22 results about "Statistical relation" patented technology

A deep learning system for time series prediction comprising a preprocessor that receives time series input sequences, truncates them by removing terminal values, and appends padding values to maintain the original sequence length. An encoder compresses these padded sequences into latent space representations, while a decoder reconstructs predicted sequences matching the original length, specifically trained to reconstruct values matching the removed terminal values in positions corresponding to the padding values. A training system optimizes the encoder and decoder by minimizing differences between original sequences and predicted sequences. The system can process multiple time horizons simultaneously while maintaining statistical properties and providing uncertainty quantification through confidence intervals. This approach enables accurate short-term forecasting while preserving both temporal patterns and statistical relationships in the predicted sequences.

A method and a system for detecting a malicious infrastructure are provided. The method comprising: receiving a request comprising at least one infrastructure element of a given infrastructure assigned with a respective tag indicative of maliciousness of the given infrastructure; searching a database to identify therein, based at least on the respective tag, and at least one respective parameter of the at least one infrastructure element, and at least one additional infrastructure element and at least one additional respective parameter thereof; determining statistical relationships between the at least one respective parameter of at least one infrastructure element and the at least one additional respective parameter of the at least one additional infrastructure element; determining, based on the statistical relationships, rules for searching the database to identify therein new infrastructure elements; and assigning the respective tag associated with the given malicious infrastructure to the new infrastructure elements, thereby updating the database.

The invention discloses a composite foundation design method based on the influence of soil around a pile on the strength of a gravel pile and application, and the design method comprises the steps that each soil layer is subdivided, and the relation between the undrained shear strength value and the depth of each subdivided soil layer is recorded; performing statistics on the data by adopting a mean square error method according to the dynamic sounding hammering number data of the gravel pile, so as to obtain the dynamic sounding hammering number N of each subdivided soil layer; counting the statistical relationship between the undrained shear strength value of each soil layer and the dynamic sounding hammering number N of the gravel pile, and evaluating the compactness of the soil layer; crushing gravel with different compactness through a large direct shear test or a large triaxial test, and establishing a relationship between the different compactness of the gravel and an internal friction angle; establishing an empirical relationship between the strength of the soil around the pile and the value of the internal friction angle of the gravel; and composite foundation parameters are selected according to the subdivided soil layer parameters. The internal friction angle parameter of the gravel pile can be obtained in a layered mode according to the distribution of the soil layer, and the actual mechanical property can be better reflected.

The invention relates to the technical field of intelligent management, in particular to an intelligent data management method and system for connector production. The method comprises the steps of constructing a time sequence production transaction for each production batch and calculating a time weight, performing synchronous pruning to obtain a multi-constraint frequent item set when a candidate item set is generated, performing cross-hierarchy association on the multi-constraint frequent item set to generate a multi-granularity candidate association rule, and calculating a confidence coefficient; and for each candidate association rule, calculating a conditional probability that a posterior item is not established under the condition that a front item of the rule is a preceding item, taking a nonlinear mapping of the conditional probability as an anti-fact stability score, constructing a causal indicator factor, and performing weighted evaluation on the candidate association rule by integrating the confidence coefficient of the candidate association rule, the causal indicator factor and the anti-fact stability score to obtain a weighted evaluation result of the candidate association rule. And strong association rules are screened out. According to the scheme, the timeliness influence of each production event on the quality can be quantified, the real causal relation and the false statistical relation are distinguished, and the risk of decision making based on rules is reduced.

The present application belongs to the field of oil and gas field development, and particularly relates to a method for characterizing the characteristics of gravel around a horizontal well in a conglomerate reservoir. The method comprises the following steps: (1) using a conglomerate reservoir gravel characteristic identification and analysis software to process FMI imaging logging images and extract gravel parameters; (2) using a multiple linear regression analysis method to carry out sensitivity analysis of the gravel parameters and logging data, and to determine the statistical relationship between the gravel parameters and logging curves; and (3) establishing a model of the logging curves and the gravel parameters, and substituting the logging curve data to measure the gravel characteristics. The present application solves the problem of characterizing conglomerate reservoir gravel, and for the first time realizes the use of conventional logging curves to characterize the gravel parameters around a horizontal well section, thereby providing an optimization direction for multi-cluster volume fracturing technology in a horizontal well section.

