25 results about "Database schema" patented technology

A system for decentralized database management includes a graphical user interface (GUI) layer configured to receive user actions for database operations. The GUI layer includes an input mapping module to translate user actions into backend commands, a real-time monitoring module to display metrics related to node status, database health, and blockchain transactions, and an error handling module to categorize errors for troubleshooting. The system includes a hybrid database system integrated within a decentralized blockchain network supporting both relational and non-relational data models, with value contracts defining database schemas and rules, and an indexing mechanism for optimizing data retrieval. A Self-Delegated Proof of Equal Stake (SDPoES) consensus mechanism is provided to validate and synchronize database operations across nodes, including a dynamic scheduling engine to coordinate block production among master nodes and a stake validation system to manage master node qualification through memory staking requirements.

The present disclosure presents systems and related methods for optimizing extraction, transformation, and loading (ETL) operations. One such method comprises receiving an extraction, transformation, and loading (ETL) job that is in current deployment on an ETL job database; obtaining ETL job metadata for the extracted ETL job; obtaining database schema for input data sources in the ETL job; using the database schema, parsing the ETL job to identify structural components of the ETL job; providing, in a graphical user interface display, one or more recommendations to perform one or more optimization actions for at least one structural component of the ETL job; and / or performing the one or more optimization actions for the at least one structural component of the ETL job.

A method for natural language to SQL (NL2SQL) statements driven by a hybrid approach combining templates and a large language model includes the following steps: constructing a pre-built SQL template library to assist the large language model in quickly matching syntactically correct SQL template skeletons; constructing a database schema information knowledge base to assist the large language model in accurately extracting entities; constructing a large-model natural language parsing and entity extraction engine to extract entity information; constructing a large-model-based template matching engine to match corresponding target SQL templates from the template library using the large language model; and constructing a large-model-based template filling engine to fill the extracted entity information into the corresponding placeholders of the selected target SQL template, generating the final executable SQL statement. By using pre-built SQL templates to ensure syntactic correctness and utilizing the large language model to parse natural language and extract entities to ensure semantic flexibility, this method solves the syntax errors and illusion problems that may occur when generating SQL using pure LLM, significantly improving the practicality and accuracy of NL2SQL technology.

The application provides a Text-to-SQL database preprocessing method fusing a hash fingerprint and a semantic vector, and belongs to the natural language processing and database technology fields. The method adopts a two-stage cooperative retrieval mechanism to construct a structured context. First, in the initial full-database retrieval stage, a multi-level screening strategy is used to locate the database entities and specific data lines related to the user query, and factual evidence is provided. Secondly, in the database schema understanding stage, combined with the original query semantics, the associated table structure and column description semantic information are obtained through vector retrieval to provide conceptual support. The two stages dynamically transfer and fuse the context through a state sharing mechanism. Through the cooperative preprocessing of instance data and schema description, the application generates a complete context input containing specific instances and abstract concepts, effectively reduces the illusion risk of large models, and significantly improves the accuracy of SQL generation in complex scenarios.

Methods and systems for retrieving data from at least one database based on a query input related to a clinical trial. A system processes a received query input related to a clinical trial using a pre-trained language model neural network. The neural network generates a structured representation of the query input. The system maps a first data field of the structured representation to a first column name and maps a second data field of the structured representation to a second column name. The system generates a database query based on (i) a database schema, (ii) the first column name, (iii) the second column name, (iv) data values associated with the first data field, and (v) data values associated with the second data field. The database query specifies an operation for joining data associated with the first column name with data associated with the second column name.

The application discloses a low-redundancy minimal data table design method and system for industrial ERP / MES, and belongs to the technical field of industrial digitization and database architecture. The application realizes low redundancy, low association, high performance and easy maintenance of the data table through full-service field collection, default single-table aggregation, controllable split constraint, weak association mapping, unified data entry and minimal query control. The application forces single-table priority, splits at most into two tables, and performs at most one association in regular query, thereby solving the problems of excessive split, complex association, poor performance and difficult maintenance in traditional design from the root, and has the advantages of minimalism, high efficiency, stability and easy expansion, and is suitable for various ERP / MES / WMS industrial management systems and has been applied in actual production environment.

