24 results about "Graph patterns" patented technology

The invention relates to the technical field of big data, in particular to an intelligent operation and maintenance management method based on a big data algorithm, and the method comprises the steps: constructing and continuously updating a dynamic fault association graph through inputting multi-source heterogeneous operation and maintenance data; starting full-graph scanning based on a predefined period, detecting an abnormal topological structure through a graph pattern recognition algorithm, and marking potential risk nodes; executing dynamic influence diffusion simulation on the potential risk nodes, calculating a business influence severity quantized value after the fault, and marking fault propagation vulnerabilities according to the quantized value; taking the potential risk node as a starting point, executing a reverse traceability algorithm for preferentially exploring a path pointing to a fault propagation vulnerable point, and outputting a fault propagation path and a source fault node identifier; and finally generating and executing a fault processing strategy. The process solves the problem that traditional operation and maintenance cannot quantitatively evaluate and discriminate the highest priority disposal object from numerous potential risks, and realizes accurate positioning and active prevention and control of weak links of fault propagation.

The invention discloses a big language model Cypher generation method and system fusing graph mode selection and reinforcement learning, and belongs to the technical field of graph databases and big models. The method comprises the following steps: firstly, constructing training data, carrying out reinforcement learning training fused with graph mode selection on a large model, selecting a sub-graph mode related to a problem from a graph database mode, and verifying the sub-graph mode; and then, taking the selected sub-graph mode as a core context, and carrying out Cypher query generation. The large model is trained through a specially-designed reinforcement learning normal form, a selection strategy of a graph mode can be learned and optimized, and a selection result is recursively used for guiding and correcting a Cypher generation process. According to the method, a graph mode selection process is used as a key reasoning step and is seamlessly integrated into a Cypher generation process, so that the focusing capability and reasoning capability of the model when facing a complex and huge graph mode are remarkably improved, and the accuracy and practicability of Cypher query are greatly improved.

The invention discloses a graph pattern generation method based on a Clarii graph, and relates to the technical field of information, and the method comprises the steps: constructing an extended fluorescent material calibration module, introducing a saturation critical light intensity reference value, and carrying out the calibration of the Clarii graph under the dual constraints of color temperature deviation distribution and brightness standard deviation; according to the method, the pattern contrast change condition is recognized through the adjusted pattern contrast, a correction index is formed, the LED backlight wavelength is corrected for the second time according to the correction index, pattern structure deviation information is fed back to a spectrum control link, and the color uniformity of the LED backlight is improved. According to the scheme, on the premise of ensuring uniform fluorescence response and inhibiting bright spots, dark spots and stripes, the Clariy pattern can be kept stable, clear and layered in an actual application environment.

The application provides a weakly supervised knowledge graph automatic construction method for professional text, relates to the technical field of knowledge graph construction, and comprises the following steps: converting a professional text into a structured text format to generate a preprocessed text set; based on a preset domain graph pattern, constructing a structured knowledge unit by using a large language model, identifying a pattern conflict, and generating a candidate new pattern description document; evaluating the comprehensive confidence of the candidate new pattern description document through a multi-dimensional confidence evaluation mechanism, and dynamically updating the domain graph pattern; optimizing and reconstructing the structured knowledge unit, calculating the quality index of the reconstructed weakly supervised knowledge graph, repeatedly constructing the weakly supervised knowledge graph until the quality index of the weakly supervised knowledge graph reaches a preset quality stability threshold, and obtaining a weakly supervised knowledge graph of a professional text. The technical problem that the preset domain graph pattern in the prior art is rigid and difficult to adapt to emerging knowledge types in the construction of a weakly supervised knowledge graph is solved.

The application is suitable for the field of model explanation, and relates to a rule mining method and device for recommending model explanation, equipment and medium. For any user-item pair in graph data, a corresponding neighborhood graph and a connected subgraph are determined, a candidate subgraph is determined from the connected subgraph, a target subgraph is determined from the candidate subgraph according to a graph evaluation score, a target graph pattern is extracted from the target subgraph, for any variable in a pattern path of the target graph pattern, a target predicate is determined from all predicates corresponding to the variable, a candidate premise condition is formed by the variable and the target predicate, a candidate explanation rule is obtained according to the target graph pattern and a candidate premise condition set corresponding to all pattern paths in the target graph pattern, and a candidate explanation rule satisfying a preset condition is determined as a target explanation rule from all candidate explanation rules. The target explanation rule reflecting the prediction principle of the recommendation model is mined from the graph data as the global explanation of the recommendation model, and the effectiveness of the explanation is improved.

