59 results about "Graph query languages" patented technology

Systems and methods are directed to incorporating large language model (LLM) prompts into graph queries using graph query language and executing these graph queries. The system receives a graph query that includes a LLM prompt in the graph query language. The LLM prompt comprises an input context and a query to be answered. Based on the graph query, the system traverses a knowledge graph to obtain context data associated with the input context. The context data may be constrained by access control policies of an enterprise providing the knowledge graph. A context table may be generated using the context data based on the input context indicated in the LLM prompt. The system then invokes an LLM to determine a result by applying the query of the LLM prompt to the context data obtained from the traversal of the knowledge graph. The result is then presented on a user interface of a client device.

The invention discloses a charging station operation maintenance method based on a multi-dimensional data system, and relates to the field of intelligent optimization, and the method comprises the steps: collecting the operation, environment, user and power grid data of a charging pile, preprocessing the data into JSON, carrying out the real-time analysis and classification through Kafka-Flume, storing the data into a distributed database, building a composite index, and carrying out the retrieval through Presto; after cleaning, combining GraphQL and a graph query language to construct a space-time atlas, and extracting a multi-dimensional feature vector; a digital twinborn model is generated, multi-source data are fused through virtual-real mapping, and residual correction and reinforcement learning optimization are combined; fusing multi-modal data by using a hypergraph neural network, and optimizing topology in combination with an entropy regularization multi-target strategy and meta-learning; power is dynamically distributed, LSTM-GAN predicts and positions faults, and PID regulates and controls output; and the block chain contract automatically triggers equipment freezing and power limiting, and on-chain data feedback optimizes scheduling and load balancing. According to the invention, through multi-dimensional data fusion and a dynamic optimization mechanism, the operation and maintenance accuracy and response efficiency of the charging station are significantly improved.

The invention relates to the technical field of data retrieval, and discloses a city multi-department heterogeneous data intelligent retrieval method and system based on a large language model and a knowledge graph, a terminal and a storage medium, and the method comprises the following steps: constructing the knowledge graph based on a concept-instance-record organization level; analyzing the natural language query based on the large language model to generate a query intention; generating a graph query language according to the query intention, and extracting an initial retrieval result from the knowledge graph according to the graph query language; and if the initial retrieval result does not meet the requirement of the query intention, extracting a supplementary retrieval result from the heterogeneous data source through a multi-agent backtracking mechanism, and integrating the initial retrieval result and the supplementary retrieval result to obtain a target retrieval result. According to the method, through knowledge graph construction, graph query language generation and a multi-agent backtracking mechanism, the retrieval efficiency and the retrieval precision of urban multi-department heterogeneous data are effectively improved, and the comprehensiveness and the integrity of a retrieval result are ensured.

The invention relates to the technical field of knowledge maps, in particular to a ferric sulfate knowledge map construction method based on multi-modal heterogeneous data, which comprises the following steps: acquiring multi-modal heterogeneous data in a ferric sulfate production process; taking a mechanical operation rhythm obtained by analyzing the multi-axis high-frequency vibration signal data as a main time reference to form a public time reference; performing alignment and association on the tomography data and the acoustic emission data based on public time reference to obtain an instantiation process normal form entity in the knowledge graph; analyzing the time sequence of the process normal form entity according to a predefined ontology relationship type, establishing a directed relationship with a state transition probability attribute, and forming a probability state machine data structure traversed by a graph query language in a knowledge graph; filling structured attributes are defined for process normal form entities in the knowledge graph. According to the method, the construction of the ferric sulfate knowledge graph is realized by fusing multi-modal data and state analysis.

The embodiment of the invention provides a graph query statement generation method, a data query method and related devices, and belongs to the technical field of natural language processing. The graph query statement generation method comprises the following steps: identifying an entity relationship in a natural language question, and retrieving a target vertex matched with the entity relationship in a graph; wherein the graph is pre-generated based on the graph structure data; acquiring local graph structure data within a specified degree from the target vertex, and generating a first prompt statement based on the local graph structure data; and inputting the first prompt statement into the large language model to generate a first graph query statement. According to the method, the accuracy of converting the natural language into the graph query language by a large model can be improved.