This invention discloses a deep learning-based intelligent question-answering system and method for suppressing illusions in large models, comprising the following steps: constructing a database schema and configuring entity mapping relationships; configuring prompt word information for large model inference to form a task prompt word sequence; forming a complete system input; using a large model with an improved NASNet network architecture to perform structured query language transformation and generate SQL query statements; generating natural language description information; performing consistency verification and obtaining consistency verification results; generating corresponding computational prompt information; obtaining the query results, and outputting the query results and computational prompt information together as intelligent question-answering results. This invention effectively suppresses illusions and semantic shifts generated by large models in structured question-answering scenarios, improving the accuracy of SQL generation and the reliability of question-answering results.

The present application discloses a database query method and apparatus, an electronic device, and a non-volatile storage medium. The method comprises: determining a knowledge graph corresponding to a database to be queried, wherein the knowledge graph is used for representing a logical structure and an association relationship of data in said database; determining similarity scores between user question text and graph nodes in the knowledge graph, and determining a target node from among the graph nodes of the knowledge graph on the basis of the similarity scores, wherein the similarity scores are used for representing the degree of association between the graph nodes and the user question text; and on the basis of the target node, generating database schema information corresponding to said database, and using a large language model to generate, on the basis of the database schema information, a structured query language statement corresponding to the user question text.

A system for decentralized database management includes a graphical user interface (GUI) layer configured to receive user actions for database operations. The GUI layer includes an input mapping module to translate user actions into backend commands, a real-time monitoring module to display metrics related to node status, database health, and blockchain transactions, and an error handling module to categorize errors for troubleshooting. The system includes a hybrid database system integrated within a decentralized blockchain network supporting both relational and non-relational data models, with value contracts defining database schemas and rules, and an indexing mechanism for optimizing data retrieval. A Self-Delegated Proof of Equal Stake (SDPoES) consensus mechanism is provided to validate and synchronize database operations across nodes, including a dynamic scheduling engine to coordinate block production among master nodes and a stake validation system to manage master node qualification through memory staking requirements.

Described herein is a cloud-based omnichannel customer interaction service, also referred to as a contact center service, that deploys an improved data model for capturing customer interaction data. From the data model, a database schema is instantiated to include a set of related database tables. The arrangement and relationship of the data and the database tables improves the functioning of the contact center service in a variety of ways, to include seamless concurrent communications between customer and agent across multiple channels, and seamless transitions between agents, who can be added to a customer interaction with relative ease.

A computer implemented method for updating a database schema. The method comprises reading (101a, 101b) and then comparing (102) first and second database schemata. Tasks are then generated (103) in dependence on the comparison. The tasks are for converting a database defined by the first schema into a database defined by the second schema. The tasks are then executed (104) in a defined order in order to carry out the database update.

A system for decentralized database management includes a graphical user interface (GUI) layer configured to receive user actions for database operations. The GUI layer includes an input mapping module to translate user actions into backend commands, a real-time monitoring module to display metrics related to node status, database health, and blockchain transactions, and an error handling module to categorize errors for troubleshooting. The system includes a hybrid database system integrated within a decentralized blockchain network supporting both relational and non-relational data models, with value contracts defining database schemas and rules, and an indexing mechanism for optimizing data retrieval. A Self-Delegated Proof of Equal Stake (SDPoES) consensus mechanism is provided to validate and synchronize database operations across nodes, including a dynamic scheduling engine to coordinate block production among master nodes and a stake validation system to manage master node qualification through memory staking requirements.

