456 results about "Gradient boosting" patented technology

The invention discloses a data security protection method and system combined with big data analysis, and relates to the field of data security protection, and the method comprises the steps: building a multi-dimensional data collection and fusion data set according to API access records, external threat intelligence and context metadata association data; based on an isolated forest algorithm, calculating a sample path length to evaluate a behavior anomaly probability score; risk indexes of external threats, data sensitivity and permission exceptions are quantified respectively; constructing a data security risk comprehensive assessment model based on a weighted summation algorithm, and updating and optimizing the weight in the model by using a gradient boosting tree algorithm; and according to an output result of the data security risk comprehensive assessment model, setting a security risk level, and according to the security risk level, dynamically responding to a protection measure. The method has the advantages that continuous and dynamic risk assessment and automatic response to third-party data access behaviors are realized by fusing multi-source data and an intelligent algorithm.

The invention provides an anesthesia virtual simulation training system fusing knowledge, skills and a thinking closed loop. The anesthesia virtual simulation training system comprises a medical knowledge base module, a clinical thinking module, a skill training module, an examination question brushing module, a knowledge graph module and an intelligent platform bottom layer framework. The intelligent platform underlying architecture comprises a data middle platform, an AI engine and a 3D engine, collects student behavior data of each module, constructs a dynamic student ability portrait through a gradient boosting tree algorithm and a collaborative filtering recommendation model, analyzes knowledge blind areas and skill shortages, plans a personalized learning path and pushes targeted training content, and provides a personalized learning result. A closed-loop process of evaluation, learning, practice and re-evaluation is formed; and deep fusion of theoretical knowledge, clinical thinking and skill operation is realized through a cross-module collaboration mechanism. The problems that traditional anesthesia teaching is high in practical operation risk, scattered in resource and insufficient in individuation are solved, and the clinical comprehensive ability and teaching quality of anesthetists are effectively improved.

The invention discloses a machine learning assisted polyethylene reaction performance prediction model training method, prediction method and device. The method comprises the following steps: acquiring a training set; screening feature items used for model training; obtaining a gradient boosting regression model for catalytic activity, a gradient boosting regression model for molecular weight and a gradient boosting regression model for molecular weight distribution; extracting feature items for model training from the data of the training set so as to obtain feature vectors; and respectively inputting the feature vectors into each model so as to train each model, thereby respectively obtaining hyper-parameters of the trained gradient-boosted regression model for catalytic activity, hyper-parameters of the trained gradient-boosted regression model for molecular weight and hyper-parameters of the trained gradient-boosted regression model for molecular weight distribution. According to the method, a model relationship between input characteristics and polymerization results (including catalytic activity, molecular weight, molecular weight distribution and the like) is established through training set learning.

The invention discloses a public building cold load short-time prediction method fusing physical information, and the method comprises the steps: collecting and preprocessing the historical cooling capacity, indoor environment, outdoor weather and equipment operation state data of a public building at a fixed time interval, and obtaining multi-dimensional input features; respectively establishing a workday sub-model and a holiday sub-model according to workday and holiday scene division; the workday sub-model and the holiday sub-model jointly form a cold load prediction model, the workday sub-model adopts a long short-term memory (LSTM) network, and the holiday sub-model adopts a light gradient elevator (Light GBM); a physical constraint loss function based on building energy balance and heat conduction residual error is introduced in the training process, and the physical constraint loss is fused into a total loss function according to a weighting coefficient so as to constrain that the output of each sub-model accords with the law of energy conservation and thermal inertia; monitoring the prediction error MAPE in real time and performing online calibration; and outputting a short-time cold load prediction result.

