49 results about "Relation classification" patented technology

The invention discloses an entity relationship recognition method based on rotation position coding and a global pointer network, which comprises the following steps of: coding an input Chinese threat intelligence text by utilizing a pre-training language model fused with the rotation position coding, and generating context semantic representation with enhanced position perception capability; based on the context semantic representation, decoding all possible entity spans and types thereof in parallel in a two-dimensional grid space through a global pointer network to generate an entity set; for a target entity pair in the entity set, constructing a structured input sequence containing entity position information; processing the structured input sequence by using a double-attention coding mechanism; and based on the output of the double-attention coding mechanism, determining the relationship type between the target entity pairs through a relationship classification module. According to the method, precise decoding of nested entities and robust recognition of cross-language terms are realized through geometric space mapping and a dynamic boundary optimization mechanism.

The invention discloses a large language model illusion detection method based on dynamic subgraph retrieval, relates to the technical field of model illusion detection, and aims to solve the problem of poor detection effect of a current illusion detection method. Comprising the following steps: acquiring and preprocessing a data set to obtain a training set and a test set; performing fine tuning according to the preprocessed data set to obtain a hop count predictor, a relation classifier, an entity classifier and a fact classifier; using cue words to extract an entity list from the question answer pair by means of a large language model; starting from entities in the entity list, extracting related evidence sub-graphs from the knowledge graph by utilizing a hop count predictor, a relation classifier and an entity classifier in a cooperative manner; and judging the correctness of the answer pair of the original question by using a fact classifier according to the obtained evidence sub-graph. According to the method, the evidence sub-graphs are dynamically retrieved through cooperation of a plurality of classifiers, and related sub-graphs of complex reasoning questions can be retrieved more accurately so as to judge illusion of question answer pairs.

The invention discloses a relation perception type two-channel entity relation extraction method. The method comprises the following steps of S1, encoding sentences through an encoder; s2, inputting the sentence vectors into a potential relation extractor, and extracting potential relation features; s3, inputting the potential relation characteristics into a dual-channel extractor, and extracting from two channels of head entity priority and tail entity priority respectively; and S4, judging the association among the subject, the relationship and the object through the bisimulation network, aggregating the entity pairs extracted by the two channels, and performing relationship classification on the entity pairs. The problems that relation recognition lacks entity semantic support and error prediction is irreversible in a relation priority type joint extraction method can be solved.

The invention discloses a visual scene graph generation method based on differentiable fuzzy logic reasoning, and solves the technical problems of semantic fuzziness, logic common sense deficiency and the like during long tail relation processing in the prior art. The method comprises the following steps: inputting a training image into a target detection module, and respectively outputting corresponding high-dimensional geometric embedding vectors to a relation classifier and a fuzzy mapping layer; outputting a visual prediction branch by the relation classifier; meanwhile, the fuzzy mapping layer outputs a fuzzy membership degree vector; the logic tensor reasoning module is combined with a common sense rule in an external knowledge base, simulates a logic reasoning process by utilizing a differentiable logic operator, and outputs a logic reasoning branch of which the relation of each pair of objects meets a preset logic rule in the training image; a gating residual fusion module performs weighted fusion on the visual prediction branch and the logical reasoning branch to generate a scene graph triple of the training image; and finally, training the network model, inputting a test image into the trained model, and outputting a corresponding visual scene graph.

The present application relates to the technical field of relation classification, and particularly relates to a small sample relation classification method based on a graph neural network, which proposes a small sample relation classification method based on graph learning for a relation classification task containing only a small amount of samples, uses a graph neural network in graph learning to construct a small sample relation classification method, uses a prototype network to further improve the effect of relation classification, simultaneously imitates the ability of humans to quickly learn new knowledge by using past learned knowledge through meta learning, reduces the dependence on a large amount of labeled data, and the model learns the generalization ability of adaptive learning between different categories through training, so that the method does not need to be changed when facing a new category for testing, and the learned generalization knowledge can adapt to the classification task of the new category.

