solution can be provided for automatic mapping of the relatives in natural languages and the predicates in RDF graph data, and converting the natural languages into the graph data to realize corresponding matching is realized.

The invention discloses a storage internal index construction method for improving query and storage performance of a novel memory. The method includes: when RDF graph data is stored, sequencing the RDF graph data, then dividing the sequenced RDF graph data into different pages, establishing a skip list capable of realizing binary search for the pages, wherein the structure of nodes in the skip list is (k, v) form, k corresponds to node information, and v corresponds to a logic address; then, storing the ordered RDF graph data into a solid state disk (SSD); and establishing a mapping table of the physical address and the logic address of the RDF graph data storage by utilizing the FTL, and storing the mapping table. According to the method, the SSD can be quickly searched and efficiently managed, and the method has a very strong competitive advantage.

The invention discloses an RDF approximate answer query method based on node context vector space. Firstly, the vector space representation of an RDF graph is generated, then the RDF query graph is partitioned, and then the approximate answer is found based on the vector space. Finally, the approximate answer results of the RDF query graph are returned to help users to define the information requirements and modify the initial RDF query graph. The invention can solve the problem of RDF inquiry empty set answers, in which the learning method of RDF representation based on node context can generate the continuous vector representation of RDF data set offline. The RDF query graph can satisfy the user's query intention when projected in continuous vector space, and the generated approximate answers and corresponding recommended queries are sorted and returned to the user for explanation. This data-driven solution method improves the quality of RDF approximate query and significantly reduces the time overhead.

The invention discloses an RDF distributed storage method based on a multi-layer partition framework. The main steps comprise (1) optimizing the vertex movement in an RDF graph through an MMA algorithm to protect small communities in the RDF graph; (2) roughening the RDF graph by an MSLM algorithm, finding the community structure of RDF graph, and reducing the scale of RDF data; (3) by means of the B_AP algorithm, realizing the k-Way segmentation, so that the amount of data between physical storage nodes is relatively balanced, and the communication cost between nodes is reduced. The inventionprovides the complete RDF distributed storage method, which lays a foundation for improving the query efficiency of the RDF.

The invention discloses a spatial RDF data keyword query method based on an abstract graph, and belongs to the field of keyword query. The method comprises the steps of dividing an original RDF graphbased on type attributes to obtain a set P of divided sub-graphs, and extracting different graph structures in the P based on a relationship between types to obtain an abstract graph set S; carrying out keyword query on P in a reverse search manner by utilizing S to generate a sub-graph set containing all keywords; and combining the sub-image sets into the input of an SK algorithm to obtain a final query result. According to the method, an abstract search layer is added on the basis of the SK algorithm, a divided sub-graph set and an abstract graph set are generated, the reverse search is conducted on the divided sub-graph set through the abstract graph set, that is, search is conducted from a keyword node from bottom to top, and a plurality of sub-graphs containing all keywords are obtained. As the number of the divided sub-graphs is far smaller than that of the nodes, the query range is reduced, and the lower bound of the path length can be quickly obtained by utilizing the abstractgraph, so that the query efficiency is obviously improved.

The invention discloses a temporal T-SPARQL query and reasoning method based on Pregel. Query processing and optimization are carried out on temporal RDF graph data in a graph parallel computing mode. The research problem can be formally defined as follows: a tense RDF query graph G and a tense query graph Q are given, and all matches of the query graph Q in the data graph G are found out. And converting the T-SPARQL query into a function executed on the vertex, performing communication on the edge by adopting a message passing mode, storing an intermediate result set in a vertex attribute, and performing message aggregation and continuous matching on the basis of the received message in the next round of iteration until the iteration is finished. According to the T-SPGX algorithm, a corresponding query plan is generated according to a given T-SPARQL query, the query plan comprises predicate label matching and time information filtering, and different time filtering methods are adopted for a single tense triple mode and a plurality of tense triple mode connection query. The query efficiency is improved from the aspect of query result optimization, the semantic hierarchy of the tense RDFS is used for reasoning the tense RDF data, an implicit result is reasoned in existing explicit query results, and a query result set is expanded. A universal and extensible solution is provided for the query method of the massive tense RDF data.