A computer-implemented method for generating by a query engine a graph of operators for a SPARQL query over an RDF graph. The method includes obtaining a graph of operators executable by the query engine, the graph comprising a plurality of basic operators, at least two of said operators being of a first type each configured to find RDF triples of the RDF graph that match a respective basic graph pattern. The method further comprises identifying a group of operators among the at least two basic operators of the graph which are of the first type. The respective basic graph patterns of the group of operators have same subject and / or predicate and / or object and the identified group of operators is replaced in the graph by an equivalent operator configured to find RDF triples of the RDF graph that match the respective basic graph patterns of the group of operators.

The invention provides a professional text-oriented weak supervision knowledge graph automatic construction method, and relates to the technical field of knowledge graph construction, and the method comprises the following steps: converting a professional text into a structured text format, and generating a preprocessed text set; based on a preset domain graph mode, constructing a structured knowledge unit by using a large language model, identifying mode conflicts, and generating a candidate new mode description document; evaluating the comprehensive confidence of the candidate new mode description document through a multi-dimensional confidence evaluation mechanism, and dynamically updating the domain atlas mode; optimizing and reconstructing the structured knowledge unit, and calculating a quality index of the reconstructed weak supervision knowledge graph; and repeatedly constructing the weakly supervised knowledge graph until the quality index of the weakly supervised knowledge graph reaches a preset quality stability threshold, and obtaining the professional text weakly supervised knowledge graph. The technical problem that in weak supervision knowledge graph construction in the prior art, a preset domain graph mode is rigid and is difficult to adapt to emerging knowledge types is solved.

The application provides a method and device for event prediction based on a timing diagram rule, the method comprising: obtaining timing diagram data, and adding event prediction results of a machine learning model to the timing diagram data to obtain timing diagram extension data; extracting a subgraph for verifying prediction rule reliability from the timing diagram extension data according to a verification ratio; training a rule creator according to the timing diagram extension data and the subgraph to obtain a timing event prediction rule set; and obtaining a prediction event set according to the timing event prediction rule set and the timing diagram data. By embedding the machine learning model for event prediction into a rule-based association relationship system as a predicate, the association rule can not only use the existing machine learning model event prediction result, but also use a logical condition to improve the prediction result of the machine learning model, has stronger expression ability, and does not have to constrain constant interval time or have a common focus constraint on a graph pattern.

The invention provides a fund transaction association relationship mining and abnormal behavior identification method based on a knowledge graph, and relates to the technical field of knowledge graphs, and the method comprises the steps: collecting fund transaction flow data, and carrying out the cleaning and time window division, and obtaining a standardized transaction data set; constructing a time sequence dynamic knowledge graph; defining a rule type exception graph mode, extracting exception sub-graphs from historical exception case data, selecting a prototype mode, and combining the prototype mode and the rule type exception graph mode to form an exception transaction graph mode library; executing graph pattern matching to obtain an abnormal sub-graph instance set; and calculating an abnormal score of each abnormal sub-graph instance in the abnormal sub-graph instance set, performing score propagation based on an account node overlapping relationship between the abnormal sub-graph instances, and aggregating the abnormal scores of all the abnormal sub-graph instances in which each account node participates for each account node to obtain an account-level abnormal score. According to the invention, the recognition accuracy of the complex hidden abnormal transaction network can be improved.

The application relates to a social network repeated event detection method based on repeated subgraph pattern mining. The method comprises the following steps: obtaining a social network snapshot graph of a current timestamp, mapping an m-edge subgraph related to a newly inserted edge e in the social network snapshot graph into a corresponding bucket of a first auxiliary data structure by using a first hash function, and updating the frequency and frequency counter state of a subgraph pattern in the corresponding bucket; obtaining the frequency of the updated subgraph pattern, and if the frequency is equal to a support threshold value sigma, mapping the corresponding four-element information of the subgraph pattern into a corresponding unit of a second auxiliary data structure by using a second hash function, and updating the four-element information in the corresponding unit; and obtaining repeated events in the current social network according to the events corresponding to the subgraph pattern in the updated second auxiliary data structure. The method can efficiently detect repeated events on a social network stream graph.

