100 results about "Relation (database)" patented technology

The invention provides an intelligent data query method based on a natural language, and relates to the technical field of intelligent data processing and natural language interaction.The intelligent data query method comprises the steps that firstly, enterprise original data is subjected to standard treatment, and a standardized theme database and a data directory and index definition document are constructed; key semantics are extracted based on unstructured knowledge, and a domain knowledge vector library is fused and constructed by combining document text fragments and vector representation of a mapping relation between historical questions of a user and an SQL (Structured Query Language). And after receiving a natural language question of a user, calling a large language model to identify a task type, and distinguishing knowledge questions and answers, data query and complex analysis. Executing corresponding operations according to different types: directly retrieving a vector library by knowledge questions and answers to generate answers; extracting keywords in data query and generating a query request in combination with context; and in the complex analysis, predefined workflow is judged and executed or intelligent agent processing is called, and query or analysis requirements are output.

The invention discloses a multi-modal data retrieval method and device, a storage medium and computer equipment. The method comprises the following steps: collecting multi-modal original data; on the basis of description information of metadata of original data, all metadata belonging to the same associated items and logic relations among all the metadata are obtained, a metadata chain is constructed, a distributed graph database is constructed on the basis of the metadata chain, the metadata chain is expressed in the distributed graph database in the form of a graph, the graph comprises nodes and edges, the nodes represent the metadata, and the edges represent the metadata. The edge represents a logical relationship between the metadata; and when a data retrieval instruction is received, traversing each node in the metadata chain along the logical relationship of the metadata chain in the distributed graph database, obtaining a target node matched with a data retrieval requirement corresponding to the data retrieval instruction, and returning original data corresponding to metadata represented by the target node. Multi-modal data dynamic association retrieval can be realized, and cross-modal information mining efficiency and accuracy are improved.

The invention provides a method and system for constructing a Neo4j knowledge graph based on an LLM natural language, and belongs to the technical field of smart tourism and knowledge graphs. The method comprises the steps that S1, text entities are recognized, and semantics are understood; s2, analyzing an entity relationship and generating a Cypher statement; s3, extracting entity attributes and performing constraint verification; s4, constructing a Neo4j knowledge graph; and S5, performing dynamic updating and quality control. According to the method, the natural language understanding ability of the large language model and the relation modeling advantage of the Neo4j graph database are fused, so that automatic construction, dynamic updating and culture compliance verification of the tourism knowledge graph are realized.

The invention discloses a global knowledge extraction method and system based on a large model and an RAG technology, and belongs to the technical field of large model data processing. According to the global knowledge extraction method and system based on the large model and the RAG technology, local sensitive hashing is adopted for duplicate removal, similar texts are combined through dynamic thresholds, and redundant data interference is reduced; a text segmentation strategy based on separator confidence ensures that segmented text blocks maintain semantic integrity and are adaptive to large model input length limitation, similarity retrieval and full-text retrieval are fused, results are reordered in combination with an RRF algorithm, semantic relevance and accuracy are considered, the multi-scene knowledge construction requirement is met, and the multi-scene knowledge construction efficiency is improved. High-correlation entity pairs are screened through mutual information, context knowledge is retrieved from a vector database in combination with an RAG system, semantic reasoning is conducted through a large language model, hidden logic relations among entities are generated, and jumping from data association to knowledge generation is achieved.

The invention discloses a query method based on a knowledge base and a large model. The method comprises the following steps: constructing the knowledge base of a database; obtaining target language information which is input by a user and contains a user question; based on the knowledge base, generating a target pseudo mode corresponding to the target language information; based on the target pseudo mode, screening out a minimum table set associated with answering the user question from the knowledge base; packaging the multi-table connection logic in the minimum table set into a query view; based on the query view and the target language information, a reference example pair list which is most similar to the user problem and indicates the mapping relation between the problem and the query statement is retrieved from a knowledge base, and each example pair in the reference example pair list represents the mapping relation between the problem and the query statement; and generating a target query statement corresponding to the target language information based on the user problem, the target pseudo mode, the query view and the reference example pair list, and performing retrieval in a database based on the target query statement to obtain a query result.