Systems and methods are directed to incorporating large language model (LLM) prompts into graph queries using graph query language and executing these graph queries. The system receives a graph query that includes a LLM prompt in the graph query language. The LLM prompt comprises an input context and a query to be answered. Based on the graph query, the system traverses a knowledge graph to obtain context data associated with the input context. The context data may be constrained by access control policies of an enterprise providing the knowledge graph. A context table may be generated using the context data based on the input context indicated in the LLM prompt. The system then invokes an LLM to determine a result by applying the query of the LLM prompt to the context data obtained from the traversal of the knowledge graph. The result is then presented on a user interface of a client device.

Systems, methods, and other embodiments associated with decision tree machine learning native to a graph database are described. In one embodiment, a method includes receiving an objective property value of a training graph for a graph-native decision tree to be trained to predict; determining feasible ranges of values from the training graph; automatically generating a graph query language grammar for graph-native decision trees based on the feasible ranges; progressively evolving output decision trees in a graph query language from the graph query language grammar to generate a final decision tree that accurately predicts the objective property value; and storing the final decision tree for subsequent execution against a target graph.

The invention provides an equipment asset management system and method based on a graph model, a data module is used for acquiring sample data, and the sample data comprises network equipment in a network service architecture and a dependency relationship among the network equipment; the graph model module is used for constructing a graph model according to the sample data, each network device in the network service architecture is a vertex of the graph model, and a dependency relationship among the network devices is an edge of the graph model; and the query module is used for receiving a query request of a user, performing query in the graph model by utilizing a graph query language based on the query request, and displaying a query result, and the query result displays the equipment dependence network with potential influence. By constructing the graph model, the dependency relationship between the devices can be effectively judged, and potential influence analysis is carried out, so that the device management efficiency is improved, and the device management effect is optimized.

The invention discloses a communication method and related equipment, and is applied to a communication system, the communication system comprises a service architecture SBA network element and an SBA semantic agent, the SBA semantic agent comprises a semantic graph and a graph query language, the semantic graph synchronizes service logic and service data from the SBA network element, and the method comprises the following steps: the SBA semantic agent obtains a graph query request; the SBA semantic agent calls a graph query language according to the graph query request; the SBA semantic agent accesses the semantic graph through a graph query language. In the application, since the semantic graph in the SBA semantic agent synchronizes the service logic and the service data of the SBA network element, the service of the SBA network element is realized by accessing the semantic graph in the SBA semantic agent, a request for calling a service interface does not need to be initiated to the SBA network element, so that the service interface does not need to be written for the SBA network element, and the efficiency of accessing the service of the SBA network element is improved.

The invention discloses a communication system and communication equipment. The communication system comprises an SBA network element and an SBA semantic agent corresponding to the SBA network element. The SBA semantic agent comprises a semantic graph and a graph query language, the semantic graph synchronizes service logic and service data from an SBA network element, and the SBA semantic agent is used for obtaining a graph query request and calling the graph query language according to the graph query request so as to access the semantic graph through the graph query language. In the application, since the semantic graph in the SBA semantic agent synchronizes the service logic and the service data of the SBA network element, the service of the SBA network element is realized by accessing the semantic graph in the SBA semantic agent, a request for calling a service interface does not need to be initiated to the SBA network element, so that the service interface does not need to be written for the SBA network element, and the efficiency of accessing the service of the SBA network element is improved.

