14 results about "Correlation test" patented technology

The invention particularly relates to an enterprise credit scoring model modeling method and system. The enterprise credit scoring model modeling method comprises the steps of obtaining data, normalizing the data, filtering out abnormal values, performing variable correlation test, and performing standardization processing on variables; preliminarily analyzing and processing unbalanced sample data; screening important features, sorting based on XGBOOST feature importance, identifying category features, realizing category feature coding, expanding to generate new combined features, analyzing feature correlation, and reducing the number of the features to a specified dimension; constructing a base model, searching for optimal model parameters, constructing a hierarchical structure model, determining features and weights of all hierarchies, performing performance evaluation on the model, and continuously performing optimization and improvement. According to the enterprise credit scoring model modeling method and system, the evaluation accuracy of the model on the enterprise credit risk is improved, the over-fitting risk of the model is reduced, the interpretability and credibility of the model are improved, and continuous optimization and iteration of the model are realized.

The invention belongs to the technical field of landslide evaluation, and particularly relates to a modeling method of a landslide susceptibility evaluation coupling model fusing an auto-encoder and a feedforward neural network, and the method comprises the steps: obtaining slope unit data and historical landslide data, carrying out the preprocessing, extracting landslide-related evaluation factors, carrying out the correlation test and correlation analysis, and obtaining a landslide susceptibility evaluation coupling model. Screening out evaluation factors for subsequent evaluation; constructing a landslide data set, balancing samples by adopting a GPU acceleration oversampling technology, and dividing a training set and a test set; constructing a landslide susceptibility evaluation coupling model based on the auto-encoder and the feedforward neural network; training the landslide susceptibility evaluation coupling model by using the training set and the test set to obtain optimal parameters; and outputting a landslide susceptibility distribution result by using the trained landslide susceptibility evaluation coupling model. According to the method, feature dimension reduction and redundant information compression can be effectively realized, and the evaluation precision and generalization ability of the model are improved.

The invention relates to the technical field of environment monitoring, in particular to a method for evaluating data quality of automatic gas monitoring equipment. According to the technical scheme, the method comprises the steps that S1, at least two to-be-tested devices of the same model are deployed under the same sampling branch pipe of the same sampling main pipe and run at the same time; s2, carrying out the following single tests on each device: a single-point check test, a real-time to-be-tested gas standard addition recovery test and a device measurement value correlation test; s3, according to the comparison result of the average concentration of the gas to be detected and the equipment quantitation limit, calculating the comprehensive score of the single equipment by adopting different weighting modes; and S4, outputting a rating result for the equipment according to the comprehensive score.

The invention relates to the technical field of airtightness detection, in particular to a compressed air energy storage salt cavern cavity airtight sealing monitoring method and system.The method comprises the following steps that a deviation sequence is generated by collecting wellhead pressure and layering pressure, brine injection amount is compared to obtain a normalized ratio, and a spatial deviation record is generated; the sampling point rate is calculated, a time record is generated in combination with correlation test, pressure recovery time is extracted to verify abnormity and form a stability record, cross comparison is performed on a multi-dimensional result to identify an abnormal area, a risk result is output, and the sealing risk identification precision and reliability are improved. According to the method, the space distribution is revealed by collecting the multi-depth pressure of the wellhead and the cavity, the injection interference is eliminated through the normalized ratio, the rate and correlation test intensifies the time discrimination, the dynamic abnormity is verified through the recovery time comparison, the space time dynamic cross validation positioning is more accurate, and the detection is expanded from the static state to the three-dimensional comprehensive discrimination. And the sealing risk identification precision is obviously improved.

The application discloses an automatic data exploration method and system based on target label contribution degree, and the method comprises the following steps: associating and integrating feature data with a target label to obtain a data set; inputting the data set into a random forest and a cross-validation model, an information gain rate algorithm and a random forest algorithm respectively to determine a sorting table of the contribution degrees of respective feature fields in the feature data relative to the target label; processing the three sorting tables in a voting manner to obtain a feature contribution degree sorting table; obtaining a preset number of feature fields with high ranking in the feature contribution degree sorting table, performing correlation test on the feature fields two by two, and removing feature fields with correlation exceeding a preset threshold; and performing data distribution analysis on the feature fields remaining after the correlation test according to the basic data types of the fields. Through the technical scheme, automatic screening of the feature fields is realized, and the efficiency of feature screening is greatly improved, and tedious and repetitive manual work is avoided.

The present invention relates to a method, device, and medium for improving an event knowledge graph based on data-driven. The method includes: collecting and preprocessing numerical data; using the Markov boundary discovery algorithm to learn event causal relationships and constructing an event causal relationship set; screening events directly related to the target event through conditional independence tests; quantitatively evaluating the causal strength of event pairs and verifying the triples in the event knowledge graph, adding the triples with causal strength lower than the preset threshold to the set to be verified; for event pairs with insufficient causality but significant correlation, retaining the events that meet the significance conditions based on correlation tests; complementing the event pairs with significant causal relationships in the data but missing in the knowledge graph, generating and bidirectionally verifying the complemented event triples; integrating the verified and complemented triples to update the event knowledge graph, and this method effectively improves the accuracy and efficiency of causal inference.

The present invention relates to the field of environmental monitoring technology, and in particular to a method for evaluating the data quality of automatic gas monitoring equipment. The technical solution of the present invention includes: S1, deploying at least two devices of the same model to be tested under the same sampling branch of the same sampling main pipe and operating them simultaneously; S2, performing the following individual tests on each device: a single-point verification test, a real-time test gas spike recovery test, and a device measurement value correlation test; S3, calculating the comprehensive score of each device using different weighting methods based on the comparison results of the average concentration of the test gas and the quantitative limit of the device; S4, outputting a rating result for the device based on the comprehensive score.