The invention discloses a water supply pipe network leakage locating method based on a lightweight gradient elevator algorithm, and belongs to the technical field of intelligent water affair and pipe network monitoring. The method comprises the following steps: integrating hydraulic modeling basic information, and constructing an initial hydraulic model by utilizing hydraulic software; actual measurement values of key node pressure and main pipe flow under different working conditions are collected and compared with simulation values to be calibrated to reach the standard, and a high-precision hydraulic model is obtained; simulating each leakage working condition in different time periods on a selected potential leakage node / pipe section by using the calibration model, collecting pressure change data of a pressure monitoring node, and performing preprocessing and data division; the data set is used for training a LightGBM model, the LightGBM model learns the relevance of node pressure in spatial topology and the dynamic change characteristics of a time sequence at the same time, a spatial position differentiation weight mechanism is introduced, and accurate positioning of leakage points and quantitative estimation of leakage coefficients are achieved. According to the method, the pipe network leakage positioning precision and efficiency are remarkably improved, and support is provided for rapid and accurate leakage detection and leakage amount evaluation.

The invention discloses a laser radar-assisted wind power generation wind shear modeling method, and relates to the technical field of wind shear modeling, and the method comprises the steps: collecting the wind speed measurement data of a laser radar at a plurality of heights in a target fan region, and constructing a measurement volume integral operator matrix; constructing a unit induction field correction model based on large eddy simulation in combination with a gradient boosting tree; coupling the measurement volume integral operator with the unit induction field correction model, and obtaining a high-precision wind speed vertical section and a corresponding wind shear parameter through inversion; on the basis of the high-precision wind speed vertical section and the corresponding wind shear parameters, physical constraints are fused, a wind shear model is generated, and uncertainty evaluation is carried out; according to the wind shear modeling method, the problems that the measurement error of the laser radar is difficult to correct, the wind shear modeling lacks the physical constraint and the result credibility cannot be quantified are solved by constructing the volume measurement integral operator and the unit induction field correction model for coupling inversion and fusing the physical constraint and the Bayesian uncertainty evaluation method.

The invention discloses a total primary productivity estimation method and system based on multi-model coupling deep learning, and the method comprises the steps: obtaining the multi-source data of meteorological data, remote sensing images, latent heat flux, sensible heat flux and solar radiation, carrying out the quality control, missing value processing and nearest neighbor interpolation of different data sources, and carrying out the prediction of the total primary productivity. Unifying to a target spatial resolution and a time resolution; in a light energy utilization rate (LUE) model family, a solar radiation phase factor is introduced into a photosynthetically active radiation absorption ratio (FPAR) to obtain a phase modulation type FPAR, drought duration is introduced into a water stress function f (W) to obtain an exponential decay type f (W), and a GPP time sequence of a plurality of improved mechanism models is calculated according to the exponential decay type f (W); extracting spatial texture features from a remote sensing image stack by using a convolutional neural network (CNN), and performing cross-modal fusion on the spatial features and the GPP estimated by the plurality of improved mechanism models in a gating mode to form fusion representation; carrying out learning and collaborative optimization on the fusion representation of the GPP and CNN spatial features estimated by the plurality of improved mechanism models by adopting a gradient lifting tree model; and high-precision estimation of the GPP is realized through multi-model collaborative optimization. The method aims at solving the problem that a traditional light energy utilization rate model is insufficient in response under the extreme environment conditions of drought and intense radiation, the adaptability limitation of a traditional single model under the complex environment is broken through, and therefore high-precision GPP estimation under the complex environment is achieved.

Particular embodiments relate to personalized document field prediction based on user behavior and feature generation. Specifically, various embodiments have the technical effect of improved accuracy with respect to field / entity value prediction (e.g., predicting that the amount due is X via a Gradient Boosting Model) relative to document processing technologies by learning through user behavior data or feedback (e.g., through continuous reinforcement learning from human feedback (RLHF)). This is at least partially because of the technical solution of accessing or generating unique features from one or more documents previously used by a user.