A system for generating a normative reference model is described. The system constructs spatial basis sets from medical scans, each basis set characterizing spatial property across the body part. A normative basis model is then generated for each spatial basis set, and a normative cross-basis model is generated from statistical relationships between the spatial basis sets. Thereafter, a normative reference model is generated from the normative basis models and the normative cross-basis model.

An apparatus and method for predicting fungible asset requirement using statistical relationship modeling. The apparatus includes at least a processor and a memory communicatively connected to the at least a processor. The memory instructs the processor to process a plurality of multimodal data associated with a first fungible asset. The memory instructs the processor to generate, using a correlation module, a correlation matrix as a function of the plurality of multimodal data. The memory instructs the processor to generate a prediction module as a function of the correlation matrix. The memory instructs the processor to generate at least an acquisition outline for a second fungible asset using the prediction module. The memory instructs the processor to transmit the at least an acquisition outline to a downstream device.

The invention discloses a multi-stage coupled climate mode optimization and uncertainty quantification method and system, and relates to the technical field of climate prediction, and the method comprises the steps: obtaining a multi-source meteorological data set, and carrying out the preprocessing of the multi-source meteorological data set; on the basis of the preprocessed multi-source meteorological data set, a multi-dimensional evaluation index system is used for screening a GCM combination with the best adaptability to a research area; based on the screened GCM combination, establishing a segmented statistical relationship between the simulation data and the measured data in the historical period, obtaining a correction factor, and performing deviation correction on the original output of the mode by using the correction factor; and on the basis of a deviation correction result, fitting the GCM data in each scene by adopting a common least square method and a smooth fourth-order polynomial, respectively calculating the variance and proportion of each uncertainty component, and obtaining the total uncertainty through linear superposition. According to the invention, the problems of large atmospheric circulation mode selection deviation, low correction precision and insufficient uncertainty management in the prior art are solved.

The invention relates to the technical field of archive storage, in particular to an intelligent archive classified storage system which comprises a character arrangement analysis module, a rule priority judgment module, a classified number registration module, a storage position extension module and an archiving path generation module. According to the method, the statistical relationship between the number of times of rule triggering and the number of times of misjudgment is constructed, the priority ranking is carried out according to the rule suitability, the dynamic nature and classification accuracy of rule scheduling are enhanced, a number hierarchy weight calculation and structure offset comparison strategy is introduced in the archiving number judgment process, the number registration precision and the structure matching degree are improved, and the efficiency is improved. According to the method, when the capacity of the archiving block is close to overflow, dynamic combination and boundary updating of the adjacent empty areas are achieved, the ductility of a classified storage structure is kept, the serial number difference value and the track moving path are subjected to combined sorting comparison in archiving path generation, the archiving calling path and the archiving execution time are shortened, and the archiving efficiency is improved. And the filing efficiency and the number matching stability are improved.

The embodiment of the invention provides an intelligent prospecting method and device based on a data control theory, and the method comprises the steps: building a variable relation graph through collecting multi-modal original mineral exploration data, taking each exploration variable as a node, taking a statistical relation between the variables as an edge, and carrying out the calculation of the multi-modal original mineral exploration data through the guidance of a dissipation structure theory; carrying out centrality measurement on the variable relation graph through a graph neural network to obtain a key variable subset in the original mineral exploration data; performing feature dimension reduction on the key variable subset to construct a feature parameter space based on the principle of Hapkine, predicting a feature state in the feature parameter space through an ordinary differential equation, and outputting a corresponding feature parameter after the state is verified through a particle filter algorithm; based on guidance of the sharp point mutation theory, the characteristic parameters serve as control parameters to be input into a preset physical information neural network, state judgment is conducted through a state discriminant, a prospecting decision signal is determined, and the accuracy and efficiency of mineralization decision can be improved.

This invention relates to the field of archival storage technology, specifically an intelligent archival classification and storage system. The system includes a character layout parsing module, a rule priority determination module, a classification number registration module, a storage location extension module, and an archiving path generation module. In this invention, by constructing a statistical relationship between the number of rule triggers and the number of misjudgments, and prioritizing rules according to their adaptability, the dynamics of rule scheduling and classification accuracy are enhanced. In the archiving number determination process, a strategy of calculating number hierarchy weights and comparing structural offsets is introduced, improving the accuracy of number registration and structural matching. When the archiving block capacity is close to overflow, dynamic merging and boundary updates of adjacent empty areas are implemented, maintaining the extensibility of the classification storage structure. In the archiving path generation, the number difference and the track movement path are jointly sorted and compared, shortening the archival retrieval path and archiving execution time, and improving archiving efficiency and number matching stability.