The invention provides a document-level relation extraction method for fusing subgraphs and displaying a constructed reasoning path, and belongs to the technical field of natural language processing. The method comprises the steps that an input text sequence is converted into a word vector sequence through an encoder; constructing a heterogeneous graph comprising entity nodes, mention nodes and sentence nodes through a document graph construction layer; explicitly defining three reasoning paths in sentences, between sentences and integration; extracting a sub-graph from the periphery of the target entity based on the document graph and the reasoning path, and reasoning by applying an R-GCN network; global encoder feature information and local subgraph feature information are sent to a fusion relation classification layer for relation distribution probability prediction; and optimizing a weighted adaptive loss function, and dynamically adjusting loss contribution degrees of different types of samples. According to the method, the reasoning ability for complex relations, the long-distance dependence capture effect and the accuracy of long-tail relation recognition are improved, and meanwhile, the accuracy, efficiency and generalization ability of document-level relation extraction are remarkably improved.

The application provides a relationship extraction method and system based on entity replacement and floating mark, and relates to the field of natural language processing.The method comprises the following steps: performing entity replacement and floating mark processing on the original text to obtain a detection text; performing language feature extraction processing on the detection text to obtain a feature vector; performing vector splicing processing and language feature extraction processing on the feature vector to obtain an entity relationship score vector; performing probability distribution prediction processing on the entity relationship score vector to obtain a category probability distribution graph; and performing decoding processing on the category probability distribution graph.The entity pair of the relationship to be extracted is replaced with a fixed mark in the application, so that the expression of the relationship is more general, and the relationship classification error caused by the name interference of the entity pair is solved; meanwhile, special representations are added before and after the entity pair, and the floating mark is marked at the end of the sentence, so that the problem of poor model generalization performance caused by insufficient representation of entity information in the relationship extraction process is solved.

The invention relates to the technical field of natural language processing and mapping knowledge domain, and particularly discloses a mapping knowledge domain relation extraction method fusing a syntactic structure and a domain rule. The method aims at solving the problems of semantic understanding deviation and insufficient domain knowledge utilization in vertical domain relation extraction. The method comprises the steps that dependency syntactic analysis is conducted on a text, a syntactic dependency adjacency matrix is constructed, and meanwhile a rule adjacency matrix is generated based on domain rule base matching; carrying out weighted fusion and filtering on the two types of matrixes to obtain an enhanced adjacent matrix; inputting the matrix and a text vector into a graph convolutional network fused with a dependency type attention mechanism, and learning to obtain a node enhancement representation; and finally constructing an entity pair feature vector to finish relationship classification. Through explicit fusion of interpretable syntactic rules and domain priori, the accuracy and robustness of relation extraction in professional fields such as a power distribution network are improved. Experiments show that the model relation extraction F1 value reaches 86.07% and is improved by 1.31% compared with a baseline model with the optimal performance.

The invention belongs to the technical field of natural language processing, and particularly relates to an entity relation joint extraction method oriented to industrial knowledge graph construction and causal reasoning driving. The method comprises the steps of obtaining a to-be-processed industrial field unstructured text and inputting the to-be-processed industrial field unstructured text into a pre-training language model for shared coding to obtain a shared feature sequence; processing the shared feature sequence by adopting a dual-channel gating module to obtain an entity feature sequence and a relation feature sequence; fusing the entity feature sequence and the relation feature sequence by adopting a causal inference module to generate a causal enhanced feature sequence; inputting the causal enhancement feature sequence into a joint decoder for entity boundary identification and relationship classification to obtain an entity relationship triple; according to the method, the task features are separated, confusion is eliminated through causal reasoning, the accuracy and robustness of entity relation extraction in a complex industrial text are remarkably improved, and particularly the effect in the aspects of processing nested entities and multi-task feature interaction is prominent.