The invention provides an intelligent scheduling method and device for a multi-hole gate and a server, and relates to the technical field of hydraulic engineering automation and artificial intelligence optimization scheduling, and the method comprises the steps: determining target input characteristics through the historical operation data and simulation data of the multi-hole gate, and improving the target input characteristics through a lightweight gradient lifting tree model; performing nonlinear regression training processing on the target input features, and determining a target traffic prediction model at the current moment; performing hierarchical expansion processing on the gate opening degree combination through a multi-constraint search model to obtain a gate expansion scheduling scheme, and performing batch prediction processing and scoring processing on the gate expansion scheduling scheme by using a target flow prediction model to obtain a gate candidate scheduling scheme set; and performing upper limit detection screening processing and error tolerance screening processing on the gate candidate scheduling scheme set, and determining a target scheduling scheme of the porous gate. The method can significantly improve the prediction precision of the outlet water of the gate and the scheduling efficiency of the multi-hole gate.

The invention discloses a medical equipment state monitoring and early warning method and system based on a full life cycle, and relates to the technical field of medical equipment state monitoring and early warning. Comprising a full-life-cycle data acquisition module, a multi-dimensional data preprocessing module, a full-life-cycle state evaluation module, a hierarchical early warning decision module, a full-life-cycle management module and an edge-cloud collaborative storage module. The full-life-cycle data acquisition module is used for acquiring multi-dimensional data of the medical equipment in the whole stage from purchase acceptance, operation and use, operation and maintenance to scrap evaluation, and outputting original data in the whole stage. Differentiation quantitative indexes are set for different life cycle stages, an evaluation model is constructed by adopting an analytic hierarchy process and dynamic weight adjustment, and an early warning threshold value and a four-level hierarchical response mechanism are dynamically adjusted in combination with a gradient lifting tree algorithm. The problems that a traditional evaluation mode is poor in adaptability, early warning misinformation lags behind, response pertinence is insufficient, and clinical safety guarantee is weak are solved.

The invention provides a database fault root cause positioning method and device based on causal discovery, and belongs to the technical field of database operation and maintenance and fault diagnosis. Comprising the following steps: generating a two-dimensional data table and a statistical information set by using a multi-source operation log of a database system; constructing a causal discovery algorithm set A, and training the gradient boosting tree model by using the training set and the path combination to obtain an agent model Fb; establishing a Monte Carlo tree, selecting child nodes of root nodes according to performance expectation and exploration rewards, and expanding the child nodes to leaf nodes layer by layer; running the Monte Carlo tree, and obtaining a causal graph G * and a performance score by using a voting mode according to all causal relationships in the causal graph obtained by each node; calculating an exploration reward and a performance expectation of each node in a complete algorithm path, and returning the exploration reward and the performance expectation upwards to a root node from a leaf node along the path; updating Fb based on the searched path; and processing the new task by using a Monte Carlo tree, and identifying an affected processing variable according to a path pointing to an abnormal result variable in the causal graph.

The invention provides a rheumatoid arthritis patient low muscle quality risk prediction method based on an uncertainty perception stacked meta-learning structure, and belongs to the technical field of machine learning. The method comprises the following steps: acquiring a rheumatoid arthritis data set; constructing a base learner set comprising a table Transform class network and a gradient boosting tree class model; splicing the out-of-fold prediction probabilities of all the base learners and the statistics thereof to form a meta-feature matrix; taking the meta feature matrix as input, and constructing and training a meta learner comprising a spectrum normalization multilayer perceptron and a random feature Gaussian process output layer; collecting to-be-detected data, inputting the to-be-detected data into the trained base learner set and the trained meta learner in sequence, and performing temperature scaling and beta-calibration on a prediction result; and taking the calibrated prediction probability as a final prediction result. Through multi-source data fusion, the defects of an existing model in the aspects of practicability, probability reliability and the like are overcome.

The invention relates to the technical field of electric leakage protection, in particular to an electric leakage identification method and system for a chemical power utilization scene. According to the method, fault feature vectors are constructed based on the total number of waveform extreme points in a residual current sequence in a preset time window, the variance of position indexes of all pulses in the residual current sequence, time domain features and frequency domain features, and the feature vectors are classified by adopting an improved lightweight magnitude gradient elevator. The improved lightweight gradient elevator is optimized on the basis of a traditional lightweight gradient elevator, a learnable value is preset for a leaf node of each decision-making tree by adding a leaf attention mechanism, output is reconstructed after leaf-level attention is calculated, then the output is fused with a residual error output by an original tree, and then the output is calculated. And finally, constructing a classifier to finish fault classification. In addition, the distribution uniformity of the initial population is improved by designing chaotic mapping, the parameter precision of the lightweight gradient elevator is improved, and then the electric leakage identification precision is effectively improved.