A computer-implemented method for generating a summary of a graph database comprising a set of RDF tuples including obtaining the graph database and generating a summary having a set of probabilistic filters. Each probabilistic filter of the set determines if at least one RDF tuple existing in the graph database corresponds to a respective basic graph pattern of the probabilistic filter with a possibility of false positive.

A method for modeling and analysing single-phase or multiphase digital signals in an angular coordinate axis defined depending on the signal's arguments and phase angles in a device having information processing capability is disclosed, a system comprising instructions operating on a device having a processor for applying the method of the invention to various signals and various angular axes, and a novel method for estimating signal vectors for multiphase signals at angular phases not provided as input signals. In the inventive method, single-phase or multi-phase digital signals received from a data source are displayed on an angular coordinate axis defined depending on the independent variables and phase angles of the signal. Depending on signal variables such as resolution amounts of the parameters desired to be displayed on the angular coordinate axis in the signal, linear or non-linear increase amounts, lower and upper limit values, a graphic pattern is created that is divided into sections called segments and layers (grids) on the angular coordinate axis. The signal is then expressed in segments and layers of the graph on the angular coordinate axis with numerical values, colours, symbols, signs or other visual elements. The method of the present invention can be used in devices with information processing capacity and can be used in modeling all kinds of signals. The method was implemented in a signal modeling system designed using the method, and various single-phase and multi-phase signals were tested on this system.

The invention discloses a cross-individual emotion online monitoring system based on electroencephalogram and map deep learning, and relates to the technical field of biomedical engineering. The system comprises an emotion induction module, a data collection module, a data processing module, a model calculation module and a stimulation presentation module, emotions are induced through video stimulation, multi-guide electroencephalogram signals are collected in real time, an electroencephalogram network connection matrix and multi-band differential entropy features are extracted after preprocessing, a multi-view space-time diagram mode is extracted by means of a graph neural network double-branch structure, and a multi-view space-time diagram mode is obtained. And in combination with transfer learning, realizing cross-individual emotion real-time decoding, and finally visually presenting a result through a graphical user interface. The system is divided into training and testing stages, the problems of poor robustness, insufficient cross-individual generalization and the like in the prior art are solved, multiple types of emotions can be accurately and dynamically monitored, and the method is suitable for actual scene application.

The application discloses a financial risk reasoning method and system based on positive and negative association, an electronic device and a computer readable storage medium. The method comprises the following steps: acquiring a financial extension graph containing positive association information and negative association information, and a financial reasoning rule set containing a graph pattern, an antecedent predicate set and a consequent predicate; matching the rule graph pattern in the financial extension graph to obtain a matching instance; checking whether the matching instance meets the antecedent predicate, instantiating the matching instance that passes the check according to the consequent predicate to generate a candidate reasoning result including a positive or negative conclusion; performing conflict detection and resolution on the candidate reasoning result to obtain a target risk reasoning result; and outputting the target risk reasoning result and reasoning explanation information. The positive and negative evidence fusion reasoning is realized, the false positive rate is reduced through conflict detection and multi-dimensional priority resolution, and auditable explanation information is outputted, so that the fine risk control and strong supervision scene demand are met.

The invention relates to the technical field of supply chain intelligent prediction, and discloses a supply chain demand prediction method and system based on artificial intelligence. The method comprises the steps of obtaining a complete transaction record in a target time period, forming a material flow time sequence, and screening flow subsequences according to a predicted target material variety, a time span and a geographical range. And generating a material flow change map based on the subsequences, and visually describing a flow dynamic fluctuation track. Scanning the atlas by using a preset atlas feature recognition rule, recognizing a specific traffic change form, marking the specific traffic change form as a to-be-analyzed feature, and extracting a corresponding original transaction record subset; constructing a deep neural network, taking the to-be-analyzed features and the transaction record subsets as input, and training the network to learn a mapping relation between the to-be-analyzed features and the transaction record subsets and subsequent material demands; according to the method, complex non-stationary demand fluctuation is captured through combination of map form identification and transaction detail multi-modal learning, and prediction accuracy and business interpretability are improved.

