71 results about "Triplestore" patented technology

The application relates to a knowledge graph updating method and device, equipment, a storage medium and a program product. The method comprises the following steps: acquiring entity vectors of multiple triplets, relation vectors of the multiple triplets and a question vector of a to-be-queried question; determining multiple candidate tail entities corresponding to the question vector according to the question vector and the entity vectors in the initial vectors; determining first candidate triplets according to the multiple candidate tail entities corresponding to the question vector; determining second candidate triplets according to the first relation vectors and the question vector; and finally updating the knowledge graph according to the first candidate triplets and the second candidate triplets. The original knowledge graph triplet data is perfected by using the method, so that the knowledge graph is updated, and the probability of obtaining a question answer from the knowledge graph by a user is improved.

This invention relates to a method for fusion of heterogeneous knowledge in the power industry, comprising: obtaining first triples containing entity relations from the ontology layers of at least two knowledge graphs; classifying the entity relations in the first triples into synonym relations and inclusion relations according to relation aggregation prompts by a language model; merging entity relations belonging to synonym relations; obtaining second triples containing entity attributes or relation attributes from the ontology layers of at least two knowledge graphs; classifying the attributes in the second triples into synonym attributes and non-synonymous attributes according to attribute aggregation prompts by a language model; merging entity attributes or relation attributes belonging to synonym attributes; obtaining entities from the instance layers of at least two knowledge graphs and dividing them into several data blocks; determining whether there are duplicate entities according to entity alignment prompts by a language model; and merging duplicate entities.

The application discloses a domain knowledge triple extraction method, system, medium and equipment, and relates to the technical field of knowledge engineering.The domain knowledge triple extraction method and system provided by the application can provide high-quality and efficient data support for triple extraction by analyzing and converting the obtained domain professional text into a structured vector knowledge base which can be efficiently searched, then the domain professional text is subjected to quantitative calculation and fusion screening, the initial entities of domain core high-frequency words which have global high frequency and cross-text universality are mined, the problem of "local optimum and global deviation" caused by traditional single word frequency is avoided, non-core term interference is effectively avoided, and the stability, relevance and efficiency of iterative extraction are improved, and further, the high-frequency word initial entities are subjected to closed-loop iteration through iterative RAG algorithm combined with the structured vector knowledge base, accurate extraction of the large model and iterative correlation entities, the explicit and implicit semantic correlation triples in the text can be mined layer by layer, the strong relevance and structural integrity of the extracted triples are ensured, and the professionalism and accuracy of the model in extracting triples are improved.

The application provides a project declaration content compliance review method and system combining named entity recognition and large models, and relates to the technical field of natural language processing. First, a project declaration full-cycle document set is obtained, including declaration documents, contracts, mid-term examination and acceptance documents, a fine-tuned named entity recognition model is used to extract structured entity triples, and a data set containing logical association relationships is constructed. Then, a compliance review prompt engineering instruction set is constructed, structured entity triples are input into a large language model for multi-document entity alignment and consistency verification, numerical evolution, semantic deviation and logical contradiction features are identified, and identification results are generated. Finally, a traceable review evidence chain set is constructed according to the identification results, the source of the problem is accurately located, and the association with the compliance rules is clarified. The application improves the review efficiency, accuracy and systematicness.

The application provides a pre-training language model construction method, system and device, relates to the fields of artificial intelligence technology and semantic processing, and mainly comprises the following steps: based on a knowledge graph, a subgraph with a triple as a node unit is constructed; based on entity and relation description text, both are encoded through a conventional pre-training language model to obtain entity representation vectors and relation representation vectors; for the entity representation vectors and the relation representation vectors corresponding to the subgraph, modeling relation information is carried out based on a graph self-attention neural network, and the parameters of the graph neural network and the entity vectors are updated through a prediction link task; the parameters of the model are updated based on a contrast learning technology; and the parameters of the model are fine-tuned for a specific task. The scheme overcomes the problem that entity in the equipment field is sparse and a conventional pre-training language model cannot sufficiently learn entity semantics, and the updated pre-training language model has better understanding and reasoning capabilities in the application of the equipment field.