The invention relates to the field of artificial intelligence, and discloses a hospital project configuration type selection scheme automatic generation method based on big data analysis. The method comprises the steps of multi-source data collection and standardization preprocessing, construction of a dynamic weight evaluation model fusing an analytic hierarchy process and a gradient boosting tree, multi-target Pareto optimization by adopting an improved genetic algorithm, introduction of a Monte Carlo simulation risk pre-judgment and linkage of a resource substitution library adjustment scheme, and output of a structured document and an interactive decision board. The system supports real-time data flow access, elastic space grid planning, post capability matrix matching, dynamic cost modeling and compliance automatic verification. Through a data driving and closed-loop feedback mechanism, automatic generation of a scheme, risk pre-control and continuous evolution of a model are realized, the resource allocation accuracy, the approval passing rate and the implementation efficiency are remarkably improved, and the hyperbranched rate and the modification cost are reduced.

The invention discloses a method for detecting and analyzing the content of liquid flavor substances in a beverage, and belongs to the technical field of beverage quality detection. The method comprises the following steps: obtaining a flavor substance dielectric response signal and a surface acoustic wave propagation signal of a beverage sample, and carrying out correlation mapping integration to form a structured original data set; preprocessing data by adopting ensemble empirical mode decomposition, integrating features through a Bayesian network multi-source data fusion algorithm, and constructing an optimized content calculation model based on a gradient boosting tree algorithm; performing time sequence consistency and dimension matching verification on the data set, and inputting the data set into a model to generate a prediction result; and extracting a related content value, carrying out deviation analysis on the related content value and a preset standard range, calculating a signal feature contribution weight, and outputting a detection report containing standard judgment and deviation traceability. According to the method, the physicochemical characteristics of the flavor substances can be comprehensively represented, accurate prediction and deviation traceability are realized, the comprehensiveness, reliability and efficiency of detection are improved, and powerful support is provided for beverage quality control.

The invention relates to the technical field of big data collection and preprocessing, and discloses a big data anomaly detection and cleaning method and system based on a multi-dimensional scoring model, and the method comprises the steps: carrying out the preprocessing of classified original data, and obtaining the processed data; an XGBoost algorithm is used, a gradient boosting tree framework is adopted to train the processed data, and parameter optimization is carried out on the gradient boosting tree framework; taking the adjusted gradient lifting tree framework as an anomaly detection model; judging whether to perform automatic data cleaning or not according to the exception score, and determining different cleaning methods according to the exception type; when automatic data cleaning cannot be carried out, an alarm mechanism is triggered; after anomaly detection, Shapley value analysis is carried out, contribution of each feature to an anomaly detection result is evaluated, and root cause analysis is carried out on anomaly data. According to the invention, the accuracy of anomaly detection and the high efficiency of cleaning are improved, and the interpretability and traceability of the result are enhanced.

The invention relates to the technical field of computer hardware intelligent configuration, and discloses a dynamic performance demand prediction method and system based on AI intelligent configuration. The method comprises the following steps: acquiring semantic demand description from user input, and converting the semantic demand description into a demand vector through a natural language processing model; querying a dynamic hardware database according to the demand vector, and extracting hardware meeting requirements to form a first candidate hardware list; verifying the adaptability of the hardware by adopting a hierarchical verification method, and removing conflicting hardware to obtain a second candidate hardware list; calculating the performance score and the comprehensive cost of each hardware combination, integrating weight adjustment mechanism evaluation, and screening out a preliminary recommendation scheme; a gradient lifting tree model is constructed according to the preliminary recommendation scheme, the hardware weight is adjusted through gradient descent, and a final hardware matching result is output through cross validation; and performing redundancy check on the final result, and generating a configuration report after confirming that no conflict exists. According to the method, user requirements can be accurately matched, the hardware compatibility and the performance adaptability are improved, and the configuration efficiency and accuracy are improved.

