601 results about "Regression analysis" patented technology

The invention discloses a territorial space planning index dynamic evaluation model, and particularly relates to the field of territorial planning, and the model comprises a multi-source heterogeneous data fusion module, an index system construction module, a dynamic adjustment model construction module, a scheme evaluation and optimization module, and a model application and dynamic updating module. According to the method, various spatial data are integrated through a multi-source heterogeneous data fusion technology, and the data quality is ensured by adopting a three-stage cleaning assembly line and an intelligent correction algorithm; the method comprises the following steps: constructing an index system based on six core contents of territorial space planning, and determining an index weight by applying a Delphi method and an analytic hierarchy process to realize adaptive adjustment of an index threshold; the model adopts time sequence analysis, regression analysis and a machine learning algorithm to predict a development trend, a dynamic feedback mechanism is established in combination with a system dynamics method, and an evaluation scheme is quantified through an analytic hierarchy process; the model performance is evaluated by adopting indexes such as accuracy rate and precision rate, and parameters are continuously optimized based on an evaluation result; finally, the model is integrated to a service system to realize real-time monitoring, and dynamic updating is realized through three mechanisms of data driving, performance triggering and regular maintenance.

The invention provides a safety monitoring method and system for a financial information service platform. Wherein a collaborative encryption transmission channel between a budgeting unit and an auditing department is constructed, a joint key is generated based on a multi-party computing protocol, and the ciphertext operation authority of a sensitive field is dynamically configured; performing ciphertext variance calculation and regression analysis on the encrypted budget execution record through the security operation node, generating a verifiable statistical result, binding the authority, and synchronously updating the sensitive field access frequency; a hierarchical security threat baseline is generated in combination with permission, frequency and historical query records, and cross-project high-frequency query and data operation deviation are associated; and when the query deviates from the baseline, calculating the risk confidence coefficient according to the deviation, triggering multi-party verification path adjustment and collaborative alarm, and dynamically limiting the ciphertext permission of the abnormal query. According to the technical scheme provided by the invention, cross-department data security sharing and dynamic risk collaborative prevention and control are realized.

The invention relates to a building material mining life cycle process carbon emission improvement potential quantification method comprising the following steps: constructing a full life cycle carbon footprint digital model covering mining, transportation, processing and restoration, and integrating a digital twinning and BIM technology dynamic correlation carbon emission factor library; key carbon emission influence factors are identified through Monte Carlo simulation and regression analysis; establishing a multi-objective optimization model, generating an optimal process parameter combination by adopting an intelligent algorithm, and quantifying the emission reduction potential; model parameters are dynamically calibrated in combination with real-time data, and a closed-loop optimization mechanism is formed; space-time distribution visualization and block chain evidence storage technologies are introduced, and accurate monitoring, intelligent decision making and credible tracing of carbon emission are achieved. According to the method, the problems of data lag and single optimization in a traditional method are solved, the accuracy of carbon emission and the feasibility of an emission reduction scheme are remarkably improved, meanwhile, carbon sink offset evaluation is supported, and a whole-process technical support is provided for mine green transformation.

The invention relates to the technical field of power system automation, and discloses an automatic control method and system for a multifunctional ring main unit. The method comprises the following steps: acquiring voltage, current and temperature parameters of a ring main unit in real time, analyzing load characteristics based on a clustering algorithm in combination with historical operation data, and generating a classification result including a load type, a seasonal trend and a load distribution thermodynamic diagram; a load change trend is extracted based on time sequence regression analysis, a fault prediction model is constructed by using a long short-term memory network, and a dynamic alarm threshold value is set; a grading alarm mechanism is triggered through comparison of real-time data and a model, a fault type and a positioning result are extracted in combination with a Bayesian network algorithm, the fault type is positioned, and finally automatic control is executed according to the fault type. Through data-driven intelligent analysis, the problems that a traditional ring main unit is complex in debugging and lagged in fault response are solved, the operation reliability, the maintenance efficiency and the power grid stability are remarkably improved, and predictive maintenance and self-adaptive control are achieved.