The invention discloses a method for associating image plane fragment trajectories by combining forward and backward prediction with speed weighting, which belongs to the field of aerospace data association technology. The method comprises the following steps: S1 pre-processes the trajectories in the scene to obtain a set of trajectories without intersection in time and space; S2 performs forward prediction on the old trajectory set and backward prediction on the new trajectory set; S3 selects a time window for the old trajectory set and the new trajectory set using the speed weighting idea to construct a test quantity; S4 performs hypothesis testing on the obtained association test quantity, and constructs a test matrix for solving the multi-target situation. The benefits of the invention are as follows: first, there is no need to model the three-dimensional motion of the target, and the target can be associated through the motion of the image plane; second, the test time window is selected using the forward prediction, backward prediction and speed weighting ideas, and the association effect is better.

The invention relates to an adjustment cluster division method for hidden abandoned risks, and the method comprises the steps: extracting hidden abandoned historical sample points from a territorial survey data set; performing contribution rate and correlation test on the initial driving factors in the initial driving factor data set; performing risk probability prediction on the recessive abandoned land by using a preset maximum entropy machine learning model; carrying out main control factor nonlinear response analysis for resisting the hidden abandoned risk; according to the non-linear response characteristics, constructing a hidden abandoned land risk adjustment effect function considering the adjustment income and adjustment difficulty of the main control factor; and identifying a main control factor adjustment effect cluster meeting a preset medium-high recessive abandoned land risk condition, and dividing the main control factor adjustment effect cluster to generate a cluster management and control strategy of the recessive abandoned land. Therefore, the problems that the recessive abandoned land in a planting-free state is neglected and the prior knowledge of the historical recessive abandoned land is not utilized to carry out risk prediction and regulation management and control research in the prior art are solved.

This application relates to the field of image processing technology, and discloses a method, apparatus, and computer device for estimating the radial distortion coefficient of an image. The method performs discrete sampling within a preset range of radial distortion coefficients to generate an ordered sequence of distortion coefficients. For each coefficient in the sequence, a parameter traversal operation is performed: trial correction is performed based on the coefficient and the original image size; straight lines are detected and filtered in the corrected image; intersection points of straight lines are randomly sampled and outliers are cleaned; and statistics describing the dispersion of the intersection point distribution are calculated in the X and Y directions. After traversal, two corresponding statistical sequences are formed. A convexity trend test is performed on both sequences sequentially, and a correlation test is performed on the sequences that pass the test. The distortion coefficient corresponding to the global minimum point in the statistical sequences that pass all tests is determined as the optimal estimate. This application achieves automatic estimation without calibration or manual intervention, and has the advantages of strong robustness and wide adaptability.

The invention discloses a multivariate data fracturing process knowledge learning method based on NURBS and KAN architectures. The multivariate data fracturing process knowledge learning method comprises the following steps: 1) collecting parameter data of a rock mass fracturing process; 2) establishing a knowledge learning set function library; 3) carrying out data processing on the collected data to obtain k-step learning data; 4) carrying out data initialization processing on the collected data; the initialization processing comprises residual error amplification and normalization processing in deep learning; (5) the processed data physical parameters and the corresponding spatial position coordinate vectors are input into the fracturing process knowledge learning model; 6) performing correlation test; 7) learning rock mass fracturing knowledge; and 8) circulating the steps 3) to 7) until the physical parameter data of the rock mass fracturing process output by the fracturing process knowledge learning model is no longer increased or reaches a set cycle index, and completing knowledge learning. The KAN architecture is used, and the correlation test is combined, so that the generalization ability of the model can be effectively improved, and the problem of data imbalance can be effectively solved.

The application provides a kind of based on one-dimensional simplified water quality model, predicts water quality mutation, calculates real degradation coefficient and pollutant migration speed of the whole year using the monitoring data of water quality monitoring station, analyzes the relationship between degradation coefficient and river pollutant migration speed, builds regression model, and obtains the optimal degradation coefficient, migration speed by using correlation test calculation and mean square error calculation, and brings it into one-dimensional model formula, and calculates time sliding to form final prediction data, output prediction result graph.

This application relates to a method, apparatus, and equipment for analyzing post-meter water leakage based on smart water meter flow data. The method sequentially processes the original time-series-cumulative reading sequence using a data quality assessment algorithm, Boolean matrix and Fisher exact correlation test, robust Mahalanobis distance, and exponential smoothing filtering algorithms to generate a high-quality standard time-series-flow sequence. This ensures the data input to the model is "clean" and complete, laying the foundation for subsequent leakage analysis. The constructed dynamic threshold leakage judgment module automatically adjusts the threshold according to changes in user water usage habits, avoiding the limitations of fixed threshold judgment. The constructed micro-leakage identification model effectively identifies small, persistent leaks. Overall, this approach more accurately identifies post-meter water leakage, effectively avoiding misjudgments and omissions, and improving the reliability of leakage judgment.

The invention relates to an event knowledge graph improvement method and device based on data driving and a medium. The method comprises the steps of collecting and preprocessing numerical data; learning event causal relationships by using a Markov boundary discovery algorithm, and constructing an event causal relationship set; screening events directly related to the target event through a condition independence test; quantitatively evaluating the causal strength of the event pair, verifying the triad in the event knowledge graph, and adding the triad with the causal strength lower than a preset threshold value into a to-be-verified set; for event pairs which are insufficient in causality but significant in correlation, events which meet significance conditions are reserved based on correlation test; complementing event pairs with significant causal relationships in the data but missing in the knowledge graph, and generating and bidirectionally verifying a complemented event triple; and the event knowledge graph is updated by integrating the verified and complemented triad. The method effectively improves the accuracy and efficiency of causal inference.