The invention relates to an intelligent decision-making method for a section drifting buoy driven by a digital twinborn model, and belongs to the technical field of decision-making optimization. A mechanical motion model of the section drifting buoy is constructed, and the digital twinborn model is obtained; establishing an agent model based on a gradient lifting tree model to represent a mapping relation between the operation speed, the operation depth and the oil discharge time of the profile drifting buoy and the energy consumption and the operation time of the plunger pump; constructing a cost function, determining a weight in the cost function based on a fuzzy rule, and setting a physical constraint; defining a decision variable, substituting the decision variable into a swarm intelligence optimization algorithm, calculating a Pareto frontier, and screening a preset proportion of Pareto frontier to obtain a candidate solution set; according to the candidate solution set, a digital twinborn model and a cost function are utilized to calculate the total cost value, a decision variable corresponding to the minimum total cost value is obtained, the energy consumption and time efficiency of the buoy floating process are improved, the decision precision is guaranteed, the calculation complexity is reduced, and the autonomous intelligent decision requirement of the buoy in the complex marine environment is met.

The invention relates to a safety monitoring method for an irrigation and drainage project operation state and an intelligent system, and belongs to the technical field of artificial intelligence. The method comprises the following steps: collecting and marking irrigation and drainage project operation monitoring data through a sensor, and constructing a training data set; completing data normalization by combining quantile and median robust scaling with adaptive nonlinear transformation, and mining and screening high-order interaction features by combining a mutual information theory and a gradient boosting decision tree; constructing a deep classification network fusing physical prior and adaptive feature interaction, introducing physical constraint and multi-scale feature fusion, and optimizing a model through adaptive marginal classification loss and physical feature manifold alignment loss; real-time data is preprocessed and then input into the model, and operation state grade classification and graded alarm are achieved. According to the method, data noise can be inhibited, a multi-index coupling relationship can be mined, the interpretability and robustness of the model can be improved by integrating a physical rule, irrigation and drainage project abnormity can be accurately identified and early warned, and the method is suitable for intelligent safety monitoring of an irrigation area.

The invention discloses an AI-driven polymer composite material process optimization method, and aims to solve the problems of data islands, process optimization lag and insufficient model timeliness in a polymer material production process. The method comprises the following steps: collecting full-link data according to a six-level customer product coding specification; the state parameters of the high-frequency equipment are safely stored in the sub-table 1 through encryption and identity authentication; constructing a structured database based on the production batch number association main table, the raw material sub-table set, the sub-table 2 and the sub-table 3; training a multi-model artificial intelligence system fusing gradient boosting regression, Bayesian optimization, a neural network and a random forest, and realizing a bidirectional linkage closed loop of formula recommendation, process optimization and performance prediction; and incremental learning is carried out by adopting a sliding window mechanism in combination with an online gradient descent and elastic weight consolidation strategy. According to the technical scheme, intelligent, efficient and safe optimization of the high polymer material process can be achieved, and the product quality and the production efficiency are remarkably improved.

The invention relates to the technical field of financial credit data evaluation, and discloses a user credit evaluation method based on multi-source credit data. The method comprises the steps of collecting multi-source heterogeneous credit investigation data in real time through a preset API interface group, and performing fusion after cleaning and standardization to form a user credit investigation data cube; extracting static, behavior sequence, association network and time sequence evolution four-dimensional features based on the cube; processing corresponding features by using a gradient boosting tree, a graph neural network and a time sequence convolutional network sub-model by using a multi-modal integrated learning framework, and generating a basic credit score through adaptive weighted fusion; the score is dynamically calibrated in combination with the real-time data flow and the macroeconomic factor, and high-risk user judgment and early warning are carried out based on dynamically adjusted behavior permission parameters by utilizing a real-time risk early warning engine; and finally, model retraining is triggered based on historical and recent default rate differences. According to the method, comprehensive, accurate and dynamic credit assessment is realized, and the assessment accuracy and timeliness are remarkably improved.