The invention provides a digital twin water conservancy hyper-fusion method for coupling multi-source data and an all-in-one machine, and relates to the technical field of intelligent water conservancy. By constructing a unified water conservancy spatio-temporal data link protocol, efficient fusion of multi-source heterogeneous data of hydrology, meteorology, engineering operation and the like is realized, and the problem of water conservancy system data islands is solved; a coupling distributed hydrological model and a hydrodynamic model are established, and real-time accurate simulation of the watershed hydrological process and the river channel hydrodynamic process is achieved; a multi-target dynamic weight reinforcement learning algorithm is adopted to optimize a water conservancy scheduling scheme, model parameters are dynamically corrected in real time through a time sequence error regression analysis model, and closed-loop adaptive optimization is achieved; according to the method, closed-loop cooperation of water conservancy data fusion, real-time accurate simulation, intelligent scheduling optimization and dynamic parameter self-correction is realized, and the real-time performance and accuracy of water conservancy intelligent decision making are remarkably improved.

The invention provides a lake water chemical oxygen demand internal and external source pollution quantitative tracing method and system, and relates to the technical field of pollution quantitative tracing. The method comprises the following steps: performing three-dimensional fluorescence spectrum analysis on water samples at different point positions of a lake to obtain fluorescence excitation / emission matrix spectrum EEMS data; performing parallel factor analysis on the EEMS data, identifying and classifying the fluorescent components of the dissolved organic matter DOM, and determining the type characteristics of each component; based on the maximum fluorescence intensity and the corresponding COD value of the fluorescence component, establishing a standardized multiple linear regression equation by adopting ridge regression analysis, and calculating the contribution percentage of each fluorescence component to the COD of the lake water body; analyzing influence paths of potential variables on DOM migration and transformation, and quantifying direct contributions and indirect contributions of different pollution sources to each fluorescent component through path coefficients; and calculating the total contribution of lake-entering rivers, bottom mud release and phytoplankton to the COD of the lake water body.

The invention discloses a BIM-based water conservancy project construction quality intelligent management method and system, and relates to the technical field of intelligent management. During operation of the system, various parameters related to the water conservancy project construction quality are acquired based on construction of a BIM model, data from a data acquisition module are preprocessed and integrated, and a BIM model is established; the method comprises the following steps: establishing a mathematical model for water conservancy project quality monitoring, establishing a dynamically updated and optimized model through regression analysis of historical data and a machine learning algorithm, predicting potential risks and quality problems in a construction process, and calculating a dynamic monitoring index Dtjk according to parameters output by a BIM model establishment module, the real-time monitoring and early-warning module is used for monitoring a BIM model, comparing the real-time monitoring and early-warning module with a preset qualified threshold value in the BIM model, giving out early warning according to the preset threshold value, reminding management personnel to take intervention measures, and providing improvement suggestions and optimizing a construction management scheme according to feedback information of the real-time monitoring and early-warning module and changes of a dynamic monitoring index Dtjk.

The invention provides a genetic disease gene detection data analysis method and system based on big data, and relates to the technical field of gene detection data analys.The genetic disease gene detection data analysis method comprises the steps that quality control and duplicate removal are conducted on gene sequencing original data, sequence comparison and variation detection are conducted, and a standard variation detection report is generated; performing multi-scale feature sampling and optimization by using an improved random field diffusion model, inputting high-dimensional feature distribution into a dual self-activation iterative network for feature extraction and integration to obtain a feature mapping matrix and a feature evolution trajectory, and inputting the feature mapping matrix and the feature evolution trajectory into an integrated predictor to obtain an integrated predictor; and in combination with Gaussian mixture process regression analysis and an improved Bayesian reasoning network, establishing a risk association network and training a deep hierarchical decision tree, and outputting a multi-dimensional risk assessment report.

The invention relates to the technical field of finance, in particular to a real-time investment decision-making system and method based on multi-modal fusion, and the method comprises the steps: integrating social media emotion data, public opinion event classification data and market structured data through a multi-modal emotion information fusion module, generating a comprehensive emotion representation vector, the market state judgment module judges market states based on historical market data and generates time-weighted market state representations, the reinforcement learning combination management module generates asset allocation decisions according to the market state representations and the time-weighted market state representations, and the multi-strategy collaborative decision module integrates reinforcement learning, decision trees and regression analysis strategies and generates collaborative decision signals. The risk self-adaptive control module predicts the market fluctuation rate, assesses the investment risk and adjusts decision signal execution parameters, the decision signal generation module generates a final investment decision signal and outputs the final investment decision signal to the transaction execution system, and the transaction execution system performs multi-modal information fusion and a time-sensitive attention mechanism. And the comprehensive understanding capability and the time sequence change adaptive capability of the market are improved.