The application discloses a talent digital representation method and device based on semantic enhancement, and the method comprises the following steps: acquiring comprehensive environment information of neighbor nodes from a talent information database, and inputting the comprehensive environment information and triplets into a semantic enhancement intelligent agent to generate a semantic enhancement text with a fusion organization background, and then encoding the semantic enhancement text into an enhanced embedding vector by using a large language model, and inputting an initial embedding vector of a center node and the enhanced embedding vector of the neighbor nodes into a potential energy aggregation algorithm to obtain talent digital representation; and the application can obtain talent representation information which accurately reflects the real potential energy and situational value of talents in a business organization.

This invention relates to a method and system for multi-source information knowledge fusion and intelligent retrieval. The method includes: acquiring multi-source heterogeneous data and performing unified knowledge modeling on the multi-source heterogeneous data to obtain a unified knowledge framework. Based on the unified knowledge framework, a large language model is invoked to perform structured preprocessing on the multi-source heterogeneous data to generate structured knowledge triples. The structured knowledge triples are then structured using attribute graphs, and the associated data of each entity is vectorized and stored using an embedding model to obtain a fused knowledge body. User query requests are received and responded to by performing deep analysis on the user input data to obtain the user's query intent and query elements. A condition graph is constructed based on the user's query intent and query elements, and the condition graph is mapped to a knowledge network for entity localization and condition graph matching to generate structured retrieval results. This reduces network overhead and scheduling latency, and improves retrieval response speed and accuracy.

The application discloses a kind of high-efficiency lightweight federal recommendation methods, comprising: server initialization article continuous value embedding matrix and article bias vector, by binary processing to obtain binary article matrix and distribute to client;Client is based on local data, by difficult negative sample mining to construct training triple, adopts the hybrid loss function combined with sorting loss, reconstruction loss and regularization loss to carry out local training, updates user parameter and calculates article gradient;Server aggregates the gradient uploaded by client to update global parameter.The application enhances model expression ability by introducing bias term, optimizes sorting performance using hybrid loss function, improves training efficiency by combining difficult negative sample mining, significantly improves recommendation accuracy while retaining the privacy protection advantages of federal learning, while significantly reducing client storage, communication and computing overhead, especially suitable for resource-constrained mobile terminal deployment.

This application relates to a cloud-native resource description framework (RPF) data storage system, method, and computer device. The system includes: a storage layer for storing triples sharing the same attribute in RPF graph data as a subgraph and storing them in cloud object storage, constructing a range-aware offset index for each subgraph; and a computation layer for receiving RPF queries, decomposing the RPF queries into multiple subqueries according to attributes, querying the range-aware offset index, obtaining the byte offset of the target data value in each subquery in the corresponding subgraph, and, based on a query processing efficiency model, determining a target access method for each subquery from multiple preset data access methods according to the byte offset, executing the target access method to output the subquery result, and concatenating the results of each subquery to generate the query answer. Using this system can significantly reduce storage and query operating costs while maintaining competitive query performance.

This invention discloses a method and system for identifying innovative points in academic papers, relating to the field of artificial intelligence technology. The method first processes the paper abstract through a concept path extraction module, including: structured semantic segmentation, concept pair extraction and verification, constraint relation triple generation, hierarchical verification, and path optimization, to construct a complete set of concept paths for the paper. Then, in a rare path discovery module, the popularity index of each concept path in the global path frequency dictionary is calculated and compared with a threshold to identify rare paths as the scientific innovative points of the paper. This invention effectively suppresses the illusion of language models through a knowledge graph constraint mechanism, significantly improving the accuracy of concept path extraction and its coverage of long-tail concepts, achieving accurate and efficient identification of innovative points in academic papers at the path level.

