20 results about "Embedding algorithm" patented technology

The invention provides an intelligent risk identification and analysis method based on multi-modal heterogeneous data fusion, which relates to the technical field of risk management, and comprises the following steps of: acquiring multi-source data, constructing symbol-nerve double-space processing, generating semantic vectors, extracting features through orthogonal matrix decomposition and a bidirectional long-short-term memory network, and obtaining an intelligent risk identification and analysis result; according to the method, multi-modal information is fused by using an adaptive weight mechanism, and a risk propagation topology network is constructed by applying a graph embedding algorithm, so that accurate identification and early warning of risks are realized, and the risk prevention and control capability and prediction accuracy are effectively improved.

The invention provides a tunnel risk reasoning method fusing a knowledge graph and a large language model, which comprises the following steps of: obtaining structured monitoring data and unstructured text data, adopting methods such as field standardization for the structured monitoring data to realize a unified format, adopting methods such as sentence segmentation and word segmentation for the unstructured text data to realize the unified format, and obtaining the structured monitoring data and the unstructured text data; the method comprises the following steps of: extracting entities from data by utilizing a model, extracting a relationship between the entities based on the entities, forming basic triads, forming a sub-graph by the basic triads, integrating to form a knowledge graph, generating a natural language, extracting the sub-graph related to the natural language from the knowledge graph, and converting the sub-graph into a sub-graph in a vector form by utilizing a graph embedding algorithm. The entities and the relation paths of the entities serve as explicit reasoning clues, the natural language, the sub-maps in the vector form and the explicit reasoning clues are input into a large language model, natural language output is generated, multi-source data information is integrated, and high-precision and interpretable tunnel risk early warning is output through the large language model.

The invention belongs to the technical field of autonomous underwater vehicle numerical simulation, and discloses an AUV physical field prediction method based on manifold learning and deep learning, comprising the following steps: performing nonlinear dimension reduction on high-dimensional simulation physical field data based on an Isomap manifold learning algorithm, and extracting low-dimensional manifold features; constructing a deep neural network (DNN) model, and establishing a mapping relation between the working condition parameters and the low-dimensional manifold features; inputting new working condition parameters to the trained deep neural network model, and predicting corresponding low-dimensional manifold features; isomap inverse mapping is realized based on a local linear embedding algorithm, and predicted low-dimensional manifold features are reconstructed into high-dimensional physical field data. Through collaborative prediction of manifold learning and deep learning, the problem that a traditional method is low in calculation efficiency in high-dimensional physical field prediction is solved, prediction precision and real-time performance are remarkably improved, and an efficient tool is provided for AUV design optimization and dynamic control.

The invention relates to the technical field of knowledge management, in particular to a knowledge slice optimization method based on semantic analysis and a semantic relation network. On the one hand, semantic knowledge atoms of time sensitivity and domain labels are generated through a multi-dimensional evaluation and standardized processing mechanism of heterogeneous data sources, the integration precision and knowledge representation consistency of multi-source heterogeneous data are remarkably improved, a reliable data basis is provided for dynamic evolution of a semantic relationship network, and the reliability of the system is improved. The problems of data format conflict, semantic association weakening and updating lag in traditional knowledge management are solved; on the other hand, through an improved TransE graph embedding algorithm and a multi-layer constraint driven knowledge slice division mechanism, self-adaptive structure optimization and multi-granularity knowledge slice dynamic recombination of the semantic relation network are realized, the problem of low user retrieval efficiency is effectively avoided, and the semantic retrieval efficiency and scene adaptability of a large-scale knowledge base are enhanced.

Implementations are directed to receiving a plurality of data samples comprising a first set of data samples associated with respective labels and a second set of data samples to be labeled; generating a random forest structure comprising a set of decisions trees, each decision tree including nodes corresponding to the first set of data samples; adding the second set of data samples into each decision tree as additional nodes of each decision tree; merging the set of decision trees to obtain a universal graph, wherein each node corresponds to a data sample; extracting, using a graph embedding algorithm, an embedding feature for each data sample that corresponds to each node included in the universal graph; determining a distance between any pair of two data samples using respective embedding features of the two data samples; and determining a label for each of the second set of data samples using the distance.