A method and system for AI based insight extraction from format-bound financial transaction data includes transforming a structured dataset in ISO format into a transformed dataset having metadata interpretable by a LLM Metadata includes descriptions of field names, expected values, entity relationships, and business rules. The transformed dataset is analyzed using machine learning models such as regression analysis, principal component analysis, predictive modelling, or anomaly detection. An intent and content of a natural language query are determined using an NLP model. Based on the intent, context, and metadata, the LLM generates a database query, which is executed on the structured dataset to retrieve relevant data. Insights are generated by combining the retrieved data with machine learning results.

The invention relates to the technical field of 5G-R network detection and monitoring, discloses a 5G-R network situation awareness method based on distributed monitoring and multi-source information fusion, and aims at solving the problems that a 5G-R network is large in scale, high in dynamic performance and complex in data isomerism. Multi-source data of a core network, a wireless network, special equipment, interface monitoring, detection equipment, a GIS and the like are collected in real time; according to the method, technologies such as Pearson's correlation coefficients, FP-Growth, Bayesian analysis, a time sequence point process, a Hookes theory, a Gaussian mixture model, a decision tree, S-ARIMA, Boxplot, N-sigma, iForest, regression analysis, a neural network and KL divergence are combined to realize multi-source data fusion, intelligent analysis and visual perception, including network alarm, application quality, operation and maintenance and resource management. According to the method, the comprehensiveness, the real-time performance, the fault diagnosis accuracy and the operation and maintenance efficiency of the 5G-R network are improved, and the requirements of low delay and high reliability of a railway scene are met.

The invention provides a prediction method and prediction equipment for a building deformation point cloud time sequence, and relates to the technical field of deformation monitoring. The method comprises the following steps: analyzing an aligned point cloud set to obtain a point cloud set of an abnormal region; based on the multi-resolution features of the point cloud set of the abnormal region, extracting topological features of a preset connection structure of the abnormal region, and constructing a topological graph of the abnormal region; carrying out stress constraint and material characteristic constraint on the topological graph of the abnormal region, and determining submerged space characteristics of the topological graph; based on a latent diffusion model, performing interpolation and extrapolation processing on the latent space features, and determining a prediction vector field and a suspected damage area of the latent space features; performing multi-scale regression analysis on the suspected damage area to obtain local parameters; and carrying out fusion and consistency evaluation on the three-dimensional deformation prediction field corresponding to the prediction vector field of the submerged space features and the local parameters to obtain a fused global deformation prediction field, and obtaining high-risk deformation distribution. According to the method, deformation can be accurately predicted.

The invention relates to the technical field of network optimization, and discloses an AI-driven real-time network optimization algorithm, which comprises the following steps: data optimization: preprocessing network state data by using a generative adversarial network, extracting optimization features and generating enhanced data; topology modeling: modeling network topology by adopting a graph neural network, and updating node features; resource optimization: dynamically adjusting calculation, storage and bandwidth resources by using deep reinforcement learning; global scheduling: optimizing resource allocation based on a multi-objective optimization method; secondary optimization: combining a Lagrangian relaxation method and a graph optimization technology to further optimize a scheduling strategy; and adaptive learning: improving the adaptive capacity of the system in a dynamic environment through self-supervised learning and meta-learning. Through a matrix operation technology based on a regular equation and in cooperation with an efficient data preprocessing process, the effect of improving the solving speed of the regression model is achieved, the problem that calculation is slow on a big data set in a traditional method is solved, and rapid and accurate regression analysis is achieved.

The invention discloses an intelligent classroom management system based on AI, relates to the technical field of intelligent classrooms, and mainly solves the problems of inaccurate teaching resource matching, inaccurate student performance mastering and unobvious feedback effect. Comprising a student information management module, a teaching resource library, an intelligent teaching auxiliary module, an interactive communication platform, a student behavior analysis module and a family-school communication module, and student demands are analyzed through a collaborative filtering algorithm to improve accurate matching of teaching resources; student behaviors are analyzed and evaluated through a regression analysis prediction model, and comprehensive mastering of students is improved; teaching experience is continuously optimized through feedback circulation, and the feedback effect is improved.