The invention is suitable for the technical field of graph data, and relates to an anti-fact interpretation generation method and device, equipment and a storage medium. Wherein a negative prediction rule of the graph neural network is obtained, and the negative prediction rule is composed of a reference graph mode and a rule attribute predicate set; a candidate space is constructed by taking a current node for performing negative prediction on the graph data by the graph neural network as a pivot, the candidate space comprises candidate node subsets corresponding to other mode variables except the current node of the reference graph mode, and the candidate node subsets comprise candidate nodes simultaneously matched with the reference graph mode and a rule attribute predicate set in the graph data; searching a candidate node with the highest perturbation cost performance and a corresponding graph perturbation operation in the candidate space, and updating the candidate space after executing the graph perturbation operation; iteratively executing until at least one candidate node subset in the candidate space is empty; and generating an anti-fact explanation corresponding to the current node according to the graph perturbation operation searched in the iteration process. According to the invention, the generation efficiency of anti-fact explanation is improved.

The application relates to the technical field of big data, in particular to an intelligent operation and maintenance management method based on a big data algorithm.In the application, multi-source heterogeneous operation and maintenance data is inputted, a dynamic fault correlation graph is constructed and continuously updated, full graph scanning is started based on a predefined period, an abnormal topological structure is detected through a graph pattern recognition algorithm, and potential risk nodes are marked, dynamic influence diffusion simulation is performed on the potential risk nodes, a quantitative value of the severity of business influence after a fault is calculated, and a fault propagation fragile point is marked accordingly, then, reverse tracing algorithm of preferential exploration directed to a fault propagation fragile point path is executed with the potential risk nodes as starting points, a fault propagation path and a source fault node identifier are outputted, and finally, a fault processing strategy is generated and executed; the process solves the problem that traditional operation and maintenance cannot quantitatively evaluate and identify the highest priority disposal object from numerous potential risks, and realizes accurate positioning and active prevention and control of a weak link of fault propagation.

Systems and methods to identify blockchain anomalies include an AI tool comprising a processor, a GPU, and a GNN model to extract graph parameters from a block transactions graph of a blockchain block, generate statistical approximations of the graph based on the graph parameters, compare the statistical approximations to at least one anomaly threshold, detect an irregular graph pattern in the graph when the statistical approximations exceed the at least one anomaly threshold, identify via the GNN model an anomaly within the block transactions graph based on the irregular graph pattern, generate via the GPU an address graph based on the block transactions graph when the anomaly is identified to display one or more addresses associated with the anomaly, and generate an alert when the anomaly is identified.

This invention relates to the field of document review technology and proposes an intelligent review method and system for contract compliance review. The steps include: preloading compliance regulations into a local node machine to build an initial knowledge base; receiving contract documents to be reviewed and review instructions uploaded by the user; performing fine-grained structural deconstruction on the contract documents to be reviewed, constructing a nested contract knowledge graph, mapping the subjects, clauses, constraints, and exceptions in the contract to entities and relationships; converting the review instructions into a graph query language, performing graph pattern matching retrieval in the nested contract knowledge graph, and extracting subgraphs related to the review instructions; inputting the extracted subgraphs and the review instructions into a large language model, guiding the large language model to perform step-by-step reasoning along the logical paths defined in the subgraphs, and generating review opinions containing compliance judgments and supporting evidence. This invention significantly improves the interpretability and accuracy of review results and effectively suppresses the illusion phenomenon of large models.

The invention discloses an efficient dynamic graph pattern mining method and system based on DAG drive execution, and the method comprises the steps: obtaining dynamic graph data of financial transactions and a to-be-mined pattern with a suspicious transaction account as a vertex and a suspicious transaction relation as an edge, representing a set operation through a DAG (directed acyclic graph), and obtaining a to-be-mined pattern; the common sub-expressions are identified and merged during compiling, so that a unified global DAG is constructed, and redundant calculation during operation is reduced; meanwhile, a DAG-based mining engine is introduced, data dependence is decoupled through fine-grained parallelism, multi-core CPU loads are balanced through a dynamic task allocation mechanism, and a global DAG is driven to execute substructure level tasks; and combining a recovery module based on a normative expression to efficiently assemble the mining result, assembling the mining result into a final sub-graph embedding result corresponding to each mapping instance, and completing a dynamic graph mode mining task of financial anti-fraud detection. According to the method, redundant calculation can be reduced, the parallel efficiency is improved, and the performance of dynamic graph mining in financial anti-fraud detection is improved.