The invention discloses a fault diagnosis method based on integrated empirical mode decomposition and a manifold structure, and aims to research an algorithm model capable of realizing effective fault diagnosis for an early fault with weak characteristics. The main core of the method is to integrate eigenmode function components obtained by empirical mode decomposition, judge the sensitivity of the eigenmode function components to early faults so as to provide a variable reconstruction strategy more sensitive to the early faults, and meanwhile, extract local features and manifold structures by using a neighborhood preserving embedding algorithm so as to improve the robustness of the early faults. And high-order statistical features more sensitive to early faults are constructed in combination with a statistical local algorithm, so that the high-order statistical features are input into a Bayesian classifier, and finally early fault diagnosis is realized. Compared with a traditional method, the method can more effectively distinguish different types of early faults, obtains higher accuracy, and is a more excellent early fault diagnosis method.

The application relates to a driving risk identification method based on an automatic timing hyperparameter optimization model, aiming to construct an automatic timing hyperparameter optimization model based on space-time trajectory data to identify driving risk behaviors, reduce training costs and optimize model precision on the basis of an existing model. The method constructs an automatic timing hyperparameter optimization model, builds an automatic machine learning framework based on vehicle space-time trajectory data, realizes automatic optimization of sliding window parameters and model hyperparameters, first synchronizes the phase coupling relationship between multiple feature data based on a dynamic time warping algorithm; second, automatically generates and reduces the dimension of features based on a deep feature synthesis-sliding window algorithm and a t-distributed stochastic neighbor embedding algorithm; finally, automatic model selection and hyperparameter adjustment are realized through Bayesian optimization, and model integration is carried out; the method considers the time sequence characteristics of data in the automatic machine learning framework, realizes the reduction of training costs and the optimization of the precision of the driving risk identification model.

The invention belongs to the technical field of multi-source track association. The invention provides a multi-source track association method based on adaptive fusion attention and contrast learning. According to the embodiment of the invention, in order to obtain track characterization with higher distinction degree, training is carried out by utilizing a self-supervised learning normal form of comparative learning, meanwhile, data enhancement is carried out on a single-source track to expand sample richness, and generalization of a network is improved. The structural features of the track under grid representation are extracted by using a graph embedding algorithm, the spatial features are enhanced, and the track is modeled more comprehensively through the two features. In order to obtain more comprehensive track characterization, complementary information between track structure features and spatial features is mined through a proposed adaptive fusion attention module. In order to solve the problem that the non-common observation target association error rate is high, ambiguity processing is introduced in the association matching stage, and the association precision is improved.

The invention provides a data dimension reduction method based on quantum mechanical characteristics. The method comprises the following steps: S1, performing nonlinear dimension reduction processing on high-dimensional data through quantum kernel principal component analysis; s2, performing linear dimensionality reduction on the output of the S1 based on a quantum neighborhood preserving embedding algorithm of mahalanobis distance; and S3, mapping the output of the S2 to a low-dimensional space by adopting a quantum variational manifold learning algorithm, and generating a final dimension reduction result. According to the method, efficient dimension reduction of high-dimensional data can be realized, the efficiency and precision of quantum machine learning preprocessing are improved, the calculation complexity is reduced, and the accuracy and reliability of a dimension reduction result are improved.

The application relates to a fault diagnosis method based on multi-head convolution and differential self-attention, first, a data sample set obtained is processed through a multi-head convolution embedding algorithm to expand the dimension, then a layer of learnable parameters is added on the top of the sample set, each element of the sample set is added with a learnable parameter, then the sample set is processed through an improved transformer encoder module to obtain the final output, only the top layer of the final output is taken to be classified through a classifier to obtain the final classification result, and then a fault diagnosis result is obtained based on the classification result; under the condition of lacking a large number of fault samples, the application can still realize high-accuracy fault diagnosis of bearings; compared with other benchmark models and existing methods, the application has higher diagnosis precision and stability, can provide effective fault diagnosis under a strong noise environment, has good robustness, and can be used for fault diagnosis under various working conditions and has strong generalization ability.