The invention discloses a large language model graph query language generation method and system based on reinforcement learning, and the method comprises the steps: constructing a supervision fine tuning training data set comprising natural language query, a reasoning process and graph query language query, and generating the reasoning process through teacher model distillation; performing supervision and fine tuning on the large language model based on the training data set; constructing a reinforcement learning training data set containing natural language query and graph query language query; reinforcement learning training is carried out on the large language model after supervision and fine tuning; and obtaining natural language query input by a user and corresponding graph mode data, and generating a graph query language through the trained large language model. According to the large language model graph query language generation method and system based on reinforcement learning, fine adjustment is performed on the student model from the teacher model distillation training data, reinforcement learning training is performed on the model after fine adjustment, the reasoning ability of the model is remarkably improved, and the accuracy and practicability of the graph query language are improved.

The invention discloses a freight unmanned aerial vehicle fault diagnosis method based on hybrid driving of a multi-relation fault knowledge graph and a large model thinking chain prompt, and the method comprises the following steps: constructing a freight unmanned aerial vehicle multi-relation fault knowledge representation, and determining the direct connection between entities and reasoning fault knowledge; entity recognition, relation extraction and knowledge fusion are carried out on the fault knowledge base under the knowledge representation category to complete construction of a multi-relation fault knowledge graph; obtaining a freight unmanned aerial vehicle fault problem, and adopting a large model thinking chain to prompt query; guiding the large model to generate a graph query language through a predefined thinking chain prompt template; and determining the logic relationship of the multi-relationship fault knowledge graph, and carrying out successive reasoning query to obtain an accurate fault reason and a corresponding solution. According to the method, through accurate knowledge modeling and logical reasoning questions and answers, the problems of expression and retrieval of rich relations hidden among fault knowledge are solved, and the strain capacity of a freight unmanned aerial vehicle system to complex factors is improved through rapid and accurate fault diagnosis.

The invention discloses a structure-aware large model graph reasoning method and system, which can convert graph data into a structure-aware sequence and support zero sample reasoning of a large model so as to efficiently and accurately complete a graph reasoning task. Comprising the steps that firstly, graph data and a graph reasoning task are prepared, the graph data comprise nodes, edges and features of the nodes, the edges and the features of the edges, then standardization processing is conducted on the graph data, and structured coding is conducted on the graph data; then, the structured coded graph data is converted into a structure perception sequence, and the sequence can be a code language or a standard graph query language; then, serialized graph data are input into the large model, and a specific input format and task description are designed, so that the large model can understand and execute a graph reasoning task; then, analyzing the reasoning result from the output of the large model, and converting the result into structured data or natural language description for a user or other systems to use; and finally, verifying and evaluating the reasoning result to ensure the reliability of the result.

The embodiment of the invention provides a method and device for processing a graph query request based on multi-stage intermediate representation. The method comprises the steps that the graph query request written by utilizing a graph query language is converted into AST; according to a target dialect set for a graph query language, the AST is converted into a dialect IR through a first infrastructure provided by an MMIR frame, and the MMIR frame further provides a Pass manager; and optimizing the dialect IR by using a target Pass pipeline configured by the Pass manager, and converting the optimized dialect IR into a physical execution plan which can be executed by a graph query engine of the graph database.

Systems and methods are directed to incorporating large language model (LLM) prompts into graph queries using graph query language and executing these graph queries. The system receives a graph query that includes a LLM prompt in the graph query language. The LLM prompt comprises an input context and a query to be answered. Based on the graph query, the system traverses a knowledge graph to obtain context data associated with the input context. The context data may be constrained by access control policies of an enterprise providing the knowledge graph. A context table may be generated using the context data based on the input context indicated in the LLM prompt. The system then invokes an LLM to determine a result by applying the query of the LLM prompt to the context data obtained from the traversal of the knowledge graph. The result is then presented on a user interface of a client device.

A method may include receiving a definition of a table user-defined function (TUDF) in a graph query language. The table user-defined function may be created based on the definition. For example, the creation of the table user-defined function may include checking and compiling the definition to generate executable code associated with the table user-defined function. Upon receiving a query including a relational query language statement invoking the table user-defined function, such as a structured query language select statement, the query may be executed on at least a portion of a graph data stored in a database. The executing of the query may include calling the executable code to execute the table user-defined function included in the relational query language statement. Related systems and computer program products are also provided.

