52results about How to "Improve semantic comprehension" patented technology

The invention provides an enhanced document generation and retrieval method based on a knowledge graph, and relates to the technical field of knowledge graphs, and the method comprises the following steps: receiving heterogeneous document input, and carrying out dynamic segmentation on the heterogeneous document; constructing a Graph-RAG model, wherein the Graph-RAG model comprises a knowledge graph-document index extraction module and a design document generation module; constructing and updating a knowledge graph based on document segmentation through a knowledge graph-document index extraction module; extracting feature vectors of the document segments by using a feature extraction model, establishing a mapping relationship between the document segments and entities in the knowledge graph through a matching strategy, and constructing a retrieval index; and receiving a query sequence of a user, retrieving the related entities and document segments through the retrieval index, and generating a target design document in combination with a preset template. According to the method, accurate retrieval and high-quality document generation based on semantic understanding are realized, and the technical problems of insufficient semantic understanding, low knowledge association degree, poor generation quality and the like in a traditional document processing method can be solved.

The invention provides a real-time data analysis and visualization method and system in a big data environment, and relates to the technical field of data processing, and the method comprises the steps: obtaining and preprocessing a to-be-processed document; carrying out hierarchical structure analysis and semantic understanding analysis; semantic index mapping is established; receiving a natural language processing instruction input by a user; performing intention understanding and semantic analysis; matching a semantic understanding task; processing the corresponding semantic fragment units; and obtaining the target document and visually displaying the target document. According to the method, intelligent processing and analysis of different types of documents can be realized, and the data processing efficiency is improved.

The invention discloses a combined visual question and answer method based on core-to-global semantic fusion reasoning, and belongs to the technical field of image and text processing, and the method comprises the steps: obtaining image data and text question data; inputting the image data into the semantic feature extraction model to obtain image global features and visual core features; inputting the text problem data into the semantic feature extraction model to obtain text problem global features and text core features; transmitting the image global feature, the visual core feature, the text problem global feature and the text core feature to a preset multi-modal semantic fusion model to execute multi-modal semantic learning of different levels; fusing the learning results of the single-level language; and according to the target answer prediction vector, calculating to obtain a target prediction answer. According to the method and the device, the technical problems of inaccuracy in reasoning and model instability caused by inaccurate semantic alignment, feature redundancy and noise interference of multi-modal information in the prior art are solved.

The invention provides a method for creating a structured document based on a deep learning model, and relates to the technical field of data processing, and the method comprises the steps: obtaining basic text information collected by a user, obtaining text data, fragmenting the text data, obtaining text fragmented data, and creating a structured document according to the text fragmented data. Analyzing the structures of the sentences in the paragraphs, determining keywords in the sentences, analyzing the logical relationship between the sentences in the paragraphs, associating the sentence keywords in the paragraphs, analyzing the association degree between the paragraphs in the chapters, identifying core keywords of the chapters, obtaining a chapter keyword data set, and analyzing the field involved in the text. The method comprises the steps of extracting field information of a text to obtain a text keyword data set, constructing a structured document model based on a graph neural network model, and inputting the text keyword data set and text data into the structured document model to obtain a structured document. According to the method, the structured document can be constructed through the deep learning model.

The invention relates to the technical field of panoramic video processing, in particular to a multi-modal fusion panoramic video intelligent content analysis method, which comprises the following steps of: acquiring a panoramic video frame and a synchronous audio thereof; performing spherical projection transformation on the panoramic image data, and converting the equidistant projection image into a spherical image; generating a plurality of plane images based on the spherical image; performing candidate target position detection based on the plurality of plane images; based on a candidate target position detection result, an image target detection model is utilized to realize instance segmentation of a target, and target features are extracted to generate a target embedded description vector sequence; based on the synchronous audio, an audio recognition processing module is used for carrying out time sequence information extraction, and an acoustic attention sequence is formed; a spectrum sensor is adopted to realize association fusion of the target embedded description vector sequence and the acoustic attention sequence, and an audio and video collaborative attention sequence is generated; and based on the audio and video collaborative attention sequence, establishing an audio and visual event identification model, and obtaining event content described by the panoramic content.