The application discloses a low-resource biomedical relation extraction method based on adaptive prompt learning and constraint enhancement, aiming at problems of low-resource annotation scarcity, poor adaptability of artificial prompts, and illegal output triplets, and constructs a double-branch training architecture of few-shot and full-shot; a VAE is used to realize soft prompt vector coding, compression and reconstruction through an adaptive prompt generation module, a gate mechanism is used to dynamically adjust the fusion proportion of prompts and texts, and context representation learning is completed through SciBERT; an entity extraction is realized through a span extraction based on a relation recognition module, a multi-label relation classification is used to predict the relation probability of entities; and a constraint enhancement decoding module filters illegal combinations and removes redundant triplets through a relation type constraint mask, entity-relation score weighted fusion, relation number prediction and pruning strategy. The application significantly improves the extraction accuracy in a low-resource scene, the output triplet structure is legal and consistent, and the application is suitable for biomedical professional texts and small-area field data scarce scenes.

The examples of the present disclosure provide a method and apparatus for classifying an entity semantic relation, a model training method and apparatus, an electronic device, and relate to the field of text identification technology. A first entity and a second entity in a corpus is determined and obtained. According to a first position distance between each word in the corpus and the first entity and a second position distance between each word in the corpus and the second entity, a feature vector corresponding to each word is obtained. Then, feature vectors corresponding to all words in the corpus are combined to obtain a model input vector corresponding to the corpus. Then an entity semantic relation type corresponding to the corpus is obtained by using the model input vector corresponding to the corpus an input of the entity semantic relation classification model.

The application provides a kind of entity and relation serial extraction method for text data.First, the text is preprocessed, and its word vector is extracted, then the word vector is input into the text entity extraction model BiLSTM-softmax to obtain the text with entity pair, then the text with ambiguity is disambiguated, finally, the text with entity pair is input into the relation discrimination model, and then the relation classification model is used to obtain the relation category of entity.The application can extract entity and relation in series for text data, and has the functions of long text processing, entity disambiguation and rich semantic function of anaphora resolution, and good entity extraction and relation extraction results are obtained.

The application discloses an electromagnetic space field entity relationship joint extraction method with additional time information, comprising the following steps: obtaining social platform open source text data by using a Python crawler technology; performing text cleaning and preprocessing on the obtained open source text to form a data set C; determining entity and relationship classification, and formulating an entity relationship sequence joint labeling strategy; labeling the data set C with a triple tag; the triple tag is (main entity Subject, relationship Predicate, and object entity Object); an entity relationship joint extraction model STERM based on fragment sorting is constructed, and an entity relationship triple prediction model is obtained through training; and a (main entity, relationship, object entity, and time) quadruple list is obtained through a sequential priority nearest matching algorithm. The method solves the problems of error accumulation and exposure deviation, entity nesting and relationship overlap, and the problem that an entity relationship triple cannot represent dynamic changes of a relationship.

Methods and apparatuses are described for generating a domain-specific knowledge graph from unstructured computer text. A computing device extracts unstructured computer text associated with pairs of entities from domain-independent documents, and trains an entity relationship classification model using a domain-independent knowledge graph and the extracted text. The computing device extracts other unstructured text associated with a first domain from domain-specific documents. The computing device identifies pairs of entities contained within the text for the first domain, and executes the trained model to determine a relationship between the entities in each pair of entities identified from the text for the first domain. The computing device generates a domain-specific knowledge graph using (i) the pairs of entities identified from the text for the first domain and (ii) the relationships between the entities in each pair of entities identified from the text for the first domain.

The application discloses a kind of multi-modal relationship extraction methods based on target-oriented description enhancement and dynamic weight, it is related to multi-modal natural language processing technical field.The method includes: obtaining original text, original image and corresponding target entity pair;Based on original text, target entity pair and preset directional guide prompt word template, target guiding description text is generated using large language model;Original image is input into target diffusion model, and semantic enhanced image is output;Original text, target guiding description text, original image and semantic enhanced image are input into pre-trained target relationship extraction network, and the relationship classification result for target entity pair is output.The application uses the target guiding description text constructed for target entity pair, can effectively drive target relationship extraction network to accurately extract the interactive features related to target entity pair in complex social media background, so as to improve the accuracy of multi-modal relationship extraction result and system robustness.