The present invention relates to the field of data retrieval technology, and discloses a method, system, terminal, and storage medium for intelligent retrieval of heterogeneous data from multiple urban departments based on a large language model and a knowledge graph. The method comprises: constructing a knowledge graph based on a concept-instance-record organizational hierarchy; parsing natural language queries based on the large language model to generate query intent; generating a graph query language based on the query intent, and extracting initial retrieval results from the knowledge graph based on the graph query language; if the initial retrieval results do not meet the requirements of the query intent, extracting supplementary retrieval results from the heterogeneous data source through a multi-agent backtracking mechanism, and integrating the initial retrieval results and the supplementary retrieval results to obtain the target retrieval results. The present invention effectively improves the retrieval efficiency and retrieval accuracy of heterogeneous data from multiple urban departments through knowledge graph construction, graph query language generation, and a multi-agent backtracking mechanism, and ensures the comprehensiveness and completeness of the retrieval results.

The invention discloses an AI coding tool enhancement method based on a graph model and multi-stage DSL generation. The method comprises the following steps: 1) analyzing a C / C + + project source code into a multi-view semantic graph model intermediate format; step 2) loading the intermediate format of the graph model to a query engine; 3) receiving an artificial natural language task of the artificial intelligence coding tool, enhancing a generation framework through multi-stage retrieval, and analyzing the task into an executable graph query language; 4, the final query, the execution result abstract and the execution context are packaged into a structured knowledge enhancement prompt, and the structured knowledge enhancement prompt is fed back to an upstream AI coding tool.By means of the method, the problems of macro semantic loss, construction environment mismatch, type system understanding shallow layer and the like in a C / C + + project are effectively solved, and the knowledge perception and reasoning capacity of a large model in a complex code scene is remarkably enhanced.

The application relates to the technical field of graph database query, and particularly discloses a governance-aware compiling method and system for ontology graph query, which comprises the following steps: receiving and analyzing a graph query request, and constructing a graph traversal abstract syntax tree; obtaining a governance strategy associated with a current query session, expressing tenant isolation constraints, access level constraints, use limitation constraints, soft deletion filtering constraints and result set size constraints in the governance strategy as governance predicate nodes, and injecting the governance predicate nodes into node scanning nodes and edge traversal nodes of the graph traversal abstract syntax tree; selecting a graph query dialect compiler corresponding to a target graph database engine, compiling the graph traversal abstract syntax tree with the injected governance predicate nodes into a target graph query language statement, and expanding the governance predicate nodes into corresponding filtering clauses during the compiling. The application realizes governance injection during graph query compiling, guarantees that the governance injection maintains semantic contractility on the graph query, and enables the same governance strategy to produce consistent governance effects on the graph query and the relational query.

The embodiment of the invention provides a corpus generation method and device based on a graph query language. In the method, a first corpus template used for constructing a conversion corpus of a graph query language is obtained, the first corpus template comprises a natural language template and a corresponding graph query language template, and the two templates comprise a first type of placeholders used for representing definition types of triple elements. Then, extracting a placeholder path formed by a first type of placeholders and a connection relationship thereof from the natural language template or the graph query language template, and generating a plurality of types of paths meeting relationship constraints based on the pattern definition and the placeholder path of the graph data, and using the definition type of the triple element in any type path to replace the corresponding first type placeholder in the natural language template and the graph query language template, and determining a first corpus based on a replacement result. Wherein the mode definition comprises definition types of triple elements and relation constraints among the triple elements. In the processing process, privacy protection needs to be carried out on related privacy data.