The invention discloses an intelligent question and answer method and system based on semantic vector optimization and dynamic prompt, and the method comprises the steps: inputting a target question into a trained question and answer model, and obtaining an answer corresponding to the target question; wherein the trained question and answer model comprises an input layer, an embedded layer, a semantic vector generation module, a feature compression module, a dynamic prompt generation module, an answer generation module, a linear layer and an activation function layer which are connected in sequence; wherein the output end of the feature compression module is further connected with the input end of the RAG retrieval module, and the output end of the RAG retrieval module is connected with the input end of the dynamic prompt generation module; according to the trained question and answer model, the training process comprises the steps that a semantic vector generation module is constructed, and the semantic vector generation module is pre-trained; and setting the pre-trained semantic vector generation module in the question and answer model, training the question and answer model, and when the total loss function value does not decrease any more, stopping training to obtain a trained question and answer model.

The invention relates to a multi-modal false information detection method based on semantic consistency, and aims to solve the problem of semantic interaction between modals in multi-modal false information detection. The method specifically comprises the following steps: firstly, acquiring multi-modal data of news in a social network, and preprocessing a text and an image; then respectively constructing global and local feature extractors of the text and the image; local semantic consistency detection is carried out through the extracted text local features and the image local features, global semantic consistency detection is carried out on the obtained global features, the two features are combined to generate a joint semantic consistency result, and the attention mechanism text and image features are adopted to carry out multi-modal fusion to obtain a joint semantic consistency result; the false information is distinguished from multiple dimensions and a loss function is established to optimize the model by comprehensively analyzing the multi-modal features after the text features and the image features are fused and the united semantic consistency result, so that the detection of the false information is realized. According to the method, the accuracy of multi-modal false information detection can be effectively improved, and the multi-modal false information can be automatically detected.

The invention discloses a construction method, system and equipment of a power dispatching knowledge question-answering system and a storage medium. The construction method comprises the steps of obtaining a power dispatching semantic document needing to be processed by the power dispatching knowledge question-answering system; sequentially carrying out character-based segmentation, proposition conversion and proposition combination on the power dispatching semantic document on the basis of the large language model and the related cue words to obtain a plurality of text blocks; based on the large language model and the related prompt words, identifying and extracting entities related to power dispatching and relationships among the entities from each text block; constructing a local graph for each text block on the basis of the large language model and the related prompt words, the entities extracted from each text block and the relationship among the entities; traversing each text block, and combining all the local graphs into a global graph; and on the basis of the large language model and the related prompt words, according to the question type queried by the user, a reply text is correspondingly generated by matching a minimum local graph which is most related to the global graph in the global graph.

The invention relates to a hyperspectral laser radar collaborative classification method based on spatial alignment and causal interaction, and compared with the prior art, the method solves the problem that in the hyperspectral laser radar collaborative classification process of an existing method, due to the fact that spatial dislocation and modal difference caused by a shooting process, an imaging mechanism and the like of a multi-modal image are not considered, the classification efficiency of the hyperspectral laser radar is greatly improved. And therefore, the problem of insufficient classification precision and robustness can be solved. The method comprises the following steps: obtaining a multi-modal remote sensing image data set, constructing a spatial alignment and causal interaction hyperspectral laser radar collaborative classification model, training the spatial alignment and causal interaction hyperspectral laser radar collaborative classification model, and obtaining an image classification result. According to the method, accurate alignment of the multi-modal image is realized by predicting the space offset, and the causal and non-causal factors of the modal are distinguished by applying a causal interaction mechanism, so that modal characteristic interference is effectively eliminated, and the accuracy and robustness of a hyperspectral laser radar collaborative classification result are ensured.

The invention relates to an image-text matching analysis method, system and device based on a large model and multiple modes. The method comprises the following steps: 1) image-text analysis: analyzing an image into an image and a text; 2) carrying out fine-tuning multi-mode large model processing on the image, and carrying out word segmentation processing on the characters; 3) constructing a bisection weight; 4) performing Few-shot cue word construction on the processed characters and the constructed equally-divided weight; 5) performing training / fine tuning on the characters and the pictures in the step 4 and the step 2 to serve as a large oracle model; according to the multi-mode image-text matching method, cooperative work of the multi-mode large model and the large language model is introduced, the advantages of the multi-mode large model and the large language model in semantic comprehension, context analysis and dynamic adjustment capacity are comprehensively utilized, a comprehensive image-text matching framework is established, and the accuracy, flexibility and interpretability of matching can be improved.