The invention discloses a method for determining a scene graph and electronic equipment. The method comprises the following steps: acquiring a spatial relationship between objects in the scene graph through an LMP; determining object classification among objects in the scene graph according to the spatial relationship; through layout-based reasoning LRM, based on the object layout information, carrying out relationship classification among representing objects; and according to the object classification and relation classification results, generating a final scene graph, and describing the objects in the image and the mutual relation thereof.

The invention provides an LLM-based industrial knowledge graph construction and intelligent question and answer method, and belongs to the technical field of industrial knowledge graphs. Multi-source material data are collected, an industrial knowledge graph is constructed, and standardized processing and entity linking are performed after user query is received; vector index query and full-text index query are executed in parallel to obtain candidate nodes, a thermodynamic-driven node importance propagation model is introduced to calculate material node temperature values as importance scores, the temperature scores and initial similarity scores are subjected to weighted fusion for reordering, meanwhile, a large language model is utilized to analyze query, a graph query statement is generated, and a graph query result is obtained; the correctness of statements is checked through the abstract syntax tree parser and the relation classifier, query is executed, natural language replies are generated, and the technical problem that the sorting quality of retrieval results is insufficient due to inaccurate node importance evaluation in industrial knowledge graph retrieval is solved.

This invention discloses a control method and system for incomplete multi-view data, belonging to the field of data recovery control technology. The invention extracts monitoring data features from each view and evaluates their quality; views below a threshold are marked as incomplete. Initial recovery results are then obtained by migrating samples from similar neighborhoods of other views. All features are mapped to a unified latent space to obtain fused semantic representation features. By comprehensively considering multiple factors such as missing data labels, data quality, spatiotemporal features, and semantic consistency, the fusion impact of each view is quantified. Then, based on the relationship between the quantified value and a set threshold, the fusion weights of each view are adaptively calculated. Finally, a global fusion representation feature is generated based on this weighted average, thereby reconstructing and recovering the incomplete view. Anomaly detection is performed, and alarm information is output, improving the accuracy of incomplete view data recovery results and solving the problem of reduced accuracy in existing technologies.

The application provides a network document relationship extraction method and system, the network document relationship extraction method comprises the following steps: replacing all mentions in each entity with the first occurrence of the mention; training a pre-training model using document-level contrast pre-training and mask language model pre-training respectively; using the trained pre-training model to encode the context of a document sample to obtain a vector representation of each token, converting the named entity recognition information and part-of-speech tagging information of each token into a vector representation respectively and splicing the vector representation of each token to obtain a word embedding representation and a document embedding representation; performing representation enhancement on the entity to obtain a final embedding representation; using a binary cross-entropy loss function to calculate the possibility of the existence of a relationship between entities, and selecting the relationship with the highest possibility as the result of relationship classification. The application can effectively extract the relationship of network documents and provide effective support for network text processing.

The application provides a target action prediction method based on a time sequence knowledge graph, comprising: acquiring multi-source data, the multi-source data comprising a plurality of entities; obtaining the correlation between the entities based on a relationship classification model, and completing the redundancy check and alignment deduplication of the correlation, wherein the relationship classification model is obtained through classification learning; after determining a target entity in the plurality of entities according to the position and characteristic information of the entity, dynamically constructing a correlation tree of the correlation between the target entities to complete the construction of the time sequence knowledge graph; based on the time sequence knowledge graph, modeling and classifying the time sequence nonlinear correlation between the target entities by using a deep neural network to generate a target space evolution model; and predicting the behavior mode and motion trajectory of the target entity based on the target space evolution model.

