65 results about "Schema graph" patented technology

Various embodiments virtualize data across heterogeneous formats. In one embodiment, a plurality of heterogeneous data sources is received as input. A local schema graph including a set of attribute nodes and a set of type nodes is generated for each of the plurality of heterogeneous data sources. A global schema graph is generated based on each local schema graph that has been generated. The global schema graph comprises each of the local schema graphs and at least one edge between at least one of two or more attributes nodes and two or more type nodes from different local schema graphs. The edge indicates a relationship between the data sources represented by the different local schema graphs comprising the two or more attributes nodes based on a computed similarity between at least one value associated with each of the two or more attributes nodes.

Tools and methods for schema matching that generate schema graphs, populate match matrices and display the schema graphs and the match matrices. These tools and methods characterize potential matches between disparate schemata in terms of both a strength of evidence indicating the potential match and an amount of evidence indicating the potential match. A number of match voters generate a set of match scores for each potential match, and these match scores are combined by a vote merger to form a single confidence value for each potential match. A number of filters display the confidence value for each potential match as a link on a graphical user interface. Machine-learning techniques may be employed to adaptively determine confidence values based on previously established matches.

An intelligent caching data structure and mechanisms for storing visual information via objects and data representing graphics information. The data structure is generally associated with mechanisms that intelligently control how the visual information therein is populated and used. The cache data structure can be traversed for direct rendering, or traversed for pre-processing the visual information into an instruction stream for another entity. Much of the data typically has no external reference to it, thereby enabling more of the information stored in the data structure to be processed to conserve resources. A transaction / batching-like model for updating the data structure enables external modifications to the data structure without interrupting reading from the data structure, and such that changes received are atomically implemented. A method and mechanism are provided to call back to an application program in order to create or re-create portions of the data structure as needed, to conserve resources.

An intelligent caching data structure and mechanisms for storing visual information via objects and data representing graphics information. The data structure is generally associated with mechanisms that intelligently control how the visual information therein is populated and used. The cache data structure can be traversed for direct rendering, or traversed for pre-processing the visual information into an instruction stream for another entity. Much of the data typically has no external reference to it, thereby enabling more of the information stored in the data structure to be processed to conserve resources. A transaction / batching-like model for updating the data structure enables external modifications to the data structure without interrupting reading from the data structure, and such that changes received are atomically implemented. A method and mechanism are provided to call back to an application program in order to create or re-create portions of the data structure as needed, to conserve resources.

The invention discloses a tax domain-oriented knowledge graph construction method. According to the construction method, a top-down and bottom-up combined mode is adopted. The method comprises the following steps: firstly, constructing a tax knowledge graph mode based on tax knowledge in a tax expert system; then, data processing including tax data source selection and acquisition, data cleaning and the like is carried out; then information extraction is carried out, and information extraction is carried out on the processed data according to different types and a mode diagram; tax knowledge fusion is carried out, knowledge sources in a tax knowledge graph are different, problems of knowledge repetition, relationship redundancy and the like need to be subjected to mode matching, entity alignment and the like, and knowledge is stored in a knowledge base after knowledge fusion is completed. And finally, knowledge feedback is carried out, and knowledge conflicts in intelligent tax model construction are solved by using a tax expert system. And finally, the problem of untight combination of the knowledge graph and the tax scene is solved.

The invention discloses a webpage information extraction method and system. The method includes the steps: acquiring a marked webpage, generating a semantic structure tree, building an information mode pattern, generating semantic attribute node information of each semantic attribute node in the information mode pattern, generating a wrapper and deriving the wrapper into a wrapper document; building an extractor for extracting webpages similar to the marked webpage; acquiring the webpages to be extracted, and recursively extracting a data extraction area or an iterative data extraction area corresponding to each semantic attribute node in the information mode pattern layer by layer from the root semantic attribute node in the information mode pattern in a DOM (document object model) tree of the webpages to be extracted by the extractor; deriving data in the data extraction area or the iterative data extraction area corresponding to each semantic attribute node as extraction results. The method has high universality, generalization capability, fault tolerance and expandability and low manual involvement degree, and online extraction efficiency is ensured, so that practicability is high.