This invention relates to the field of artificial intelligence technology and discloses a workflow intelligent orchestration method based on database MCP service, including the following steps: S1, user input; S2, LLM semantic parsing; S3, database schema knowledge base invocation; S4, SQL generation; S5, database MCP service invocation; S6, AI output analysis results. This method breaks down technical barriers, enabling non-technical personnel to participate in the construction of complex data workflows, significantly reducing the technical threshold and development costs, and achieving business-driven agile development. By providing a unified interactive entry point through visualized intelligent nodes, it achieves "zero-code" rapid delivery of data query functions, significantly improving development and orchestration efficiency. By decoupling business logic (natural language description) from execution logic (SQL code) and managing metadata through a centralized knowledge base, it greatly enhances the system's flexibility and maintainability, easily responding to business and data changes.

The disclosure provides a structured query language generation method, which comprises: obtaining original query text input by a user; rewriting the original query text based on a large model to obtain rewritten query text; extracting sub-database schema information related to the rewritten query text from a database based on the rewritten query text; and generating a target structured query language statement based on the rewritten query text and the sub-database schema information. Thus, by rewriting the original query of the user first, ambiguities in the original expression, such as colloquialism, specialization or context dependence, can be eliminated, making the query more matched with the database structure; then, based on the rewritten query, sub-database schema extraction is performed, only retaining table structure information related to the current query, greatly reducing the input size; finally, the SQL statement is generated based on the simplified and highly relevant information, which can effectively reduce the model hallucination risk and significantly improve the accuracy and reliability of the generated SQL.

The application discloses a data processing method and device, network equipment and a storage medium. The method comprises the following steps: in the case that a database architecture corresponding to a database comprises X database shards and agent middleware, receiving a command sent by a user end, wherein the command comprises a read command or a write command, and X is an integer greater than 1; splitting the command according to the command content of the command to generate Y sub-commands, wherein the Y sub-commands correspond to Y databases in the X database shards, the Y databases store data information related to the Y sub-commands, Y is less than or equal to X, and Y is an integer; and executing the Y sub-commands in the Y database shards.

One or more embodiments of the specification provide a multi-copy data storage method, device, equipment and medium. The method comprises: obtaining a pre-set copy deployment mode in a target data table in a database; in the case where the copy deployment mode is not set, obtaining a copy deployment mode indicated by a configuration file of the database; setting storage formats of multiple nodes according to the copy deployment mode; and storing data in the target data table in the nodes according to the storage formats of the nodes. By setting the row storage and column storage nodes in the same set of database architectures, different requirements for the storage formats under different query load scenarios can be met, and compared with setting the row storage and column storage in different databases, the maintenance of the database can be realized by using a set of software systems, and the nodes with different storage formats do not need to rely on data synchronization tools outside the database, which is easy to monitor and maintain.

An embodiment may involve: identifying an entry in a database table representing an individual; determining, based on one or more capabilities database tables, a set of current capabilities of the individual; determining, based on one or more target capability database tables, a set of target capabilities of a target role; identifying, based on the set of current capabilities and the set of target capabilities, a capability gap of the individual; based on the capability gap, identifying steps that the individual can take to satisfy the target capabilities; and generating a graphical user interface component identifying the capability gaps or the steps.

The invention discloses an automatic discovery method for database mode association for penetrating audit. The table field portrait data and the topological compatibility matrix are used for performing dual verification on the proposal association pair set, so that the illusion problem generated when a large language model processes a numerical type can be avoided, the proposal association pair set can be summarized to generate a global context, the conflict between transmission redundancy and a close-to-parent structure is eliminated, and the accuracy of the whole context is improved. Then, the logic rationality of a graph structure is guaranteed through global context auditing, so that key auditing clues are not missed, complex semantics of auditing data can be coped with, the accuracy and recall rate of association discovery are improved, meanwhile, the association mining problem caused by foreign key missing is avoided, and the accuracy and recall rate of association discovery are improved. Therefore, the accuracy of the restored auditing evidence chain is greatly improved, the bottom layer quality of data governance and knowledge graph construction is ensured, the auditing evidence obtaining result is stable and controllable, and high efficiency and leakproofness are achieved.