The invention discloses an RDF triad-oriented data custom mapping method. The method comprises the following steps of 1, obtaining relational database data; step 2, directly outputting the database table structure and the data as an RDF graph by utilizing a Direct Mapping mapping language; customizing a specific database structure and a word list by utilizing an R2RML mapping language, and outputting the data relationship library as an RDF data set; and step 3, taking a Direct Mapping mapping language as a default mapping scheme, and taking R2RML as a personalized customization scheme. The Direct Mapping mapping language provides an initial rule for automatic conversion of a machine, the R2RML provides a manually self-defined logic framework and a grammar rule, and the two mapping modes supplement each other, so that data in the relational database is better converted into triple data of the RDF.

The invention designs a distributed keyword query method based on an RDF graph, and belongs to the field of information retrieval. The method comprises the steps of firstly, converting the RDF data graph into an RDF sentence graph; secondly, by utilizing a conditional depth-first algorithm and a simulated annealing algorithm, segmenting the RDF sentence graph according to two most basic principlesincluding minimum of an edge cut set and data balance between sub-graphs after segmentation; and finally, refining the segmented RDF sentence graph into the RDF data graph, obtaining a vertex cut setof the RDF graph, and by utilizing a reverse search algorithm and a Hadoop distributed computing framework, realizing efficient and fast query of keywords. Under the condition of guaranteeing the atomicity and semantic integrity of RDF data, the limitation of a traditional algorithm on the segmentation efficiency of a large-scale data set is effectively eliminated, and the query efficiency of thekeywords is greatly improved.

The invention provides a query method and system for large-scale time sequence RDF (Resource Description Framework) graph data. The query method comprises the following steps: S1, uniformly loading and storing time sequence RDF graph data in a quintuple format into memories of a plurality of machines in a key-value storage mode; s2, creating a plurality of client threads and a plurality of working threads on each machine; s3, the client thread receives the query request of the user, analyzes the query request of the user, and sends the analyzed query request to the working thread of the corresponding machine; s4, executing the query task by the working thread to obtain a final query result; and S5, returning the query result to the client thread by the working thread.

The invention discloses an information integration method and device, equipment and a readable storage medium. Different types of data sources including a plurality of object integration relationships in a certain local area network are used as knowledge sources of a graph, comprehensive integration relationships in the local area network are summarized, knowledge extraction is performed on each data source subsequently, and extracted knowledge is recorded according to an RDF triple format to obtain RDF data; and the RDF data is fused and is converted into an RDF graph based on a knowledge fusion technology, and the RDF graph is visually displayed. The obtained RDF graph clearly records the integration relation between objects such as systems and interfaces in the local area network, and unified management and maintenance of the object integration relation can be achieved based on the RDF graph. Correspondingly, the invention provides an information integration device, equipment and a readable storage medium, which also have the above technical effects.

The invention discloses a tense RDF (Resource Description Framework) query method based on SparkGraphX. The method comprises the following steps: converting a tense RDF graph into a tense attribute graph with a label on the basis of a Spark GraphX distributed graph processing framework, performing sub-graph matching query on tense RDF graph data on the basis of an optimized Pregel interface in GraphX, converting T-SPARQL query into a function executed on a vertex, performing query processing and optimization on the tense RDF graph data in a graph parallel computing mode, and performing query processing and optimization on the tense RDF graph data. And two strategies of vertex calculation optimization and communication cost optimization are provided to improve the effectiveness of the algorithm. Various methods exist for storing tense RDF data and processing T-SPARQL query, but the graphic property of the RDF data causes a large amount of iterative connection, which has great influence on system performance, and the research of a tense RDF model mainly lies in integration of tense information and a classical RDF model, so that the relationship between the classical RDF model and the tense information needs to be coupled well. According to the method, large-scale query of the tense RDF data is carried out on a distributed platform, the characteristics of the time dimension and the graph structure are combined, storage and query of the tense RDF are efficiently carried out, the data set of the tense RDF model is better managed, and the query speed of the data set is increased.