The invention relates to the technical field of port safety management, in particular to a ship navigation risk alarm information intelligent identification method and system. According to the method, when entities and relationships are extracted, iterative updating of a graph attention network of directional entity perception is performed on an initial node feature matrix, and a final high-order node feature matrix is obtained; inputting the final high-order node feature matrix into a classifier, and performing entity recognition and relationship classification; in an iterative updating link, a brand-new final high-order node feature matrix is output, and each row in the matrix is a depth feature which is rich in global context information and directional entity perception and customized for entity recognition and relationship classification. And the depth feature is directly input into a subsequent classifier to finish a final extraction task, so that the extraction precision and robustness of the complex alarm statement are remarkably improved.

In a model training method, feature extraction is performed on a plurality of words in a first training sample to obtain first feature vectors of the plurality of words, the first training sample including at least a first part of training samples for pre-training a first model. Relation classification is performed to obtain a first classification probability of the first training sample. A plurality of sample-relation pairs formed by the first training sample and each relation of a plurality of second relations are determined. Similarity calculation is performed to obtain similarities corresponding to the sample-relation pairs. A first weight of a sample-relation pair is determined. A first loss of the first model is determined based on the first weight of the sample-relation pair and the first classification probability. A parameter of the first model is adjusted based on the first loss.

The application provides a remote supervision relationship extraction method based on multi-granularity contrast learning, and relates to the technical field of information extraction. The method comprises the following steps: obtaining a sentence in a data sample; segmenting the sentence into a word sequence; marking the head and tail of the word sequence respectively to obtain a marked sentence; obtaining initial sentence features by using a bidirectional encoder; obtaining initial positive sample features by using a positive example pair encoder; obtaining constraint features of a relationship and an entity pair according to a graph encoder and fusing the constraint features into corresponding initial sentence features and initial positive sample features to obtain enhanced sentence features and enhanced positive sample features; fusing the enhanced sentence features into bag-level features; inputting the bag-level features and the enhanced positive sample features into a contrast learning relationship classifier to extract the relationship of the data sample and obtain a classified relationship. The application solves the problem of low accuracy and efficiency of relationship extraction and weak long-tail relationship extraction capability caused by ignoring the interaction between different hierarchical features.

The application discloses a kind of based on global transformation prototype network's small sample relation classification method and device, belong to natural language processing technical field.The method mainly includes: (one) based on pre-training language model and entity mark's relation instance representation extraction mechanism;(two) based on global transformation prototype network of relation instance representation;(three) based on the global prototype representation obtained by the above network, the small sample relation classification method of being carried out.The pre-training language model and entity mark's relation instance representation extraction mechanism of the application can be fully encoded to the entity pair and relevant context included in relation instance;Based on global transformation's prototype network can be fully obtained by support set Global feature of new task, based on this, new relation prototype representation is transformed, so that better domain generalization is carried out.

The application discloses a kind of based on knowledge graph's work order whole process electronization management method and system, it is related to work order management technical field, including: definition work order management field core entity, entity interrelation to obtain core entity relationship chain, determine each entity attribute and construct field ontology diagram;Structured and unstructured data are divided and respectively preprocessed, and entity-relation-attribute chain is obtained;Accordingly, build initial knowledge graph, after its service is connected with work order management system;Integrating work order initiation channel using system, data acquisition, data are classified according to initial graph entity-relation and are matched with optimal resource, and the standard graph of optimization graph is obtained;Optimized solution and knowledge push list are generated in combination with standard graph, and work order processing process information is recycled;Archive work order and secondary optimization graph obtain preferred standard graph.The advantage lies in: realize work order whole process intelligentization, improve processing efficiency, guarantee knowledge sediment reuse.