The application discloses a knowledge triple construction method and device based on semantic distillation and a medium, and relates to the technical field of machine learning. The method comprises the following steps: performing feature extraction on the to-be-processed multi-modal data to obtain semantic vectors of a unified dimension, and inputting the semantic vectors into a unified decoder to decode and output a teacher triple containing a head entity, a relationship, a tail entity, an attribute dictionary and an evidence mask; inputting the teacher triple and an intermediate feature map corresponding to the teacher triple into a to-be-trained student network, and training the to-be-trained student network through a preset distillation objective function and a dynamic temperature annealing strategy until the to-be-trained student network converges, so as to obtain a lightweight student network; calculating the Fisher information value of the trained parameters in the lightweight student network to insert a lightweight adaptive module into the lightweight student network; updating the lightweight adaptive module based on the multi-modal data, and outputting knowledge triples in real time by the lightweight student network.

This invention discloses a method, apparatus, and storage medium for converting natural language to SQL in the securities industry, relating to the field of natural language processing technology. The invention first matches natural language with a knowledge graph to obtain a target sub-graph. Then, it extracts all triples from the target sub-graph and matches each triple with the original text fragment from the natural language. Each triple can be used as a smaller unit of the text fragment for semantic querying, thereby retrieving the original text fragment from the natural language. This invention uses triples in the target sub-graph to segment the natural language, thus filtering out the most reasonable M text fragments. Because the process of determining the text fragments is relatively reliable and reasonable, the table template determined based on the text fragments is also more reliable. Filling the table template with natural language yields a more reliable tabular language, and ultimately, based on the tabular language, a more accurate SQL statement can be obtained.

The application discloses a knowledge graph construction method and system, and relates to the technical field of knowledge graph construction.The knowledge graph construction method comprises the following steps: performing a streaming decoupling operation on a received unstructured document to obtain a pure text data stream and a visual data stream; performing semantic direct reading and extraction on the pure text data stream and the visual data stream to obtain a triple pool; performing induction and standardization processing on the triple pool to obtain standardized triple data; mapping the standardized triple data into nodes and directed edges; and constructing a knowledge graph according to the nodes and the directed edges.The knowledge graph construction method can improve the completeness and numerical accuracy of data extraction from a complex visual chart.

This invention discloses a method and system for extracting triples for knowledge graph construction. The method includes: learning multi-dimensional feature representations from input natural language text to generate deep semantic embedding vectors that integrate vocabulary, syntax, and discourse structure; dynamically generating relation-aware contextual feature representations based on the deep semantic embedding vectors; synchronously decoding entity boundaries and relation types using a dual-pointer network to output a preliminary set of candidate triples; performing graph structure consistency verification and conflict resolution on the candidate triple set, and calibrating confidence based on global knowledge distribution to generate final purified structured triple knowledge. Using this invention, the accuracy and reliability of triple extraction from complex text can be improved, providing an effective technical means for the automated construction of high-quality knowledge graphs.

A system and method are provided for explaining ontological sub-graphs. The system and method include querying an ontology to determine a match between a query graph and a portion of an ontology graph. When there is a match, a subgraph representing the match is first translated into a simple summary using a simple language (e.g., triplets which include a subject and object corresponding to pairs of connected nodes in the subgraph and a verb / predicate representing a relation / edge in the subgraph that connect the pair nodes). This simple summary is then fed, as part of a prompt, to a large language model (LLM) that generates a human-readable summary based on the prompt.

The application relates to the technical field of intelligent question and answer interaction, in particular to an intelligent question and answer interaction method fusing a private database and an enterprise knowledge graph, which comprises the following steps: in view of the influence of enterprise-specific terms and implied semantics on the connection of relations, semantic similarity, co-occurrence frequency characteristics and syntactic dependency degree characteristics in homologous text data are analyzed in depth to reflect the credibility of the three tuples formed between entity subwords; in view of the influence of the multi-source nature of enterprise data, the consistency of enterprise identification and the semantic equivalence characteristics between entity subwords are analyzed to analyze the confidence of the connection relationship between entity subwords in different source data; the completeness of the construction of the enterprise knowledge graph is improved, and the defects that the returned answers of the intelligent question and answer are not accurate enough and the fluency is low are remedied.