The invention discloses a natural language-oriented content diversification searching method. The method mainly includes: understanding a query statement of a user, and then converting a natural-language query into a query pattern graph with ''query focus'' through technology of entity recognition, text representation, entity relationship extraction and the like; then using a method of structure matching to carry out matching calculation on a query pattern graph and a knowledge map to obtain a series of results satisfying a query requirement; and finally, obtaining Top-k query results, which are highly related to the ''query focus'' and diversified, through diversification calculation, and thus better satisfying the query requirement of the user. According to the method of the invention, more diversified retrieval contents can be provided for the user, and searching speed is high.

One or more embodiments of the invention provide a continuous sub-graph matching method and a continuous sub-graph matching system for mode graph change, and equipment. The continuous sub-graph matching method comprises the steps of: constructing a data structure based on a network topological graph G and a mode graph P, wherein the data structure is used for storing a local matching result of thenetwork topological graph G and the mode graph P; constructing a maintenance model based on the dynamic change of the mode graph P; updating the data structure based on the dynamic change of the modegraph P by means of the maintenance model, so that the data structure stores a real-time local matching result; constructing a cost model based on a matching sequence of the mode graph P; calculatingthe minimum matching cost of the mode graph P by using the cost model, and acquiring a matching algorithm based on the minimum matching cost; and acquiring a final matching result of the network topological graph G and the mode graph P based on the real-time local matching result and the matching algorithm, so as to complete network threat detection. According to the continuous sub-graph matchingmethod and the continuous sub-graph matching system, abnormal network attack behaviors can be monitored, an impending network attack mode can be predicted, and an attack target of a network hacker can be speculated.

The invention discloses a sensitive recognition method based on a knowledge graph, and the method comprises the steps: firstly carrying out the preprocessing of an obtained original data set in orderto build a user commodity knowledge graph, building a mode graph of a user-article through the preprocessed data, and then building the knowledge graph according to the preprocessed data and the modegraph; secondly, in order to identify the sensitive data, complementing the sensitive relationship which originally does not exist between the user and the object in the knowledge graph through the constructed sensitive relationship reasoning rule; and finally, inquiring sensitive data from the whole knowledge graph and outputting the sensitive data. The recognition speed is increased.

An automated method for cyberattack detection and prevention in an endpoint. The technique monitors and protects the endpoint by recording inter-process events, creating an inter-process activity graph based on the recorded inter-process events, matching the inter-process activity (as represented in the activity graph) against known malicious or suspicious behavior (as embodied in a set of one or more pattern graphs), and performing a post-detection operation in response to a match between an inter-process activity and a known malicious or suspicious behavior pattern. Preferably, matching involves matching a subgraph in the activity graph with a known malicious or suspicious behavior pattern as represented in the pattern graph. During this processing, preferably both direct and indirect inter-process activities at the endpoint (or across a set of endpoints) are compared to the known behavior patterns. The approach herein provides for systematic modeling of inter-process behaviors for characterizing malicious or suspicious patterns among processes.

The invention discloses a multi-task external memory mode diagram processing method based on I / O scheduling, includes streaming partitioning graph data to obtain graph partition, evenly placing graphpartition in multiple external storage devices, selecting target external storage devices from multiple external storage devices based on I / O scheduling, and taking graph partition in the target external storage device that has not been accessed by graph processing task as designated partition; Judging whether the synchronization field of the designated partition is not mapped into the memory according to the synchronization field of the designated partition, if so, mapping the designated partition from the external storage device into the memory, and updating the synchronization field of thedesignated partition; Otherwise, the graph partition data is accessed directly through the address information mapped to memory by the specified partition. Through I / O scheduling, the invention selects the external storage device with the least number of tasks to access, thereby controlling the sequence of accessing the data of the external storage diagram partition and balancing the I / O pressure.By setting the synchronization field to realize the data sharing of graph partition, the repeated loading of the same graph partition is reduced, so as to reduce the total I / O bandwidth and improve I / O efficiency.

The invention discloses a graph pattern matching method based on multi-source-point parallel exploration. The graph pattern matching method can be used for graph pattern fuzzy query starting from a determined point in a graph database. According to the invention, the to-be-queried mode graph is decomposed according to the hierarchical structure, graph traversal query is carried out by taking the graph layer as a unit, the exploration depth can be remarkably reduced, and the parallel exploration performance is improved. Concepts of a center mode set and an edge mode set are provided and used for controlling the exploration process, an exploration task is converted from a sub-graph as the center to a point as the center, and an algorithm can be achieved on a universal distributed graph computing platform. According to the matching result refining method for merging the exploration results of the multiple source points, repeated exploration is carried out by specifying the multiple auxiliary source points, constraint on the matching results is enhanced by utilizing layer differences brought by different visual angles, and the matching precision is improved.