The invention relates to a data correction method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring an original structured statement and original annotation information; determining knowledge data corresponding to the original structured statement in a target knowledge database based on a metadata query engine; the knowledge data comprises graph relation metadata and semantic vector metadata; according to the data correction large model, the graph relation metadata and the semantic vector metadata, error recognition is carried out on the original structured statement and the original annotation information, data correction is carried out on error data obtained through error recognition, and a target structured statement and target annotation information are obtained. By adopting the method, the accuracy of structured statement and annotation error correction can be improved.

The invention discloses a digital main line automatic modeling method and system based on a large language model, and the system comprises a database system, an embedded model, a vector database, a matching processing unit, a front-end interface, a cue word generation module, and a large language model interface. The digital main line automatic modeling method based on the large language model comprises the following steps: S1, extracting data information of a standard database and a database to be matched and converting formats; s2, storing the data information of the standard database in a vector database in a vectorization manner by using an embedded model; s3, vectorizing to-be-matched business database data information, searching similar fields in a vector database, and selecting candidate fields; s4, the user checks the candidate similar fields and confirms the mapping relation; s5, the data matching relation is input into the large language model, and a data transfer SQL is generated; and S6, executing SQL to realize data integration and automatic modeling.

The invention discloses a multi-modal retrieval method and device, and relates to the technical field of multi-modal retrieval, and the method comprises the steps: obtaining a query vector corresponding to a user query text, and carrying out the query in a vector database according to the query vector, and generating a query result; vector codes corresponding to the multi-level content block structures corresponding to the multiple multi-modal table documents are stored in the vector database; the multi-level content block structure of any multi-modal table document comprises an atomic layer, a semantic layer and a relation layer; the atomic block comprises target element content and structure information of any cell in the multi-modal table document; the semantic block comprises aggregation content corresponding to any row, any column or any whole table in the multi-modal table document; the relation block comprises a text for describing the incidence relation between the multiple atomic blocks and the incidence relation between the multiple semantic blocks; the structure information comprises row and column indexes corresponding to each cell and a cross-modal association relationship among the multi-modal elements. According to the method, the multi-modal retrieval accuracy can be improved.

The invention belongs to the technical field of databases, and particularly relates to an LLM-Text2SQL-oriented database table relation exploration method. According to the method, a multi-stage cooperative processing strategy is adopted, database metadata and data content features are integrated, and semantic enhancement and structured completion are performed on original information through a large language model. And on the basis, a potential association candidate set is screened in combination with an algorithm based on feature similarity, multi-dimensional verification and judgment of an association relationship are performed by fusing a predefined rule and a large language model reasoning mechanism, and finally an entity relationship graph supporting interactive editing and iterative optimization is generated. According to the method, end-to-end automatic processing from a heterogeneous database with constraint missing and data integrity impaired to a standardized ER graph is realized, and a solid foundation is laid for remarkably improving the success rate of natural language SQL generation based on a large language model.

The invention discloses a wind turbine generator abnormal knowledge association method and device based on a knowledge graph and a medium, and the method comprises the steps: carrying out the collection of distributed multi-source heterogeneous data and knowledge based on the operation and maintenance demands of the wind power industry; establishing an OPC UA information model of the wind power equipment node; the method comprises the following steps: guiding an industrial multi-modal industrial large model to carry out standardized description and identification on node data through cue word engineering, completing entity extraction and attribute extraction, carrying out relation extraction and expression, and importing into a graph database to construct an abnormal knowledge graph; the semantic relation between the node entities is optimized and updated, knowledge merging and processing are completed, and the abnormal knowledge graph is dynamically updated; and for an abnormal knowledge application scene, performing knowledge reasoning based on abnormal knowledge graph mining entity association. According to the multi-modal knowledge graph construction method based on graph structure learning and fine tuning of the multi-modal industrial large model, relevance of different anomalies is explored, the knowledge base rich in abnormal semantics is constructed, and it is ensured that industrial abnormal faults are correctly processed.