The invention relates to the technical field of scenic spot explanation systems, in particular to an intelligent tour guide robot and AR scenic spot explanation system, and the system comprises a cloud server side which comprises a scenic spot 3D model database, an augmented reality resource package database, a deep learning semantic understanding service module and a multi-language text translation service module; the intelligent tour guide terminal comprises an intelligent tour guide robot, and the robot is provided with a walking mechanism, a voice recognition module, a voice synthesis module, a natural language processing module and a human-computer interaction interface; the AR glasses terminal comprises AR glasses, and the glasses are provided with a camera, a display screen, a processor, a memory and a communication module; through the combination of deep learning semantic understanding and neural machine translation, the accuracy of cross-language communication is improved, and the ability of the system to understand the cultural background is enhanced. The AR technology and the knowledge graph are fused, so that unprecedented knowledge visualization experience is created, and abstract historical and cultural knowledge becomes vivid and perceptible.

The invention discloses a three-dimensional point cloud open scene understanding method based on double-flow feature fusion, and belongs to the field of multi-modal learning, and the method comprises the steps: preparing data, collecting three-dimensional point cloud data V, carrying out the data marking, obtaining a three-dimensional point cloud open scene understanding data set A, and carrying out the data set division of the obtained three-dimensional point cloud open scene understanding data set A; constructing three-dimensional point cloud open scene supervision information, constructing a foreground interest region, constructing open semantic information, and fusing double-flow feature information; designing a model M comprising a three-dimensional point cloud feature extractor E, a pre-trained text feature extractor theta text and a multi-modal discriminator D; training the model M to obtain a model M *; testing on the model M *; according to the three-dimensional point cloud open scene understanding method based on double-flow feature fusion, three-dimensional point cloud scene understanding on an open set is achieved, and the precision of tasks such as open set semantic segmentation is remarkably improved.

The invention discloses a deep learning-based question and answer knowledge graph construction method and system, and the method comprises the steps: collecting multi-dimensional question and answer data of a target domain from a plurality of data sources, carrying out the cleaning and standardization processing of the question and answer data, so as to eliminate noise information, and generating a preliminary data set; performing semantic analysis on the question and answer data by using a deep learning model, and generating a context embedding representation of each question and answer text in the question and answer data; knowledge entities and relationships among the entities are identified from the context embedding representation, and a preliminary knowledge triple is constructed; and based on the knowledge triad, constructing a preliminary question and answer knowledge graph for the identified entities and relationships, and allocating a unique identifier to each node in the preliminary question and answer knowledge graph to generate a structured question and answer knowledge graph. By utilizing the embodiment of the invention, the question and answer knowledge graph construction efficiency and the semantic understanding capability can be remarkably improved, and the application of an intelligent question and answer system in a complex scene is promoted.

The invention relates to the technical field of field detection, in particular to a zero-sample field detection method and system based on multi-task fine tuning and knowledge enhancement of argument data, and the method comprises the steps: firstly, searching related knowledge between a text and an object from an external knowledge base through a prompt project, and carrying out the knowledge enhancement of fine tuning data and test data; then analyzing and discussing that a plurality of tasks are included in the corpus data, and meanwhile, preprocessing data of each task; and finally, performing multi-task fine tuning on the open-source large model by using the discussion corpus data and external knowledge, and performing field reasoning by using the model after fine tuning. According to the method, multi-task fine tuning is carried out on the open-source large model by utilizing discussion corpus data with rigorous logic and relatively low noise, so that the performance of model site detection is improved; and meanwhile, the model takes related knowledge returned by an external knowledge base as a context, so that the accuracy of the model in text understanding and analysis aspects is enhanced.