The invention discloses a DDoS attack traffic identification and detection method under an SDN. The method comprises the following steps: S1, collecting flow table information; S2, calculating a traffic characteristic value with DDoS attack characteristics according to the traffic table information, and converting the traffic characteristic value into characteristic vector data; s3, utilizing the characteristic vector data to construct a random forest model, adopting the constructed random forest model to screen the traffic characteristics with little effect of identifying the DDoS attack, and obtaining an optimal characteristic subset; S4, converting the optimal feature subset into a feature vector Fnew, measuring the feature vector Fnew and the Mahalanobis distance of each node in the feature mode graph, if a node having the nearest Mahalanobis distance with the feature vector Fnew exists, determining that the feature mode corresponding to the node is the category of the feature vector Fnew, and if the category is a benign category, determining that the traffic is normal; otherwise, determining that the traffic is the DDoS attack traffic.

The invention discloses an operation optimization decision-making method for a gas-steam combined cycle unit, and belongs to the technical field of gas turbines. The operation optimization decision-making method for the gas-steam combined cycle unit comprises the following steps: step 1, mining and processing data in a distributed control system of the gas-steam combined cycle unit, and summarizing effective operation data; step 2, decomposing layer by layer based on a calculation relationship and a mechanism relationship of the operation parameters of the gas-steam combined cycle unit, and carrying out classification processing on the effective data; step 3, combining the classified effective data, and carrying out calculation processing by utilizing a performance calculation principle and a consumption difference analysis principle; step 4, drawing the optimal operation mode graph or curve graph according to a calculation result; and step 5, feeding back to an operation center through an information network to assist the unit to optimize operation. According to the invention, the performance state of the unit is monitored in real time, the optimal operation curve is sought through the performance and consumption difference analysis and calculation, and the safe and efficient operation of the unit is guaranteed.

The invention provides a medical information acquisition method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of databases. The medical information acquisition method comprises the steps of performing entity recognition operation on unstructured case information, and generating an entity sequence; determining a mode diagram matched withthe type of the medical information; when it is detected that the entity sequence comprises an entity matched with the type, executing matching operation on the entity sequence and the mode diagram to determine an entity relationship in the entity sequence; converting the unstructured case information into structured case information based on the entity relationship; and acquiring the medical information based on the structured case information. According to the technical scheme of the invention, the configuration of the mode diagram can achieve the definition of a more complex knowledge rule, and enables the medical information extracted based on the structured case information to be higher in accuracy.

The present invention discloses a server connection method and system, relating to the technical field of communications and used for implementing an optimal network using a server having identical specifications, thus reducing the complexity of routing. The method comprises: providing 10 servers; each of the servers comprising 5 nodes and the 5 nodes of each server being connected serially and end to end; connecting in one-to-one correspondence the 5 nodes of any one of the 5 servers and the 5 nodes of each of the remaining of the 5 servers, forming an optimal network of a Hoffman-Singleton graph.

The embodiment of the invention provides an isomorphic pattern recognition method and device, equipment and a medium. The method comprises the steps: receiving a target graph and a mode graph, wherein each of the target graph and the mode graph comprises nodes and edges connected with the nodes; screening the nodes in the target graph based on the labels of the nodes in the target graph and the labels of the nodes in the mode graph, and obtaining a sub-graph of the target graph according to the screened nodes in the target graph; and searching the mode graph in the sub-graph of the target graph to obtain an isomorphic graph of the mode graph. According to the embodiment of the invention, the efficiency of identifying the isomorphic pattern of the mode pattern from the target pattern is improved, and the time complexity and time consumption of identifying the isomorphic pattern of the mode pattern from the target pattern are reduced.

The invention discloses a method and a system for detecting abnormal power consumption behaviors of a power consumer. The method comprises the following steps: 1) obtaining user power consumption data through non-intrusive load monitoring; 2) processing the user power consumption data by adopting a deep Q network learning method to obtain a user power consumption behavior pattern pattern, wherein the user power consumption behavior pattern pattern comprises a checked abnormal user power consumption behavior pattern pattern; 3) extracting power consumption behavior characteristics from the checked abnormal user power consumption behavior pattern graph; 4) establishing a plurality of analysis models according to the extracted power utilization behavior characteristics; 5) adopting a Q network learning method to train and evaluate the plurality of analysis models, and selecting the analysis model with the optimal evaluation effect; and 6) detecting whether the power consumption behavior of the user is abnormal or not by adopting the analysis model with the optimal evaluation effect. Through monitoring and reasonable extraction and utilization of the electric energy data, reasonable power distribution of the power distribution network is facilitated, and the existing electric energy is used at the maximum efficiency.