The invention discloses an intelligent MBSE modeling and man-machine collaboration method and system based on a large language model and feedback enhancement. The method comprises the steps of input understanding, knowledge representation, model generation, model application and man-machine collaboration feedback loop. In the input understanding step, semantic analysis and intention recognition are performed on natural language requirements by using a pre-trained large language model, and a structured requirement element set is generated; in the knowledge representation step, the demand is mapped to an MBSE knowledge graph through a graph embedding algorithm, and a knowledge-enhanced demand representation vector is generated; the model generation step is used for automatically generating an initial system model conforming to SysML / UAF specifications by using a sequence-to-sequence model based on the vector; in the model application step, the model is deployed for simulation verification and a result is output; and a feedback loop step of receiving user feedback, and updating model parameters by using a reinforcement learning algorithm to optimize subsequent processing. According to the method, the automation level and accuracy of MBSE modeling are improved, the man-machine cooperation efficiency is enhanced through closed-loop feedback, and the method is suitable for complex system development.

The application provides a student performance prediction method based on a woa (whale optimization algorithm), a T-SNE (t-distribution stochastic neighbor embedding) algorithm and an LSTM model, and comprises the following steps: A, obtaining a student data set and cleaning abnormal data; B, using the woa algorithm for data weighting; C, using the T-SNE algorithm for nonlinear dimension reduction processing of the data; and D, predicting the processed data through an optimized Adam-LSTM model to obtain student predicted performance. The application provides a student performance prediction method based on a deep learning model LSTM and a data set processing algorithm WOA and T-SNE. In the experiment, an Adam optimizer is fused with an LSTM model, the weight of the model can be more efficiently updated, and the stability and prediction efficiency of the LSTM model are improved. Meanwhile, in order to make the model pay more attention to important samples in the training process and improve the performance of the model, the WOA algorithm is applied to the weighted processing of the data set. In this paper, the T-SNE dimension reduction technology is also used to transform the original data by generating a new attribute set, which not only significantly reduces the number of attributes, but also ensures that most of the effective information in the original data is retained, further optimizes the LSTM model and obtains more accurate student ability prediction results.

The invention discloses a hyperbolic graph neural network recommendation method fusing comment scores. The method comprises the steps of obtaining a comment data set; converting the comment data into comment vectors by using a BERT-Whiting algorithm, and constructing a user-project comment perception graph and a user-project graph; introducing a hyperbolic embedding algorithm into the graph neural network to obtain a hyperbolic graph neural network, and inputting the graph into the hyperbolic graph neural network for learning to obtain an embedded representation of the graph; constructing a cross-view contrast loss function of the user and the project based on the embedded representation; mapping the graph, performing score prediction according to a mapping result, and constructing a loss function of score prediction; optimizing the hyperbolic graph neural network based on the loss function to obtain a prediction model, and outputting a prediction score of the user on the project; according to the method, the expression ability of the model for a complex graph structure can be enhanced, the generalization ability for user-project interaction data is improved, the discrimination of user and project expression is enhanced, and the accuracy and reliability of a recommendation result are improved.

The application provides a large language model optimization method and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: constructing a knowledge graph based on a knowledge base corresponding to an industry field, the knowledge graph comprising a plurality of nodes and connection relationships between the nodes, and the nodes representing knowledge points contained in the knowledge base; generating a prompt template corresponding to the knowledge graph based on a graph embedding algorithm; and optimizing a basic large language model based on the prompt template to obtain a professional large language model corresponding to the industry field. The application can improve the accuracy of the large language model in the application process.

The present application belongs to the technical field of mutual inductor measurement error identification, and particularly relates to a current transformer measurement error online identification method, device, equipment and medium, wherein the method comprises decomposing the three-phase current signal at the secondary side of the current transformer by VMD, and screening out the residual signal after decomposition; performing wavelet decomposition on the residual signal to obtain an approximate coefficient set; performing nonlinear dimension reduction processing on the approximate coefficient set by a t-distributed stochastic neighbor embedding algorithm to obtain low-dimensional feature data; estimating the probability density distribution of the low-dimensional feature data, and combining the preset confidence to determine the measurement error overrun threshold; comparing the low-dimensional feature data with the overrun threshold, if the low-dimensional feature data is less than the overrun threshold, it indicates that the current transformer is in a normal operating state at this time; otherwise, it is considered that the current transformer may be in an abnormal operating state at this time, solving the problem of inaccurate online identification of current transformer measurement error in the background technology.