The invention discloses an equipment fault diagnosis method and device based on a dynamic knowledge graph and a large model fine tuning technology. The method comprises the following steps: firstly, identifying a core entity from multi-source heterogeneous equipment fault data through a named entity identification model for fine tuning of domain data and a relation extraction model for special fine tuning of a fault diagnosis domain corpus, mining deep semantic association, and injecting the deep semantic association into a graph database after cleaning to form an initial knowledge graph; receiving user natural language fault description, realizing term and standard entity linking through editing distance fuzzy matching and Sension-BERT semantic vector similarity calculation, and combining bidirectional retrieval and attention mechanism fusion to obtain an enhanced context; and finally, generating a structured diagnosis report containing thinking chain reasoning based on an enhanced context by utilizing a specialized fine-tuning fault diagnosis large language model. According to the method, the limitation of a traditional diagnosis method is effectively solved, high-precision and interpretable equipment fault diagnosis is realized, and the diagnosis efficiency and reliability are improved.

The invention discloses a database transaction load sequencing method based on a conflict relationship, which is characterized in that non-conflict operations are grouped into operation batches in parallel by analyzing data access conflicts among the operations based on an original execution log of a database transaction load; the original execution sequence of conflict operations is deduced according to operation types, version dependence and lock competition conditions, then operation batches are sorted according to the sequence, an operation batch sequence is constructed, and therefore it is ensured that the execution result of the sequence is equivalent to the original execution result of a transaction load. Compared with the prior art, the method has the advantages that a kernel of a database system does not need to be intruded, the analysis speed is linearly expanded along with the scale of the original execution log, and good expandability and application prospects are achieved.

A computer-implemented method is provided of generating a system prompt for a large language model to convert a natural language query (NLQ) to a structured query language(SQL). 5 The method includes generating (1606) the system prompt for the NLQ based on information extracted from a first database. The information identified from a knowledge graph includes a structured view of data including telecommunications domain knowledge and relations within the data filtered by a chain of thought prompt that represents a logical relation between a plurality of nodes in the knowledge graph. The method further includes passing (1614) the system prompt to 0 the LLM to convert the NLQ to the SQL.

The invention relates to the technical field of databases, and discloses a data query method and device, equipment, a medium and a product, and the method comprises the steps: obtaining a plurality of query statements for a target database; the query statement is generated by the large language model according to an input natural language; according to the relation between the statistical tables corresponding to the multiple query statements, the multiple query statements are combined, and at least one combined statement is obtained; and executing the at least one merged statement to obtain query results corresponding to the plurality of query statements. According to the scheme, the multiple query statements can be merged according to the relation between the statistical tables corresponding to the query statements, so that the query result corresponding to the query statements can be obtained by executing the merged statements, compared with direct processing of the multiple query statements, the IO frequency of the database is reduced, the IO load of the database is reduced, and the query efficiency is improved. The unnecessary interaction overhead is reduced, and the data query efficiency is improved.

The invention discloses a knowledge indexing and retrieval method for an AUTOSAR hierarchical architecture, and the method comprises the steps: taking a hierarchical structure of an AUTOSAR CP knowledge system as a trunk, carrying out the vectorization of an input document, constructing an entity-relation hierarchical tree, obtaining a knowledge graph indexing structure, carrying out the query matching through the similarity during retrieval, and carrying out the retrieval. Sorting retrieval output entities and relationships, constructing a context, and outputting a structured answer; comprising the following steps: step 1, indexing: intelligently partitioning external database content input by a user and performing vectorization storage, operating general entity extraction and / or AUTOSAR special extraction and combining results, constructing a tree structure, performing hierarchical clustering to form a knowledge graph, and generating a community report; and 2, retrieval: receiving user query, finding related entities through vector similarity search, aggregating contexts in combination with tree and graph dual structures, and starting universal cue words and / or AUTOSAR special cue words to generate structured answers. The invention further discloses a system for implementing the method, and the system has wide application value.