The application provides a scene graph semantic information extraction method and system for a graph query. The method analyzes the entity, relationship and attribute structure in the Cypher language of a user query, and inputs the structure information as a prompt input of a scene graph generation model. The scene graph fuses the graph structure information of the query requirement to generate simple scene graph data, aligns the scene graph data with the original data model in the graph database, stores the scene graph, and finally realizes efficient query of semantic information in a picture by using a graph query language. The application is guided by the user query requirement, converts a picture into graph structure data by using a scene graph extraction model, and performs data alignment, solves the short board that a previous graph query language cannot query a picture, and innovatively and efficiently extracts scene semantic information according to the user query requirement, realizes efficient query, and improves the shortcomings of the previous scene graph extraction model, such as long extraction time and waste of storage space.

The present invention discloses a method for the operation and maintenance of a charging station based on a multi-dimensional data system, which relates to the field of intelligent optimization. It includes collecting data on the operation of charging piles, the environment, users, and the power grid. After preprocessing it into JSON, it is parsed and classified in real time through Kafka-Flume, stored in a distributed database, and a composite index is established, and retrieved through Presto. After cleaning, a spatio-temporal map is constructed by combining GraphQL and graph query language, and multi-dimensional feature vectors are extracted. A digital twin model is generated, and multi-source data is fused through virtual-real mapping, and combined with residual correction and reinforcement learning for optimization. A hypergraph neural network is used to fuse multi-modal data, and combined with an entropy regularization multi-objective strategy and meta-learning to optimize the topology. Power is dynamically allocated, LSTM-GAN is used to predict and locate faults, and PID is used to regulate the output. The blockchain contract automatically triggers equipment freezing and power limiting, and the data on the chain is fed back to optimize scheduling and load balancing. The present invention significantly improves the accuracy and response efficiency of the operation and maintenance of the charging station through a multi-dimensional data fusion and dynamic optimization mechanism.

The invention discloses a train control system test scene and test case management method based on a knowledge graph, and relates to the technical field of railway signal and train operation control system testing. Comprising the following steps: constructing a knowledge graph in a graph database; establishing an ordered many-to-many relationship between the test scene set and the test case set, and setting execution sequence attributes on the ordered many-to-many relationship; importing into a graph database, and constructing a test knowledge graph; transactional dynamic adjustment operation is carried out in the database, consistency verification is executed at the same time, and if verification fails, automatic rollback is carried out; aiming at any test scene, utilizing a graph query language to sort the associated test cases, and automatically extracting an ordered test case set; and filling of the test execution document and change influence analysis are realized. According to the method, the compiling efficiency and accuracy of the test sequence in the joint debugging and test stage are improved.

The application relates to the technical field of knowledge graphs, in particular to a ferrous sulfate knowledge graph construction method based on multi-modal heterogeneous data, which comprises the following steps: acquiring multi-modal heterogeneous data of a ferrous sulfate production process; taking mechanical operation rhythms obtained by analyzing multi-axis high-frequency vibration signal data as a main time reference to form a common time reference; aligning and correlating tomographic imaging data and acoustic emission data based on the common time reference to obtain instantiated process paradigm entities in a knowledge graph; analyzing time series of the process paradigm entities according to predefined ontology relationship types, establishing a directed relationship with a state transition probability attribute, and forming a probabilistic state machine data structure that can be traversed by a graph query language in the knowledge graph; and defining a filling structured attribute for the process paradigm entities in the knowledge graph. The application realizes the construction of a ferrous sulfate knowledge graph by fusing multi-modal data and state analysis.

A method comprises obtaining a prompt indicating that when a user query satisfies a set of criteria, the user query is to be translated into a set of graph queries in a graph query language; receiving a specific user query in natural language indicating access to an ontology and satisfying the set of criteria, data of the ontology being stored in one or more databases, including a graph database; incorporating the specific user query into the prompt to obtain an extended prompt; executing, with the extended prompt, a first large language model; obtaining, from the executing, a set of database queries including one or more graph queries in the graph query language to access the ontology; submitting the set of database queries to a set of databases to obtain a database query result; transmitting the database query result in response to the specific user query.