This invention discloses a holomorphic embedded power system security verification method based on the compensation method. First, based on the holomorphic embedding theory, a holomorphic embedded power flow equation is constructed according to different node types. Second, a compensation method is introduced to correct the coefficient matrix of the holomorphic embedded power flow equation. Finally, combining the holomorphic embedded power flow equation and the compensation method, an efficient power system security verification algorithm is proposed, which avoids the inversion of high-dimensional matrices and greatly reduces the amount of calculation. The method of the present invention combines the advantages of the holomorphic embedding algorithm and the compensation method, and can take into account the accuracy, robustness, and efficiency of power system security verification.

The invention discloses a semantic reasoning method and device, equipment, a storage medium and a program product, and belongs to the technical field of knowledge maps. The method comprises the steps of constructing a knowledge graph based on a project text; based on the knowledge graph, vector representations of entities and relationships are obtained through an embedding algorithm; on the basis of the vector representation, a context vector and a historical behavior vector of the generated project are fused; based on the knowledge graph and the vector representation, identifying a cross-project collaborative path and calculating a path score; inputting a unified semantic representation fused with the context vector, the historical behavior vector and the path score into a multi-task learning model, and obtaining prediction scores of a plurality of evaluation dimensions in parallel; and extracting a semantic path associated with the prediction score from the knowledge graph, forming an interpretable reasoning chain, and outputting the interpretable reasoning chain and the prediction score together. According to the embodiment of the invention, the accuracy of semantic reasoning can be effectively improved.

The application discloses a traffic flow prediction method based on a Transformer adaptive adversarial graph neural network, and comprises the following steps: performing embedding operation on original data by using a degree embedding algorithm and a distance embedding algorithm to obtain traffic flow data; inputting the traffic flow data into a TDN module of a generator to perform attention mechanism operation to obtain aggregated time information Y; taking Y as input of an SDGCN graph neural network to obtain aggregated space information Z, and then inputting Z into an MLP with two hidden layers to obtain prediction data; connecting the prediction data with the traffic flow data X, denoted as X r ; inputting X p and the real data X p into a discriminator together; calculating loss functions L r of the discriminator and the generator; calculating L d and L g again; calculating a loss function L of the whole model; and finally updating parameters of the model through stochastic gradient descent. The application improves the accuracy of traffic flow prediction and the global consistency of prediction results, and has certain advantages in convergence speed.

The application discloses a shed factory microclimate prediction model construction method based on big data analysis, aiming to solve the defects of single data dimension, conventional algorithm and weak generalization ability of the existing model. The method first synchronously collects conventional environment, implicit association, production activities and historical time sequence multi-source heterogeneous data, and after preprocessing such as cleaning, normalization and time sequence alignment, extracts core features by using an improved sparse local linear embedding algorithm, constructs an improved autonomous echo state network prediction model fused with a snow ablation optimizer, and outputs prediction results combined with a sliding window dynamic optimization and error feedback correction mechanism. The application has comprehensive data coverage, strong algorithm innovation, high prediction accuracy and excellent generalization ability, can be adapted to various shed factory scenes, provides reliable support for accurate environmental regulation, and has outstanding practicality.

The application discloses a kind of based on the multi-element flexible resource scheduling method, system and terminal of holomorphic embedding algorithm, method includes, the kind of acquisition multi-element flexible resource and the parameter of each kind flexible resource belongs to power grid;Based on holomorphic embedding algorithm, the tide flow equation including multi-element flexible resource is constructed;Determine optimization target and constraint condition, linear programming model is constructed based on the tide flow equation;Solve linear programming model, obtain the coefficient of each variable series each order term, generate the operation instruction value of multi-element flexible resource.The application uses holomorphic embedding method to describe power flow constraint, so that the model has the constraint ability of considering tide flow, voltage, power;Relax high-order term linear, combined with optimization control target and flexible resource, power grid topology operation constraint, establish linear programming model, facilitate to use mature optimization solver to improve the convergence speed of algorithm, improve the feasibility of multi-element flexible resource control instruction.