The invention discloses an NL2SQL and NL2VIS method based on large model collaborative intelligent agent, and belongs to the technical field of natural language processing. The NL2VIS method comprises the steps that candidate database modes related to natural language query are obtained from database modes, and task complexity corresponding to the natural language query is obtained according to the number of tables contained in the candidate database modes; in combination with the task complexity, natural language query is converted into a preliminary SQL query statement; verifying, repairing and feeding back the SQL query statement based on the initial SQL query statement until a final SQL query statement is generated; and converting the final SQL query statement into a visual query language. According to the method and the device, the database table, the field and the association relation of the database table and the field involved in the query intention of the user can be more accurately identified, and connection errors, field mistaken selection and information omission caused by mode misunderstanding are fundamentally reduced.

The invention discloses a prediction method, device and equipment and a readable storage medium, and the method comprises the steps: obtaining a relational database containing a to-be-predicted object in response to a prediction request, taking a business object in an entity table in the relational database as an entity node, taking a business event in an event table as an event node, and carrying out the prediction of the to-be-predicted object. And determining edges between the entity nodes and the event nodes according to foreign keys corresponding to the business events in the event table, and constructing a target topological graph according to the entity nodes, the event nodes and the edges between the entity nodes and the event nodes. Therefore, the target features of the to-be-predicted object are automatically extracted from the whole target topological graph, original data in all tables in the relational database are fully utilized, human intervention of feature engineering is reduced, the prediction efficiency is improved, the complex relation between the to-be-predicted object and other data in the relational database can be better captured, and the prediction accuracy is improved. And the accuracy of model prediction is improved.

The invention discloses a relational data importing method for large-scale multimode data, and belongs to the technical field of computers. Obtaining an original data file, a storage format and a data model; analyzing, reading and structuring the data information of the original data file through a data conversion module; generating a unified relation data model file from the data information through a block parallel writing method, and storing the unified relation data model file as a same file storage format CSV; performing preprocessing through a data association module; extracting mode information, sending the mode information to a mode conversion module, and automatically generating a relation table and establishing a DDL statement by the mode conversion module; and sending the DDL statement and the CSV file to an SQL processing terminal, uniformly converting the multi-mode data into relational data through the COPY statement and the DDL statement of the database, and storing the relational data into a bottom storage structure of the relational database. According to the method, the multi-mode data are uniformly stored in the same storage layer, so that query can be directly called from the database storage layer, and the data conversion efficiency is improved.

The invention provides a retrieval enhancement mode linking method and device based on multi-round iteration, relates to the technical field of natural languages, and can accurately select a first candidate table and a first table element related to input content by combining the input content, key information of the input content, description information of tables in a database, description information of fields and enumeration values. And the accuracy of the first candidate table and the first table element is improved. The first candidate table and the first table element are further screened by utilizing the large language model, so that the semantic comprehension capability of the large language model can be fully utilized, and the accuracy of the second candidate table and the second table element is further improved. Through a table relation directed graph of a database, a retrieval enhancement generation technology is utilized in a multi-round iteration mode, the complex incidence relation between tables is fully considered, the range of a mode link result is gradually expanded, it is ensured that the finally obtained mode link result can completely cover input content, and the recall rate and the accuracy rate of mode link are increased.

The invention relates to the technical field of knowledge maps, and provides a screening type dynamic classification self-adaptive relation extraction method, device and equipment, and the method comprises the following steps: defining an entity type list, inputting a text by a user according to a cue word template in a labeling format, judging the length of the text, and extracting the entity type list according to the length of the text; selecting a corresponding strategy according to the text length to generate a candidate entity pair; the association strength of each candidate entity pair is calculated by combining static corpus statistics and a dynamic semantic hybrid scoring model, and the candidate entity pairs with low association are filtered; constructing a dynamic prompt request LLM to generate a natural language abstract describing the subject entity and the object entity, and classifying the abstract into a predefined relationship type to obtain a structured triple; and importing the triple into a graph database to construct the knowledge graph. According to the method, the knowledge extraction efficiency is improved, and the problems of calculation redundancy and high cost are solved.