The invention discloses a method for constructing an urban traffic space-time mapping knowledge domain, which relates to the technical field of intelligent traffic, and comprises the following steps: preprocessing multi-source traffic data, constructing a space reference grid unit, and outputting a standardized data set with a grid identifier and a time batch identifier; based on the standardized data set with the grid identifier and the time batch identifier, setting a map mode including a state entity, a road segment entity, a grid entity and an interest point entity, and outputting a map mode definition; creating a grid entity, a road segment entity and an interest point entity in the graph database according to the graph mode definition, and outputting and displaying a static knowledge graph sub-graph of a spatial relationship; and injecting the vehicle track data in the standardized data set with the grid identifier and the time batch identifier into the static knowledge graph sub-graph containing the display space relationship. According to the method, explicit modeling of the continuous time-space evolution process of the traffic flow is realized, and a dynamic data basis is provided for analyzing congestion propagation.

The application provides a reasoning program automatic conversion and operator fusion method and system for a heterogeneous platform, and belongs to the field of deep learning compilation optimization, and comprises the following steps: S1: in the PyTorch model reasoning execution process, the operator call information of the model is captured through a runtime interception mechanism, and the tensor operation information corresponding to the Python layer is also captured at the same time; based on the operator call information and the tensor operation information, a node-granularity adaptive hybrid computation graph is constructed; S2: the hybrid computation graph is scanned according to a topological sequence or an execution sequence, a local structure meeting a fusion condition is identified by using graph pattern matching for fusion, and a fused hybrid computation graph is obtained; S3: based on the fused hybrid computation graph, operator matching and mapping are performed by using an operator support table, parameter analysis is performed by using an operator conversion function, and Ascend C code is generated. The method can significantly reduce the development cost of model cross-platform deployment, and improve the execution efficiency of the model on the Ascend device.

The present application is applicable to the field of model explanations, and relates to a rule mining method and apparatus for recommendation model explanations, and a device and a medium. The method comprises: for any user-item pair in graph data, determining a corresponding neighborhood graph and connected sub-graphs, determining candidate sub-graphs from among the connected sub-graphs, and on the basis of graph evaluation scores, determining a target sub-graph from among the candidate sub-graphs; extracting a target graph pattern from the target sub-graph; for any variable in pattern paths of the target graph pattern, determining a target predicate from among all predicates corresponding to the variable, and using the variable and the target predicate to form a candidate precondition; and on the basis of the target graph pattern and candidate precondition sets corresponding to all the pattern paths in the target graph pattern, obtaining candidate rules for explanations, and determining, from among all the candidate rules for explanations, candidate rules for explanations that meet a preset condition as target rules for explanations. Target rules for explanations that can reflect a prediction principle of a recommendation model are mined from graph data and used as a global explanation for the recommendation model, thereby improving the effectiveness of explanations.

The invention discloses a big data search optimization and association method based on a dynamic knowledge graph, and relates to the technical field of computer information retrieval, and the method comprises the following steps: executing an enhanced query, carrying out graph pattern matching on a local sub-graph structure and a big data source, and retrieving a preliminary search result set based on a matching result; and performing association mining on the preliminary search result set, extracting a new entity and a new relationship in the result, and fusing the new entity and the new relationship into the initial knowledge graph to form an extended knowledge graph. According to the method, the shortest association path for connecting the query entity and the entity corresponding to the result is searched in the extended knowledge graph, the shortest association path is converted into the association interpretation information described by the natural language and is output along with the final result, and meanwhile, the extended knowledge graph is fed back to be used for processing subsequent search query. The interpretability of the search result is enhanced to improve the credibility of the user, and the iterative optimization of the search effect is realized.