The invention discloses a ship industry knowledge graph construction and analysis method. The method comprises the following steps of: pre-defining a relationship between entities, and constructing a mode graph of a ship industry knowledge graph; obtaining knowledge according to the constructed mode graph, and obtaining the entities and relational data of the ship industry and links between the entities; performing knowledge fusion on the obtained entities and relational data, and performing entity alignment, entity disambiguation and data classification on the data; performing knowledge storage on data obtained after knowledge fusion, storing the data into a relational database so as to process the data, and storing a processed triple into a graph database to form a ship industry knowledge graph; and finally, based on the ship industry knowledge graph, analyzing a ship industry structure from three aspects, namely, regional industry layout, enterprise community discovery and focus enterprise discovery, and updating the predefined entities in the mode graph and the relationship between the entities based on an analysis result. According to the method of the invention, research and management of a ship industry cooperation network are facilitated.

The invention provides an Ethereum address clustering method and device based on neighbor information aggregation. The method comprises the following steps: constructing an address target graph and an address transaction graph based on Ethereum historical transaction data; establishing a strong association address node mode graph and an exchange address mode graph based on neighbor node information in the transaction graph, and performing address node neighbor information aggregation; establishing a subgraph isomorphic CSP model, and performing clustering to obtain a strong association address node set and a transaction exchange address set. Compared with the prior art, the Ethereum address clustering method based on neighbor information aggregation provided by the invention can track the relevance between the address node and other address nodes, so that the mutual relation and source between the address nodes in the block chain transaction can be identified more accurately.

The embodiment of the invention provides a graph matching method and device, electronic equipment and a storage medium, and the method comprises the steps of obtaining a graph matching statement, and analyzing the graph matching statement to obtain an initial mode graph; determining a root node from the initial mode graph, and decomposing the initial mode graph based on the root node to obtain a target star-shaped sub-graph; accessing nodes in the data graph through an iterator interface provided by a storage engine, and matching the star-shaped sub-graphs in the data graph to obtain a compressed matching result corresponding to the target star-shaped sub-graph; and processing the matching result corresponding to the star-shaped sub-graph through a parallel pipelined multi-path connection algorithm to obtain a matching result of the initial pattern graph. According to the method, the matching statement is analyzed, the target star-shaped sub-graph is used for matching, random access to a disk in matching can be avoided in a storage mode with the node as the center, the matching efficiency is improved, and efficient graph matching is achieved.

The invention discloses a knowledge graph-based question and answer implementation method. The method comprises the following steps of: when a natural language query instruction of a user is received,firstly, carrying out entity mention extraction and entity linking on a query statement of the user to obtain a candidate entity set and a candidate relationship set of a query text; scoring the relationship matching degree between the query text and the candidate relationship set by using a relationship extraction technology, selecting the relationship with the highest score as a semantic relationship corresponding to the natural language query text, filtering out invalid entities in the candidate entity set based on the semantic relationship of the query text, and finding out semantic entities of the query text; constructing a query mode graph based on the extracted semantic entities and semantic relationships, and finding all matching items matched with the query mode graph in the knowledge graph, the matching items being answers retrieved by natural language query based on the knowledge graph. The invention further discloses a question and answer implementation system based on theknowledge graph. According to the invention, the user experience is optimized.

The invention discloses a task type dialogue state tracking method fusing slot association and semantic association, which is used for fusing slot association information and mining the relationship between slots by using a graph neural network. Firstly, a multi-relation mode graph is constructed for slots, and then attention scores of a relation layer and a node layer are respectively calculated by utilizing a hierarchical graph attention network, so that an explicit or implicit relation between the slots is captured. In order to fuse the semantic relationship, the method introduces a word-level semantic similarity vector in a slot gate mechanism to obtain local semantic matching information. According to the mode, the local semantic features of the slot position and the current dialogue can be obtained, and the similarity vector is used as the enhanced feature of the slot gate mechanism, so that whether the slot position is involved in the dialogue or not is better judged, and the prediction mode of the slot value is determined. And finally, decoding different prediction modes through two sub-modules in the slot value decoder, so that the method can effectively improve the prediction precision and speed of the dialogue state.