This invention proposes a method and system for causal verification and attack tracing in power systems, belonging to the field of power system network security technology. The method includes: collecting multi-source data from the power system; standardizing the multi-source data; extracting entity-relationship-attribute triples; cleaning the triples to generate a triple dataset; constructing a power attack tracing knowledge graph based on the triple dataset and combining entities and their inter-entity business relationships; selecting candidate event sets from the power attack tracing knowledge graph based on abnormal equipment events; constructing an attack causal graph based on the candidate event sets; verifying the causal relationships between events; and reconstructing the attack chain to locate the attack source. This invention improves the physical credibility of attack paths, the ability to identify attacks specific to power scenarios, and the supporting value and practicality of tracing results for defense decisions; it reduces the false positive rate caused by accidental correlations and improves the accuracy of attack chain and source location.

This application discloses a semantic communication method and a related apparatus. The method includes: determining a first set, where the first set includes at least one triple corresponding to to-be-transmitted data, and each triple includes a first head entity, a first tail entity, and a first relation between the first head entity and the first tail entity; determining a third set based on a probability corresponding to a second relation in a second set, where the second set includes at least one quadruple, each quadruple includes a second head entity, a second tail entity, a second relation between the second head entity and the second tail entity, and a probability corresponding to the second relation, the third set includes at least the first head entity and the first tail entity; and sending the third set to a receiving side to determine the first set.

This invention proposes a method, system, device, and medium for relation completion in cement-based materials, belonging to the field of cement-based composite materials technology. The method includes: constructing a set of triples for a material spectrum based on original text samples of the cement-based material's formulation, process, and properties; training an encoder based on the triple set to obtain a first relation encoding model; mixing unlabeled samples into the original text samples and using the first relation encoding model as a model base, semi-supervised training of the first relation encoding model using a self-adversarial loss function to obtain a second relation encoding model, and extracting the source node embedding, target node embedding, and relation embedding of all triples; constructing a positive and negative sample pair input self-interference decoder to predict missing relations in the material spectrum; traversing the material spectrum, inputting the embeddings of any two nodes, and completing the missing relations based on the self-interference decoder. This invention achieves relational semantic reasoning for cement-based materials, thereby improving the accuracy of relation completion.

The application discloses an open domain scientific knowledge discovery method and device based on a pre-trained language model, constructs an input template comprising a head entity, a first prompt, a second prompt and a tail entity mask, fills the pre-trained embedding of the head entity of each triple containing a target relation, the discrete tokens of the first prompt corresponding to the target relation and the second prompt tokens into the input template, processes the tail entity mask, forms input sample data, constructs a single pre-trained language model for each target relation, trains the pre-trained language model using the input sample data corresponding to the target relation, optimizes the embedding representation of the first prompt and the second prompt, and predicts the missing entity in the triple using the optimized embedding representation of the first prompt and the second prompt and the pre-trained language model, thereby improving the efficiency and accuracy of the pre-trained language model in discovering knowledge.

The application relates to the technical field of knowledge graphs, and discloses a knowledge graph updating method, device and equipment and a readable storage medium, which comprises the following steps: acquiring public text data related to a target enterprise; using a preset large model to analyze the public text data and generate a candidate triple set representing the target enterprise and its associated subject events; performing credibility evaluation and semantic alignment processing on the candidate triple set to obtain a high-credibility event knowledge set; performing conflict detection on the high-credibility event knowledge set based on preset knowledge graph constraints to determine a target knowledge set; and writing the target knowledge set into an enterprise risk knowledge graph to generate updated enterprise risk knowledge graph data. The application significantly improves the automation degree and accuracy of enterprise risk knowledge graph updating, reduces the cost of manual participation and rule maintenance, and provides more stable and reliable knowledge support for enterprise risk identification, early warning and analysis.