Embodiments of the present disclosure provide an end-to-end system for extracting statistical relationships from scientific documents using a variety of machine learning (ML) and statistical models, including generative large language models (LLMs). The system aims to identify and extract two types of statistical relationships: a'universal 'or effect magnitude relationship, and a'pairing' or inter-group comparison relationship.

This application provides an intelligent mineral exploration method and apparatus based on data control theory. The method includes: collecting multimodal raw mineral exploration data, constructing a variable relationship graph with each exploration variable as a node and the statistical relationship between variables as an edge, and using dissipative structure theory as guidance to measure the centrality of the variable relationship graph through a graph neural network to obtain a subset of key variables in the raw mineral exploration data; based on the Haken enslavement principle, performing feature dimensionality reduction on the subset of key variables to construct a feature parameter space, and predicting the feature states in the feature parameter space through ordinary differential equations. After the state is verified by a particle filtering algorithm, the corresponding feature parameters are output; based on the cusp catastrophe theory, the feature parameters are input as control parameters into a preset physical information neural network, and the state is judged through a state discriminant to determine the mineral exploration decision signal. This application can improve the accuracy and efficiency of mineralization decision-making.

The application discloses a method, system, device and medium for predicting the development degree of volcanic rock cracks, and relates to the technical field of volcanic rock oil and gas reservoir exploration. The method comprises the following steps: determining a three-dimensional fusion seismic attribute body of a target area; determining a three-dimensional volcanic rock lithology data body according to the three-dimensional fusion seismic attribute body and a lithology division criterion; determining a crack development rate data body predicted by a thickness factor and a crack development rate data body predicted by a control area fault according to the three-dimensional volcanic rock lithology data body, a statistical relationship curve determined based on a sampling point and a drilling identification result and a well point fusion seismic attribute; and performing weighted calculation on the two crack development rate data bodies to obtain a final crack development rate data body. The application can effectively predict the crack development rate of volcanic rock cracks and provide support for the quantitative evaluation of the crack development degree.

The invention provides a seabed slope failure probability evaluation and analysis method based on a BP neural network algorithm, and the method comprises the steps: simulating the stability of a seabed slope in two scenes of rock-soil parameter uncertainty and wave load uncertainty through a random finite element numerical value, and obtaining the safety coefficient or stability label of the seabed slope. Geotechnical parameters and wave load uncertainty parameters are used as input, a safety coefficient or a stability label is used as output, a training set of a machine learning agent model is jointly formed, the two scenes are analyzed and trained through a machine learning BP neural network algorithm, and the two scenes are analyzed and trained. Complicated and time-consuming Monte Carlo random finite element numerical simulation is converted into a statistical relationship, a proxy model for rapidly calculating the seabed slope failure probability is established, the calculation cost of seabed slope reliability evaluation is remarkably reduced, a basis is provided for seabed slope engineering decision making, and the safety of seabed slope engineering is better guaranteed.

A system and a method for privacy preserved and fair synthetic data with proofs is disclosed The system a processor, and memory with instructions to: receive seed data with differing schemas; combine samples into a schema-consistent dataset by reconciling names, types and missing semantics; encode columns by characteristics to optimize representation; predict training requirements (batch size, learning rate, convergence) to optimize training; convert encoded data to training format; train generative models to produce synthetic records preserving statistical relations; sample models to produce records that maintain mathematical consistency with key statistics; generate proofs validating fairness and accuracy between synthetic and real records; assess privacy and fairness risks via attribute sensitivity and representation ratios and use results to guide parameters and sampling; and present via a user interface about job statuses, fairness indices, privacy scores, utility metrics and volumes to a user for governance.

This invention discloses a method for optimizing and reconstructing bridge scour formulas based on symbolic regression, comprising the following steps: obtaining statistical data on bridge scour, which are used as training and testing sets respectively; determining the initial structure of the bridge scour formula; firstly determining candidate operators, then introducing a group of functions with significant physical meaning and statistical relationships in scour to determine candidate operational variables; using a symbolic regression method based on genetic programming to generate an optimized formula using the data from the training set; adjusting parameters to generate a group of candidate formulas; evaluating the formula performance based on the testing set, and selecting the formula form with the best prediction effect and the simplest formula structure. This invention, based on existing standard formulas, uses a symbolic regression method based on genetic programming algorithms, integrating machine learning methods with nonlinear representation capabilities to more accurately mine rich features in the data. Simultaneously, by combining prior knowledge of scour and physical relationships, it can improve the predictive performance and generalization of the scour calculation formula.