An example operation may include at least one of loading an Artificial Intelligence (AI) model from the storage, wherein the AI model is one of a diffusion-based model or a flow-based model, receiving tabular input data for execution by the AI model, wherein the tabular input data is scaled with a class-conditional scaler, creating a multi-output Gradient Boosted Tree (GBT), creating a Scalable AI (SAI) model from the AI model by using the multi-output GBT as a function approximator, and generating synthetic data by at least one of: executing the SAI model on the tabular input data or implementing a trained SAI model with the tabular input data, wherein the generating synthetic data reduces processing and memory resources.

The invention relates to a user behavior fused flow casting dynamic pricing method and device, equipment and a medium. The method comprises the steps of obtaining an original behavior log of a user, performing feature extraction processing on the original behavior log, generating a structured behavior feature vector, performing quantification processing on the structured behavior feature vector, and generating a user-level behavior willingness index; historical transaction data is obtained, the user-level behavior willingness index and the historical transaction data are input into a price elasticity prediction model for processing, a user-level real-time price elasticity coefficient is output, and the price elasticity prediction model is constructed based on a random forest algorithm and a gradient boosting algorithm; and performing dynamic bidding calculation processing based on the user-level real-time price elasticity coefficient and a basic bidding parameter preset by the advertiser to generate a final optimized bidding. According to the method, user behavior feature analysis, price elasticity prediction and dynamic bidding calculation are fused, individuation of flow casting dynamic pricing is achieved, the rate of return on investment of advertisement putting is effectively increased, and cost control is optimized.

The invention provides an FFU filter screen leakage early warning triggering system and method based on a particle counter, and relates to the technical field of filter screen leakage early warning, and the system comprises a data collection module which calculates the actual particle concentration change rate based on the collected data of an FFU; the reference curve construction module is used for establishing an associated reference curve of the rotating speed-particle concentration of the FFU motor under multiple particle sizes based on historical normal working condition data; the anomaly recognition module is used for calculating a deviation value between the actual particle concentration change rate and the associated reference curve, and marking leakage to be detected if a preset anomaly judgment standard appears; and the early warning module detects the abnormal score of the leakage to be detected through an isolated forest unsupervised algorithm, outputs the leakage probability in combination with a gradient boosting tree supervised algorithm, and triggers early warning based on the leakage probability. According to the method and the device, the technical problem of relatively low early warning accuracy of the leakage of the FFU filter screen in the prior art can be solved, and the technical effect of improving the early warning accuracy of the leakage of the FFU filter screen is achieved.

The invention discloses a satellite precipitation two-stage error correction method based on an actual measurement site, and the method comprises the steps: extracting drainage basin DEM data, obtaining a mask file of a drainage basin, and extracting the satellite precipitation of the drainage basin; performing inverse distance weighted interpolation on the rainfall data of the actual measurement site to a corresponding resolution to obtain grid data of the actual measurement site; error correction data are obtained through a deep learning model or dynamic quantile mapping; calculating a residual error between the rainfall data after error correction and grid rainfall of the actual measurement site, and learning a nonlinear relationship between the rainfall data and the residual error by using a gradient lifting regression tree model; and adding the precipitation residual error of the nonlinear residual error correction model to the error-corrected satellite precipitation error to obtain a residual error-corrected precipitation product. According to the method, the satellite precipitation product is corrected based on the actual measurement site data, the high-precision precipitation product is obtained, point-to-surface conversion of the precipitation data can be realized, and data support is provided for input of a refined hydrological model.

The invention discloses a flight landing time prediction method based on machine learning, and relates to the technical field of flight pre-judgment, and the method comprises the steps: carrying out the preprocessing of data based on historical flight data and historical meteorological data, and outputting feature sample data; performing data cleaning and standardization processing on the feature sample data, and outputting a training data set and a test data set in combination with actual landing and planned landing moment information of historical flights; establishing a landing prediction model, and performing optimization by combining reinforcement learning and a generative adversarial network; using the average absolute percentage error to test the landing prediction model and output a trained landing prediction model; and outputting a dynamically updated landing prediction model by adopting a concept drift detection algorithm, online gradient lifting and adaptive weighted fusion. Through combination of the deep Q network and multi-agent reinforcement learning, the prediction result is optimized in real time, and dynamic adjustment can be performed according to changes of actual flights.