The application discloses an intent structured SQL sentence generation method based on a large language model, and comprises the following steps: obtaining a natural language query sentence input by a user, database schema information corresponding to a target data source, and semantic layer resources; performing normalization processing on the natural language query data; inputting the normalized query sentence into the large language model; constructing a retrieval unit set based on a query intent intermediate representation, and performing retrieval in a semantic layer resource set; based on the query intent intermediate representation and the semantic retrieval result set; generating an SQL main skeleton based on a compilable query intent intermediate representation and a template; calling an improved PICARD algorithm to perform a parser-in-the-loop multi-dimensional constraint incremental decoding processing on a to-be-generated SQL structure; and filling the constrained SQL segment into the to-be-generated SQL structure to generate a target SQL sentence. The application adopts an intent structuring and constraint decoding mechanism to realize controllable generation from natural language to SQL.

This application discloses a database cardinality estimation method and program product based on query embedding, relating to the field of database processing technology. The database cardinality estimation method based on query embedding extracts features of data distribution information through a data distribution embedding module, enabling cardinality estimation to perceive changes in data distribution in real time. It also extracts semantic features of database query statements through a query statement embedding module, allowing cardinality estimation to rely on stable features in the database schema (such as data distribution and column correlation) for inference even when features are missing, reducing the risk of overfitting the cardinality estimation model to surface-level query semantics. Furthermore, it identifies the correlation between query semantics and data distribution through an attention mechanism, and the features of both are weighted and fused to generate a query feature vector. This generated query feature vector simultaneously contains the semantic logic of the query intent and the real-time state of the data environment, improving the robustness of cardinality estimation.

A system for decentralized database management includes a graphical user interface (GUI) layer configured to receive user actions for database operations. The GUI layer includes an input mapping module to translate user actions into backend commands, a real-time monitoring module to display metrics related to node status, database health, and blockchain transactions, and an error handling module to categorize errors for troubleshooting. The system includes a hybrid database system integrated within a decentralized blockchain network supporting both relational and non-relational data models, with value contracts defining database schemas and rules, and an indexing mechanism for optimizing data retrieval. A Self-Delegated Proof of Equal Stake (SDPoES) consensus mechanism is provided to validate and synchronize database operations across nodes, including a dynamic scheduling engine to coordinate block production among master nodes and a stake validation system to manage master node qualification through memory staking requirements.

The application belongs to the field of information technology and relates to a database schema linking method and device based on sampling optimization. The method comprises the following steps: inputting a database schema and a user question into the context of a large model; obtaining a diversity-enhanced sampling result of the large model through database schema enhancement and noisy inference of the large model; obtaining a database schema linking result according to the diversity-enhanced sampling result of the large model; constructing a database schema relationship graph; and extending the database schema linking result based on the database schema relationship graph. The application effectively improves the overall performance of the database schema linking technology based on the large language model, can significantly improve the recall rate while maintaining a high accuracy, can promote the popularization and development of the automatic schema linking technology in practical applications, and can lay a solid foundation for various downstream tasks such as database question answering and information extraction.

This invention discloses a document information extraction method, system, and program product that combines logical constraints and reinforcement learning, relating to the fields of artificial intelligence and intelligent document processing technology. Firstly, by explicitly incorporating arithmetic rules into the reward function, this invention significantly reduces "numerical illusions" and substantially improves logical consistency. Secondly, it solves the problem of extracting borderless tables, deriving the table structure through semantic content and logical constraints instead of relying on visual dividing lines, making it particularly suitable for borderless complex tables such as medical invoices. Thirdly, this invention uses reinforcement learning to transform discrete business rules into dense supervisory signals to guide the optimization of the document information extraction model. Furthermore, the document information extraction model of this invention directly outputs JSON format conforming to the database schema, eliminating the need for complex post-processing steps and achieving direct conversion from images to a queryable database.