The application discloses a kind of intelligent management system and method of product oil transaction fusing knowledge graph, it is related to the technical field of intelligent management, including the following steps, step S100, based on nonlinear fork type bifurcation model, the stable boundary of entity association is obtained and entity is filtered, step S200, construct heterogeneous graph convolution network, based on graph query language and heterogeneous graph convolution network, obtain knowledge spectrum diagram, step S300, based on knowledge spectrum diagram identification current abnormal node and trace back first transaction path.The application is by introducing nonlinear dynamics method to process the stability screening of transaction entity, eliminate risk entity from the root, realize the abnormal tracing based on graph reasoning, provide the complete transaction path of explainable, upgrade knowledge graph from static knowledge base to dynamic analysis engine, promote product oil transaction management from experience-driven to data intelligent-driven, have significant technical advantages and application value in improving regulatory efficiency, preventing and controlling transaction risk, supporting tax compliance and the like.

The invention provides a domain adaptation graph query language generation method and device based on a large language model, and relates to the field of natural language processing. The domain adaptation graph query language generation method based on the large language model comprises the steps of obtaining natural language query information; a pre-trained large language model is used for conducting domain adaptation processing on the natural language query information, graph query language information of the target domain is generated, and pre-training is used for enabling the large language model to be aligned with a graph database of the target domain. According to the method, the business logic described by the natural language is understood through the pre-trained large language model, the graph database query statement is directly generated, and the bottleneck of traditional query statement programming is avoided. Meanwhile, the graph query language generated by the large language model has cross-platform and cross-database mobility, and is suitable for different business fields and rules.

The present invention provides a method and device for generating a domain-adaptive graph query language based on a large language model, and relates to the field of natural language processing. The method for generating a domain-adaptive graph query language based on a large language model includes: obtaining natural language query information; using a pre-trained large language model to perform domain adaptation processing on the natural language query information, and generating graph query language information of the target domain, wherein the pre-training is used to align the large language model with the graph database of the target domain. The present invention understands the business logic described in natural language through a pre-trained large language model, and directly generates a graph database query statement, thereby avoiding the bottleneck of traditional query statement programming. At the same time, the graph query language generated by the large language model has cross-platform and cross-database portability, and is suitable for different business fields and rules.

The invention provides a demand analysis generation method and device based on a knowledge graph and a large model. The method comprises the steps that a demand analysis appeal of a user for a target system is obtained, a programming large model is adopted to convert the demand analysis appeal into a graph query language, query processing is conducted on the graph query language in a knowledge graph, an analysis sub-graph is obtained, and the demand analysis appeal represents an analysis appeal for modifying or querying the target system; the demand analysis appeal is retrieved according to a system detailed arrangement knowledge base, system detailed arrangement documents corresponding to the demand analysis appeal are obtained, and the system detailed arrangement knowledge base comprises system detailed arrangement documents of a plurality of target systems; and inputting the system detailed document and the analysis subgraph into the large reasoning model to obtain a demand analysis report corresponding to the demand analysis appeal. The problem that the preset function point cannot cover the demand scene in the demand analysis process when the business demand evaluation is carried out through the preset function point in the prior art is solved.

The embodiment of the invention provides a query statement generation method and device in a graph query language. According to the method, an existing first query statement written through a graph query language is obtained, a first path query mode of the first query statement is determined, and the first path query mode comprises a path composed of triple elements. Then, the first path query mode serves as a query subgraph, subgraph query is conducted in a mode definition graph of graph data, a plurality of query results are obtained, the query results are paths formed by definition types of triple elements, the definition types in the query results are used for replacing the corresponding definition types in the first query statement, and the query result is obtained; and generating a new query statement. In the query statement generation process, privacy protection needs to be carried out on involved privacy data.