The invention discloses a method for predicting liquid chromatography retention time of traditional Chinese medicine active ingredients. The method comprises the following steps: detecting retention time of flavonoid compounds or anthraquinone compounds under different chromatographic conditions through a reverse high performance liquid chromatograph; performing structure optimization on the flavonoid compound or the anthraquinone compound; performing molecular descriptor calculation on the optimized compound structure to obtain a molecular descriptor data set; performing genetic algorithm screening on the molecular descriptor data set to obtain characteristic molecular descriptors; combining the characteristic molecule descriptors with the different chromatographic conditions into a complete data set; the complete data set serves as an input variable, retention time serves as an output variable, and a gradient elevator GB algorithm or a random forest method RF is adopted for modeling to obtain a QSRR model; and predicting the retention time of flavonoid or anthraquinone compounds under different chromatographic conditions by using the constructed QSRR model.

The invention discloses an APP interface visual communication adaptive optimization method and system based on user satisfaction, and the method comprises the steps: constructing a visual element-satisfaction association prediction model, taking historical satisfaction score, operation feedback and scene feature data as a training basis, and carrying out the training through a gradient boosting tree algorithm; collecting real-time scene features and initial interaction data of a target user, and generating candidate visual element parameters and prediction scores; constructing an instant preference vector, and screening a to-be-verified visual scheme in combination with cosine similarity; a local gray level display scheme is adopted, and multi-modal feedback is collected; constructing an evaluation matrix, and calculating a real-time satisfaction comprehensive score by using an analytic hierarchy process; determining a scheme and incrementally updating the model if the standard is reached, and regenerating parameters if the standard is not reached; the system comprises six components such as a model iteration engine and a real-time data collector, and dynamic adaptation is achieved cooperatively. According to the method, user preferences and scenes can be accurately matched, the visual experience and the operation efficiency are improved, continuous optimization can be achieved through model iteration, and the method is suitable for various APPs needing personalized interfaces.

The invention relates to the technical field of oil-immersed transformer fault diagnosis, in particular to an oil-immersed transformer fault diagnosis method and system based on feature enhancement and deep isomerism. The method comprises the steps that a minority class oversampling method is used for synthesizing minority fault sample data into new samples conforming to physical constraints; an enhanced input space is constructed based on deep features generated by an auto-encoder VAE, and deep features of fault gas in a new sample are fully extracted; based on the extracted deep features, using a heterogeneous integrated model to simulate a lightweight gradient elevator Light GBM to capture a shallow relationship of the deep features; using a convolutional neural network 1D-CNN to identify local features of the fault gas; establishing a global dependency relationship on the basis of a Transform model; the robustness of the model is remarkably enhanced through a dynamic weighting mode, and therefore it is guaranteed that reliable and stable diagnosis results are continuously output in various complex and high-uncertainty practical application scenes.

The invention relates to the technical field of index structure recombination, in particular to a nursing process management database index optimization method, which comprises the following steps of: performing cost calculation on multiple indexes by adopting a gradient lifting regression tree, comparing the cost calculation with a threshold value, and forming a recombination candidate priority list under the support of a numerical result; a potential recombination object has a quantifiable judgment standard, predicted consumed time is calculated in combination with the low-ebb IO capacity and compared with the time slice length, the screening process can correspond to the system resource bearing condition, a B + tree algorithm is adopted for calculating a loading factor and the continuous scanning length, and through merging, splitting and redistribution operation, the screening efficiency is improved. A node organization structure is finely adjusted, it is guaranteed that an index structure and an access mode are kept coordinated, information entropy is utilized to sort and screen a field sequence, the field arrangement with high distinction degree is preferentially used for constructing a new index, atomic switching operation is executed to replace an old index, and the field sequence has adaptability in a dynamic data environment.