The invention discloses a pedestrian tumble detection method based on NanoDet-Plus and fine tuning CLIP. The method comprises the following steps: S1, collecting data to make a data set; s2, a NanoDet-Plus target detection model is trained, and a trained algorithm model weight file is obtained; s3, performing pedestrian recognition on the image by using a target detection model NanoDet-Plus, and independently storing the recognized pedestrian image as an image file; s4, training the CLIP model and the improved Resnet50 model; s5, performing preliminary classification prediction on the pedestrian attitude image by using a CLIP model, performing further classification prediction by using an improved Resnet50 model, and connecting two classification prediction results so as to obtain a pedestrian attitude prediction result; and judging the current state of the pedestrian according to a posture prediction result, and taking feedback measures. And the recognition accuracy of the pedestrian falling event in the complex scene is improved.

The invention relates to a BERT and multi-module fusion-based track circuit fault label classification method and device and a medium, and the method comprises the steps: obtaining a to-be-classified track circuit fault text, and carrying out the cleaning and standardization processing of the to-be-classified track circuit fault text; performing word segmentation and vectorization processing on the track circuit fault text to generate a lexical symbol sequence conforming to BERT encoder input specifications; extracting a context semantic vector representation of the input lexical symbol sequence through a BERT encoder; performing feature extraction through a TextCNN module and a self-attention module which are parallel to obtain local features and weighted global features, and splicing the local features and the weighted global features to obtain a comprehensive feature vector; and synchronously generating a multi-label classification result of a fault phenomenon, a fault reason and a solution measure corresponding to the current fault text through a multi-label classification head based on the comprehensive feature vector. Compared with the prior art, the method has the advantages of high discrimination, high accuracy, high practicability and the like.

The invention provides a text sentiment classification method, electronic equipment and a storage medium. The method comprises the steps of obtaining a to-be-analyzed financial text; inputting a financial text to be analyzed into the FinBERT model to obtain an embedded representation form; inputting the to-be-analyzed financial text into the self-supervised learning model, learning the semantic role relationship of sentences, and inputting the to-be-analyzed financial text into the supervised learning model to obtain global features and local features; wherein the supervised learning model comprises a first module for capturing local features of a text, a second module for capturing global features of the text and an attention mechanism layer; and determining a text sentiment classification result corresponding to the financial text to be analyzed according to output results of the self-supervised learning model and the supervised learning model. According to the method and the device, a mode of combining the FinBERT model, the self-supervised learning model and the supervised learning model can be adopted, so that the limitation of a single type of model on financial text information identification and extraction is avoided, and the accuracy of an emotion classification result is improved.

The invention discloses an intelligent customer service cross-project experience migration method in combination with migration learning. The method comprises the following steps: S1, generating a source project feature set; s2, generating a source project optimization model containing a shared feature layer; s3, performing global optimization on shared feature layer parameters in the source project optimization model through Bayesian optimization; s4, generating an optimized target project feature set; s5, constructing a target project model; s6, performing fine tuning training on the target project model; s7, performing semantic consistency evaluation, response accuracy evaluation and business logic matching degree evaluation on the fine-tuned target project model on the target project data set; and S8, deploying the target project model passing the performance verification to a target project intelligent customer service system. According to the method, efficient deployment of the cross-project intelligent customer service system is realized, the migration efficiency and the service quality of the model are remarkably improved, and the method can be widely applied to intelligent customer service scenes of multiple industries.

This specification provides a text processing model training method, apparatus, device, medium, and product. The text processing model training method includes: acquiring first training data and training an initial text processing model based on the first training data to obtain an intermediate text processing model, wherein the first training data includes first sample query information and tag documents corresponding to the first sample query information; and acquiring second training data and training the intermediate text processing model based on the second training data to obtain a target text processing model, wherein the second training data includes second sample query information, sample documents corresponding to the second sample query information, tag relevance information, and tag thought chains, wherein the tag thought chains are derivational information used to characterize the tag relevance information between the second sample query information and the sample documents.

The present application belongs to the technical field of ground laser radar forestry data processing, and discloses a broad-leaved forest single tree segmentation method and system based on branch information guidance, taking ground-based laser radar broad-leaved forest point cloud as a processing object, using RANSAC cylindrical fitting to combine the growth characteristics of the tree trunk to detect the tree trunk; starting from the top of the tree trunk in the low vegetation area, the tree branches are extracted by segmenting and growing the branches and combining the thickness changes of the branch segments; starting from the end of the branch, the tree crown leaf point cloud is segmented by layer-by-layer growth. The present application has stronger trunk detection capability under the conditions of lush low vegetation and complex terrain; on the other hand, the present application can accurately segment the tree crown when the large and small crowns are intertwined and the multiple crowns are closely surrounded. The algorithm is efficient, simple and easy to use, and has great practical significance for improving the semantic understanding ability of forest scenes, assisting forest resource investigation, vegetation ecological research, and satellite remote sensing product calibration and verification.