The invention discloses a method for path discovery in a knowledge graph, which comprises the following steps: acquiring the knowledge graph G = (V, E) meeting strong mode constraint, V being a node set comprising a plurality of nodes, E being an edge set comprising a plurality of edges, each node having a node type thereof, and each edge having an edge type thereof; obtaining a mode graph = (,) of the knowledge graph, which is a node type set comprising a plurality of node types and an edge type set comprising a plurality of edge types; determining one or more mode graph paths of the node type of the first node and the node type of the second node in a mode graph; and performing a traversal lookup in the knowledge graph along each of the one or more pattern graph paths to discover one or more paths between the first node and the second node. The invention also relates to other path discovery methods and associated systems, devices and media. According to the scheme, path discovery can be efficiently executed.

The present invention relates to the field of information retrieval. The present invention aims to solve the problem that the existing knowledge map retrieval is not convenient enough, and proposes a knowledge map retrieval method based on a Bayesian classifier, by establishing the corresponding relationship between query sentences and query structure graphs , convert the query sentence into the corresponding semantic label sequence, and then calculate the probability that the query sentence is mapped to each type of corresponding query structure graph through the Bayesian classification model, select the query structure graph with the highest probability value as the query pattern graph, and perform Knowledge graph retrieval. It builds a bridge between flexible natural language query and knowledge graph semantic retrieval, making information retrieval more convenient for ordinary users, and is suitable for film and television retrieval or other knowledge graph retrieval.

The invention discloses a ship-based segmented logistics mode mining and visualization method and system, and the method comprises the steps of collecting the ship segment original logistics data, andgenerating a ship segment logistics transportation original data table; preprocessing the original ship segment logistics transportation data table to generate a ship segment logistics feature vectorof each ship segment; carrying out the hierarchical clustering analysis on the multiple ship segment logistics feature vectors to generate a clustering tree diagram, and generating the multiple shipsegment categories and ship segment sets corresponding to the multiple ship segment categories according to the clustering tree diagram; and performing statistical analysis and normalization processing on the set and logistics transportation original data tables to generate a logistics mode feature statistical table, and drawing a ship segment logistics mode graph of a plurality of ship segment types according to the feature statistical table. According to the method, problems existing in an existing ship section logistics mode can be obtained, and the discovered problems are analyzed or solved in a targeted mode by combining the characteristics of specific ship sections corresponding to each logistics mode.

Various embodiments virtualize data across heterogeneous formats. In one embodiment, a plurality of heterogeneous data sources is received as input. A local schema graph including a set of attribute nodes and a set of type nodes is generated for each of the plurality of heterogeneous data sources. A global schema graph is generated based on each local schema graph that has been generated. The global schema graph comprises each of the local schema graphs and at least one edge between at least one of two or more attributes nodes and two or more type nodes from different local schema graphs. The edge indicates a relationship between the data sources represented by the different local schema graphs comprising the two or more attributes nodes based on a computed similarity between at least one value associated with each of the two or more attributes nodes.

The invention discloses a method for establishing a distributed graph routing protocol of a wireless sensor network under an asymmetric link, which belongs to the technical field of computer networks. This method designs a distributed graph routing calculation method for wireless sensor networks without a central management server; defines the format of graph routing interaction messages; designs asymmetric link state downlink (DOWN-LINK), uplink ( UP-LINK), peer to peer (peer to peer, P2P) three modes diagram routing establishment, interaction, update process. The invention aims to solve the routing asymmetry problem when wireless sensor networks establish routes in an asymmetric link environment. At the same time, in the absence of a central management server, the distributed routing calculation method is used to solve the hardware bottleneck of centralized graph routing. question. The distributed graph routing protocol proposed by the invention can effectively reduce the management and control overhead of traditional graph routing, enhance the robustness of the network, and optimize the routing performance of the wireless sensor network as a whole.

The invention relates to the technical field of home decoration design, in particular to a graph processing method in home decoration design, which comprises the following steps of S1, area measurement, S2, engineering drawing, S3, size marking, S4, contour point marking, S5, drawing scanning, S6, graph conversion, S7, data storage, S8, data export and S9, graph display. According to the method, the contour points are marked on the contour positions of the target house area, and the positions of the area contour points can be effectively and clearly expressed in the process of converting image information into graphic information by utilizing a contour tracking mode of photoshop software, so that the target area in a graph is more comprehensively displayed with an optimal effect.