The application belongs to the technical field of data mining, and provides an entity relationship joint extraction method and system based on an interactive double-affine mechanism.The application divides the entity relationship joint extraction task of text data into three subtasks of subject entity identification, object entity identification and subject-object pair relationship classification which can be jointly learned.In the entity identification and relationship classification stage, through a preset double-affine mechanism model, the subject, the object and the subject-object pair relationship can be respectively mapped into a subject entity table, an object entity table and a subject-object pair relationship classification table.The subject entity identification, the object entity identification and the subject-object pair relationship classification can further interact bidirectionally at the subtask level, improve the entity relationship extraction effect, and solve the problems of error propagation, entity redundancy and insufficient information interaction between subtasks in the identification and classification process.

The application relates to the technical field of pipe network data processing, and discloses a pipeline network spatial topological relation analysis method and system based on a graph neural network, which comprises the following steps: collecting pipe section objects, pipeline nodes and constraint objects. A spatial topological index structure is constructed, which comprises a spatial block index, a topological adjacency index and a standard constraint index. A constraint template is selected according to a pipeline category and a topological relation type to determine candidate object pairs and calculate topological indexes, and finally, a preliminary topological relation classification result is obtained. The relation classification input feature is input into a relation classification model, feature transmission and aggregation are carried out on a graph structure formed by the pipe section objects, the pipeline nodes and the constraint objects, and a topological relation classification result is output. Version identifiers are set for spatial blocks and candidate object pairs, and a spatial topological relation analysis result of the pipeline network is determined according to pipeline network data update conditions. The application improves the accuracy of topological rule violation identification and reduces the consumption of computing resources.

The invention provides a compiling period code generation method and system and electronic equipment, and the method comprises the steps: obtaining a source code, generating an abstract syntax tree of the source code, extracting annotation information based on the abstract syntax tree, carrying out the classified storage of the annotation information, and obtaining a first mapping relation corresponding to class annotations and a second mapping relation corresponding to field annotations, the structural bodies in the abstract syntax tree are subjected to description conversion in a classified mode according to the first mapping relation and the second mapping relation, metadata used for describing the structural bodies are obtained, the metadata is associated with corresponding preset template files according to the second mapping relation, and the preset template files comprise template codes of plain texts. Performing code format conversion on the metadata according to an associated preset template file to obtain a to-be-compiled code file, and compiling the to-be-compiled code file to obtain an executable file corresponding to the source code; according to the method, code generation is realized in a non-intrusive manner in the compiling period, and the consistency, efficiency and safety of code generation are improved.

The application provides a method, device and storage medium for entity structuring and disambiguation, the method comprising: collecting text data; inputting the text data into a first training model to output label classification of all first entities; when the label classification is unambiguous, inputting the text data into a knowledge base to output the closest result; when the label classification is ambiguous, inputting the text data into a second training model to output relevant relationship classification and another first entity corresponding to each relationship; when there is an equivalent relationship, inputting the another first entity into the knowledge base to output the closest result; when there is only a similar relationship, inputting the first entity and the another entity into the knowledge base to output the closest result as the disambiguation result of the ambiguous first entity. The method can further supplement and structure the first entity in the text data by using the correlation of the entities and combining the knowledge base, and can effectively eliminate the ambiguity of the ambiguous entity.

The application provides an acquisition method and device of a commodity attribute word hierarchical relationship dictionary, and relates to the technical field of data processing. The method comprises the following steps: extracting candidate commodity attribute words from multiple data sources, performing clustering analysis on the candidate commodity attribute words, and obtaining a first commodity attribute word of any cluster; for the first commodity attribute word of any cluster, grouping the first commodity attribute words two by two to obtain a commodity attribute word pair; inputting any commodity attribute word pair into a target classification model to identify the hierarchical relationship type, so as to obtain the hierarchical relationship of any commodity attribute word pair and generate a hierarchical relationship dictionary of commodity attribute words. In the application, the coverage of commodity attribute words is improved, the hierarchical relationship between commodity attribute words is fully mined, rich samples are provided for subsequent full-potential hierarchical relationship classification judgment, the accuracy of acquiring commodity attribute hierarchical relationships is significantly improved, and more commodity attribute hierarchical relationships can be mined.