The invention discloses a single dual-polarization radar data self-filling method based on physical constraint and dynamic optimization, and relates to the technical field of data filling. Comprising the following steps: acquiring original observation data, preprocessing the original observation data, and calculating a beam blocking rate; establishing a reliability criterion, and selecting an optimal filling path for each pixel point; when the relative differential phase is determined to be reliable, constructing a core path based on a-relationship; when the differential phase is not reliable but the differential reflectivity is reliable, an auxiliary path is constructed based on a statistical relationship, and data filling is executed hierarchically; and solving a reflectivity field which is finally filled and is physically coherent by minimizing a cost function. Smooth transition of the filling area and the surrounding effective observation area is ensured, and a high-quality data field is formed.

The present invention discloses a multi-physics field collaborative physical property inversion method for a mineral cluster area based on interactive guidance, which belongs to the field of mineral exploration technology. The technical key points are as follows: Step 1: Subdividing the underground space into grids according to the horizontal position distribution of its observation points; Step 2: Calculating the forward kernel matrix of each geophysical field to obtain the mapping relationship between the observation point and the underground discrete grid unit under the unit physical property, and establishing the observation data through the forward kernel matrix. d To Model m relationship; Step three: obtain the three-dimensional discrete grids corresponding to the observation points of different physical fields, and realize the forward and inverse calculation of the three-dimensional discrete grids of each physical field by interactively guiding the calculation process. This has the advantages of solving the problems in the existing technology of strong dependence on rock physical property measurement and statistical relationship, and the inability of structural coupling mode to adapt to the sensitivity differences of different physical property parameters, and realizing high-precision and high-resolution multi-physical field physical property modeling of mining areas.

The invention relates to the technical field of weather forecast correction, in particular to a PM2.5 concentration-based logarithmic value mode temperature forecast correction method, which comprises the following steps of: S1, collecting hourly hourly temperature data and daily highest and lowest temperature data in the past three years, and simultaneously acquiring PM2.5 concentration data in a corresponding time period; s2, deviation analysis is carried out; and S3, selecting 155g / m as a PM2.5 concentration threshold value, and dividing the data sample into a high-concentration sample (Chigh) and a low-concentration sample (Craw). And S4, determining which regression model is used to carry out temperature forecast correction by predicting the PM2.5 concentration of the current day. According to the method, on the basis of the statistical relation between the PM2.5 concentration and the temperature forecast deviation, the forecast deviation under different pollution concentration conditions is coped with through a double regression model strategy, effective correction of traditional numerical mode forecast is achieved, and the accuracy and reliability of temperature forecast are improved.

The invention discloses a method for reducing climate mode estimation uncertainty based on variable coordination relationship constraint, which comprises the following steps of: identifying a physical coordination relationship among key variables, and constructing a regression model of multi-mode historical constraint variable data and future constrained variable data; and forming a statistical relationship, namely an emergence constraint equation, which can be checked by the reference data, and further accurately calculating the uncertainty of the corrected global climate mode estimation data according to the multi-model data layer, the reference data layer and the constraint relationship layer in combination with the reference data. According to the method, a physical constraint mechanism is established by mining the cooperative relationship among climate variables, and the uncertainty is calculated by adopting layering, so that the reliability of mode estimation is remarkably improved, and the uncertainty is reduced.

The application discloses a water quality monitoring index prediction method based on multiple regression and random forest. The application firstly analyzes the correlation degree between monitoring indexes based on various monitoring indexes of water quality monitoring by using Pearson correlation coefficient, so as to obtain the correlation coefficient between the monitoring indexes, further obtains the statistical relationship between the highly correlated parameter indexes by using a multiple linear regression algorithm, uses the form of the regression equation to express the causal relationship between the monitoring variables, and finally realizes the prediction of the dependent variable index by using the independent variable index in water quality monitoring by using a random forest algorithm, so as to reduce the monitoring items and the monitoring cost. The application can effectively improve the efficiency of water quality monitoring and reduce the cost of monitoring index items, and has the characteristics of high prediction accuracy and strong generalization ability.