The invention discloses an intelligent mining method for dispatching knowledge of a water-wind-light complementary system based on a large language model, which belongs to the technical field of dispatching of the water-wind-light complementary system and comprises the following steps: S1, constructing an ontology model in the field of water-wind-light dispatching according to'demand traction-reuse verification-concept extraction-semantic modeling-formalized implementation '; s2, collecting and processing data; s3, the RoBERTa-BiLSTM-CRF architecture is trained, and entity mining is achieved; s4, realizing relation mining by combining a mixed relation mining method with a three-layer semantic constraint mechanism; s5, performing multi-dimensional fusion on entity semantics; s6, relying on Neo4j graph database storage, combining a man-machine natural language interaction interface of the generative LLM and realizing knowledge base iteration updating. According to the intelligent mining method for the scheduling knowledge of the water-wind-light complementary system based on the large language model, multi-source data collaborative fusion and mining are achieved, and core knowledge support is provided for multi-energy collaborative scheduling and optimization decision making.

The invention provides a cross-database approximate nearest neighbor search method and system and a calculation framework, and the method comprises the following steps: constructing a graph index, storing the graph index in a relation table, and obtaining a graph index table; acquiring and storing a data set and a query set in a structured relation table form; the dismantling approximate nearest neighbor search process comprises a plurality of SQL operation stages including a candidate node screening stage, a neighbor expansion stage, a distance calculation stage, a result combination stage and a priority queue maintenance stage; and based on the graph index table, executing an iterative search process of the plurality of SQL operation stages on each query point in the query set, and finally outputting an approximate nearest neighbor search result of the query set. According to the method, the graph index is combined with the relational database, and the approximate nearest neighbor search is realized by adopting a plurality of SQL operation stages, so that the high-dimensional vector retrieval efficiency and the cross-database compatibility are remarkably improved, and the large-scale application of the vector data in a multi-element scene is promoted.

The invention discloses a large model illusion suppression intelligent question answering system and method based on deep learning, and the method comprises the following steps: constructing a database Schema, and configuring an entity mapping relation; configuring cue word information reasoned by the large model to form a task cue word sequence; complete system input is formed; executing structured query language conversion by adopting a large model of an improved NASNet network architecture to generate an SQL query statement; generating natural language description information; performing consistency verification to obtain a consistency verification result; generating corresponding calculation prompt information; according to the method, the hallusion and semantic offset generated by a large model in a structured question and answer scene are effectively inhibited, and the accuracy of SQL generation and the reliability of the question and answer result are improved.

The invention provides a parameterized three-dimensional modeling method and system based on natural language and database driving, and belongs to the technical field of civil engineering modeling. According to the method, a performance function library (comprising a structure creation unit, a section operation unit and a parameter management unit) is constructed, a natural language dialog box is used for receiving a user instruction, the instruction is analyzed into modeling parameters by means of a natural language processing technology, and corresponding performance functions in the performance function library are automatically called. And then, according to a performance function calling result, generating information such as node coordinates, a geometric topological relation and section attributes, storing the information into a database, and finally, generating a three-dimensional model based on the database. According to the parameterized three-dimensional modeling method and system based on the natural language and database driving, professional software operation is not needed, the learning and modification cost is remarkably reduced, manual input errors are reduced, the modeling efficiency and the team cooperation ability are improved, and the parameterized three-dimensional modeling method and system can be widely applied to building, bridge, tunnel and steel structure engineering.

The invention discloses a production data query method and device based on a graph, and the method comprises the steps: extracting a metadata relation set from an equipment production database, and carrying out the semantic coding of a query text, and obtaining an instruction semantic vector; constructing an enhanced graph by the metadata relation set; aligning a preset optimized semantic vector with the enhanced atlas to identify an entity association path; generating a query instruction according to the entity association path; performing data query in the enhanced atlas by using the query instruction, and monitoring the state of data query in real time so as to update the enhanced atlas according to the state; and performing data query in the updated enhanced atlas by using the instruction semantic vector. The enhanced atlas constructed by the method not only can store the attributes of the entities, but also can show and store the association relationship between the entities in a displayed manner, so that efficient and accurate data query is realized.