The application discloses a modal correlation degree prediction method and device, equipment, a storage medium and a program product, and relates to the field of machine learning. The method comprises the following steps: acquiring a sample content set, extracting a second modal feature vector corresponding to second modal data of the sample content, determining a feature vector center corresponding to a plurality of sample classifications respectively, determining the distance between the second modal feature vector and the feature vector center as the correlation degree label between the second modal data and the first modal data in the sample content, training a candidate correlation degree identification model based on the correlation degree label to obtain a correlation degree identification model, using the model to identify the correlation degree between the first modal data and the second modal data in target content, and extracting the semantic feature representation of the target content based on the correlation degree. The correlation degree identification model can better learn the correlation between the modes in the multi-modal content, thereby assisting in enhancing the understanding ability of the article semantics.

The invention discloses a public security event identification method based on small sample learning and a knowledge graph, which comprises the following steps: generating a virtual scene similar to a real event scene through the combination of small sample data and the knowledge graph, and dynamically activating nodes related to the event through a knowledge activation threshold value, freezing and storing the unrelated nodes, and dynamically releasing frozen nodes; identifying dynamic evolution of the event by monitoring node relation fluctuation in the knowledge graph; a multidimensional event analysis result is generated through mutual reinforcement of a graph fission mechanism and an interactive training mechanism; generating a risk path identification result through the event aggregation trigger point; the scheme has the characteristics of realizing efficient learning of small samples, enhancing semantic understanding ability, dynamically adapting to diversified scenes, improving event identification flexibility and constructing a global risk assessment framework.

The application belongs to the technical field of target detection, and discloses a wind turbine blade defect detection method based on an improved RT-DETR. The method introduces a SWRepBlock module in the backbone network, thereby effectively solving the problem that a fixed receptive field of a traditional network cannot adaptively extract target features of different scales. In addition, the method replaces an AIFI module in an encoder with a DyT-AIFI module to improve the semantic understanding effect of long-distance feature interaction. In addition, the method introduces a CAA-HSFPN module in the encoder, thereby effectively solving the problem of a semantic gap of a traditional feature pyramid and insufficient distinction of feature importance. The wind turbine blade defect detection method can not only significantly improve the detection capability of small damages on a complex surface texture background of a wind turbine blade, but also effectively identify multiple types of defect features such as cracks, erosion and oil leakage.

The application belongs to the technical field of point cloud processing, and discloses a multi-modal semantic mapping method and system based on a master-slave architecture. The method comprises the following steps: S1, extracting feature points in point cloud data and calculating the covariance matrix of the neighborhood point cloud of each feature point; S2, solving the optimal pose corresponding to the minimum of the target function; S3, converting the obtained point cloud data into a world coordinate system using the optimal pose, fusing all frame point cloud data in the world coordinate system to form a three-dimensional point cloud map, projecting the three-dimensional point cloud map to a camera coordinate system, and inputting the three-dimensional point cloud map projected to the camera coordinate system into a semantic segmentation neural network to obtain the semantic label corresponding to each point cloud in the three-dimensional point cloud map, thereby achieving semantic mapping. Through the application, the problems of weak dynamic adaptability, insufficient model generalization ability and long-term running precision decay of the traditional three-dimensional semantic mapping method in a dynamic complex scene are solved.

The application provides a kind of digital resource retrieval system and method based on multi-modal and AI intelligent agent, it is related to electric digital data processing technical field, its method includes obtaining multi-modal query data and pre-processing, feature extraction and cross-modal semantic alignment, generates query vector set;Identify user category and accordingly the query vector set is semantically enhanced, and generates enhanced query vector;The enhanced query vector is matched with knowledge graph to determine the target node, and the extended query vector is generated along the semantic diffusion of knowledge graph;Based on the agent collaborative mechanism, the extended query vector is sequentially executed, and the parallel retrieval, fusion and sorting and confidence correction are carried out, and the modified sorting result is obtained;According to the scene adaptation factor, the modified sorting result is reordered, and the digital resource retrieval result is output in combination with the user category, so that user perception, semantic expansion and multi-agent collaborative retrieval for multi-modal education scene can be realized, and the accuracy and personalized adaptation degree of digital resource retrieval are improved.