This invention belongs to the field of natural language processing technology, specifically relating to a ransomware question-answering method and apparatus based on attack chain stage constraints. The method includes: constructing a ransomware domain knowledge vector library with attack chain stage labels; parsing user queries and identifying target attack chain stages to generate structured query vectors; retrieving and filtering candidate knowledge fragments based on the target stage, and after stage consistency filtering and attribute metadata sorting, constructing an evidence subgraph using structured triples; completing conflict elimination and path completion under attack chain stage order constraints; fusing knowledge evidence and inputting it into a large language model to generate question-answering results. This invention improves the stage consistency of retrieval and evidence fusion through attack chain stage constraints, suppresses multi-stage information mixing and model illusion, and improves the accuracy and reliability of ransomware question answering.

This invention belongs to the field of code security detection and automated repair technology. Specifically, it is a closed-loop system for automated repair and verification of code vulnerabilities based on unit tests. It completes vulnerability root cause localization and feature extraction through program slicing and static analysis. Then, based on the vulnerability root cause, it generates reproducible test cases with multi-dimensional assertions and regression test cases that meet coverage standards. Candidate patches are generated through retrieval enhancement and invalid patches are filtered out by static pre-verification. Patch verification is completed in an isolated sandbox using core and auxiliary dual-dimensional rules. Failure results are classified into scenarios and structured semantic feedback is generated to guide iterative repair. Finally, successful cases are structured and stored in a knowledge base in the form of triples to achieve self-evolution. This invention solves the core problems of existing technologies, such as unstable reproducible test cases, low iterative convergence efficiency, inability to identify overfitted patches, and lack of self-evolution capabilities. It can achieve fully automated repair and verification of code vulnerabilities, meeting the requirements for production environment application.

Systems and procedures for a demonstration uncertainty-based artificial intelligence model for open information extraction. A large language model (LLM) can generate initial structured sentences using an initial prompt for a domain-specific instruction extracted from an unstructured text input (110). Structural similarities between the initial structured sentences and sentences from a training dataset can be determined (120) to obtain structurally similar sentences. The LLM can identify relational triplets from combinations of tokens from generated sentences using the structurally similar sentences (130). The relational triplets can be filtered based on a computed demonstration uncertainty (140) to obtain a filtered triplet list.A domain-specific task can be performed using the filtered triplet list (150) to support the decision-making process of a decision-making entity.

This application provides a method, system, and storage medium for expense reimbursement review, including parsing expense reimbursement application data packets to extract reimbursement reason information, invoice information, and supporting material information; constructing a triplet data structure containing business entities, invoice entities, and physical evidence; performing semantic vectorization and spatiotemporal attribute alignment operations on the business entities, invoice entities, and physical evidence based on the triplet data structure to generate a multidimensional feature vector set; performing cross-validation operations on the content consistency between invoice entities and business entities, the evidentiary consistency between invoice entities and physical evidence, and the spatiotemporal logical consistency between business entities and physical evidence based on the multidimensional feature vector set to generate a consistency verification result; and generating a review result in natural language format based on the consistency verification result. This invention can accurately identify hidden violations such as genuine invoices for fake purposes and invoices issued in other locations, significantly improving the accuracy and automation level of the review.

The application provides a hypernym-hyponym relation extraction method based on concept definition and data enhancement, which comprises the following steps: extracting concept pairs from natural text by using a keyword extraction technology, constructing concept triples based on the concept pairs and the hypernym-hyponym relations corresponding to the concept pairs, and taking the set of concept triples as a training data set; obtaining concept vectors in each triple in the training data set, offset vectors between the concept vectors, and vectors of concept definitions; constructing a hypernym-hyponym relation prediction model with the training data set as the input and the fused vectors of the offset vectors between the concept vectors, the concept vectors, and the vectors of concept definitions as the output, training the hypernym-hyponym relation prediction model according to the training data set and the fused vectors; obtaining a to-be-predicted concept triple in a test text, inputting the to-be-predicted concept triple into the trained hypernym-hyponym relation prediction model, and predicting whether the to-be-predicted concept triple has a hypernym-hyponym relation according to the output components.