The invention belongs to the technical field of intelligent question counting, and discloses a large model and knowledge graph combined intention recognition method and system and a medium, and the method comprises the steps: constructing a semi-structured language system MLS for a big data platform intelligent question counting scene, and taking the MLS as a semantic bridge of a natural language and an SQL (Structured Query Language); receiving and analyzing natural language query through a large model LLM, and extracting three elements including known conditions, a target object and a limited relation; performing semantic mapping and reasoning on the three elements by using a predefined knowledge graph containing sets, items, values and various semantic relationships in the MLS; natural language query is converted into an MLS expression containing operators such as I, C, V and Q according to the reasoning result; and finally, the MLS expression is translated into a relational calculation form and an SQL statement which can be executed by a database, so that intention recognition and conversion from a natural language to a computer instruction are realized, and the accuracy and efficiency of intelligent number asking are improved.

The embodiment of the invention provides a conversational intelligent analysis method and system fused with time-space relation modeling, and the method comprises the steps: carrying out the light ontology atomization extraction and time-space evidence binding of an input text, and generating a structural unit carrying an evidence anchor point; constructing a three-library index architecture in which a vector library, a graph database and a relational library are coordinated, and realizing responsibility separation of semantic retrieval, graph structure propagation and space-time constraint; when a user query is processed, time-space and semantic intentions are analyzed, time-space hard filtering and semantic hybrid retrieval are executed, a personalized initial vector fusing time-space semantic signals is constructed based on a retrieval result, and a context evidence set of steady-state sorting is obtained through propagation sorting of a time-space perception graph and is assembled into a structured evidence packet; an evidence packet constraint large model is used to generate an answer with reference and confidence; and performing machine self-inspection on the answers, starting attribution re-check on high-risk items, constructing a performance portrait based on audit feedback and driving self-evolution of strategy parameters to form a full-link audit closed loop.

A method for performing a backup in relation to a separated database, the method includes creating, by a processing circuit, a new backup log structured merge (LSM) tree of the separated database, wherein the new backup LSM tree is associated with a new point in time (PIT) and belongs to a group of LSM trees that comprise a primary LSM tree and one or more backup LSM trees that are associated with different PITs, and share a key value (KV) database that is mutable and is separated from the group; wherein the creating comprises storing the new backup LSM tree to a destination.

The present invention discloses a relation-enhanced retrieval method, device and readable medium based on a graph database. The method obtains a natural language question and performs word segmentation processing to obtain a word sequence, and uses a pre-trained language model to generate word vector features of the words in the question, word vector features of table names and word vector features of column names; a graph database is constructed based on the word vector features of the table names and the word vector features of the column names, and a self-attention mechanism is used to generate deep relation fusion features based on the word vector features of the words in the question and the graph database; the word vector features of the words in the question, the word vector features of the table names, the word vector features of the column names and the deep relation fusion features are spliced ​​to obtain a joint feature vector, and the joint feature vector is input into a trained Text2AQL model to obtain a syntax tree. According to the syntax tree, an AQL query statement is obtained by using the abstract syntax description language specification mapping, which can reduce the difficulty of interaction between users and the graph database and improve the user experience.

The invention discloses a multi-modal database-oriented adaptive index structure selection method, which comprises the following steps of: analyzing statistical characteristics such as data dimensions, variances, sparseness and distance distribution of each modal, calculating hidden dimensions of the modals, and describing effective data complexity of the modals; based on a preset index adaptation rule, automatically mapping each mode to an optimal index type in the candidate index structure set, and completing automatic construction of a local index and binding of the local index and a global routing structure according to the optimal index type; on the basis, the multi-modal query can be automatically routed to the corresponding index to execute retrieval according to the index mapping relation. According to the method, an index structure self-adaptive selection mechanism without manual configuration is realized, the index maintenance cost of the multi-modal database can be remarkably reduced, and the query efficiency and expandability in a complex retrieval scene are improved.