The application discloses a kind of key place-oriented abnormal behavior intelligent identification and pre-emptive disposal system, preliminary abnormal event sequence is generated by collecting multi-modal environment data, it is fused with the semantics of preset scene knowledge graph to form composite abnormal event description information, then based on large language model inference generates dynamic disposal preplan, is decomposed into collaborative control instruction set by central dispatch intelligent agent, drives video analysis intelligent agent, broadcast dissuasion intelligent agent, anxi linkage intelligent agent executes collaborative disposal operation, and generates situation evolution information in shared event canvas through environment feedback data, realizes the dynamic optimization adjustment of disposal strategy.The system realizes the full-process intelligentization of abnormal event from identification to disposal, improves the semantic understanding ability of system to complex scene and the efficiency of multi-agent collaborative response, enhances the pertinence and adaptive adjustment ability of disposal preplan.

The invention discloses a dialogue intention information determination method and device, equipment and a storage medium. The method comprises the following steps: acquiring target speaking information of a target user in a communication dialogue; and performing complexity analysis processing according to the target speaking information, and determining a comprehensive complexity value corresponding to the target speaking information. And screening and determining a target reasoning module for intention analysis from a pre-constructed reasoning module set according to a pre-constructed complexity level threshold value and the comprehensive complexity value. And performing intention recognition processing on the target speaking information according to the target reasoning module, and determining dialogue intention information of the target user. According to the technical scheme, different resources are called according to the complexity of the user dialogue information, so that stable recognition of the intention category and intention intensity of the user is achieved, and the computing power and the cost are controlled while the recognition effect is guaranteed.

The present invention relates to a positive and negative sentiment analysis method, device, equipment and storage medium, and belongs to the technical field of public opinion analysis. The method, device, equipment and storage medium pre-construct an entity recognition word library and a domain subject word library, thereby automatically extracting subject words and entity words from the data to be analyzed, corresponding prior knowledge to subject words and entity words, converting the data to be analyzed into model input format data to be analyzed, and inputting it into a pre-constructed positive and negative sentiment analysis model to obtain the positive and negative sentiment probability values ​​of the data to be analyzed, wherein the automatic extraction of entity words and subject words in the field avoids the technical problem of manually annotating words one by one, liberates most of the manual annotation work, saves manpower and time, and introduces prior knowledge of subject words in professional fields, further improving the semantic understanding ability of the model.

The present invention relates to the field of network models, and specifically provides a model construction method for commodity name classification, which has the following steps: S1, open source Wikipedia and Books Corpus corpus collection, tax commodity name corpus collection; S2, data set operation on open source Wikipedia and BooksCorpus corpus; S3, introduction of BERT network and modular programming; S4, combination of various programming modules, construction of a complete End-to-End pipeline; S5, pre-training of the adjusted network using pre-processed open source data; S6, fine-tuning of the network to obtain the final tax field commodity name classification transfer learning model. Compared with the prior art, the model construction method for commodity name classification of the present invention can realize tax commodity name classification and has good promotion value.

A front-end and back-end combined hidden danger analysis system and method thereof, the system comprising: a front-end camera module, a video target recognition module, an image definition recognition module, an image similarity recognition module, a front-end slice compression module, a back-end image analysis module, a back-end hidden danger analysis result module and a front-end loudspeaker module, the method comprising: one: image acquisition; two: hidden danger target recognition; three: image definition recognition; four: image similarity recognition; five: video slice compression; six: image analysis; seven: hidden danger analysis; eight: hidden danger broadcast reminder; the video frame is preliminarily detected through the front-end video target recognition module, only the frame containing the hidden danger target is reserved for subsequent processing; combined with the image definition recognition module and the image similarity recognition module, further eliminate the fuzzy and repeated frames, generate high-quality video segments, effectively reduce the image data amount required to be processed by the back-end multi-modal large model, thereby greatly reduce the consumption of token.