The invention relates to the technical field of lake chlorophyll a inversion, and provides an inland lake chlorophyll a inversion method and system based on hyperspectral data, and the content of chlorophyll a in a lake range is inverted through a multiband index model and a regression method. Firstly, hyperspectral image data of a lake is obtained and preprocessed to remove noise and improve data quality. Then, extracting a water body range by adopting an NDWI index, and extracting wave band information of different sampling points; based on optical characteristics of lake water quality, a multiband index model considering pigment absorption and remote sensing reflectivity is constructed, and an inversion model is optimized in combination with regression analysis. According to the method, the hyperspectral remote sensing image data are collected and fused only through actual measurement of a small number of points, the constructed model is applied to the lake chlorophyll a distribution characteristics on the scale of a monitoring area, accurate inversion of the concentration of the chlorophyll a in the whole area of the lake can be achieved, and the problem that the accuracy is low when remote sensing data is singly adopted for inversion is solved.

The invention provides a power load prediction method based on a space-time diagram convolutional network in extreme weather. Comprising the following steps: collecting historical load data and regional meteorological element data of a plurality of load nodes in a power system, and screening key meteorological characteristics which have obvious influence on loads through a mode of combining model interpretation and regression analysis to construct a meteorological characteristic vector; multivariable empirical mode decomposition and singular value decomposition are adopted to carry out multi-scale reconstruction on load data, and smooth and effective load feature tensors are extracted. On the basis, a graph network structure is constructed in combination with a node physical connection relationship, and load and meteorological characteristics are fused in a time dimension to form node time sequence characteristics. And predicting the load by using the space-time diagram convolutional network model. According to the method, the space-time dependency relationship of the load data can be effectively modeled, the prediction accuracy of the load change under the extreme weather condition is enhanced, and the method has good robustness and generalization ability.

A method and system for improving the fuel efficiency of an aircraft flight, and a method for training a machine learning module to predict the most probable descent flight trajectories are disclosed. The machine learning module uses two stages of clustering and regression analysis to analyse historical flight data, so that the trained model can determine the most probable flight trajectory for the descent phase of a future / ongoing flight. This can then be used to determine an adjusted top of descent and output to the pilot of the future / ongoing flight, or used to control an associated autopilot system.

The invention provides an ecological system quality quantitative analysis method and system, and the method comprises the steps: carrying out the standardization processing of a multi-dimensional data factor, and obtaining the evaluation results of three dimensions, i.e., an ecological system background state, an ecological system landscape structure, and an ecological system service function, based on the processed data factor. And performing weight distribution and comprehensive calculation on evaluation results of the background state, the landscape structure and the service function of the ecological system to obtain an ecological system quality index. Based on this, regression analysis is performed on the ecological system quality index and a plurality of data factors, and then the core driving factor is accurately identified. According to the scheme, multiple dimensions such as the background, the function and the structure of the ecological system are coupled to perform ecological system quality evaluation, the defects of single dimension, evaluation framework limitation and the like in the prior art are avoided, and an effective basis can be provided for regional ecological environment management, ecological restoration project layout and protection decision by determining the core driving factor.

The invention provides a tunnel construction carbon emission prediction method based on an LCA and MLP neural network model, and the method comprises the following steps: (1) dividing a tunnel construction process into material production, transportation and construction stages, quantifying the carbon emission of each stage based on a carbon emission factor method, and carrying out the accumulation to obtain the total carbon emission; (2) extracting initial influence factors from construction design parameters and geological conditions, screening significant variables through univariate regression analysis, and carrying out refined analysis on the surrounding rock grade in combination with SHAP value contribution degree quantification; (3) constructing an MLP neural network model by taking the I-level key index as an input variable, and optimizing the performance of the model through hyper-parameter tuning; and (4) generating a multivariable nonlinear empirical formula based on the optimized MLP model weight matrix and an SHAP value analysis result, and converting a model prediction result into a rapid estimation tool suitable for tunnel construction carbon emission by the empirical formula through combining a linear term, a nonlinear term and a variable interaction term.

The invention discloses an intelligent dynamic prediction method for a coal and gas outburst dangerous area of a working face, and the method comprises the steps: building an integrated early warning information analysis platform, dynamically collecting key parameters in a mine, and constructing a dynamic data set; and calculating a coal seam firmness factor and a gas emission factor by utilizing a data fusion technology and combining historical data and regression analysis. Furthermore, in combination with a static data set, such as geological and ventilation system distribution data, a multi-parameter coupling prediction model is constructed. The model is trained and verified through a machine learning method so as to predict the coal and gas outburst risk of the working face in real time. The system automatically triggers early warning according to a model prediction result and a preset threshold value, a scientific basis is provided for coal mine safety management, and coal and gas outburst accidents are effectively prevented.

The invention discloses a method for automatically adjusting the riveting stroke of a rivet-free rivet, relates to the technical field of riveting, and is used for solving the problem of riveting precision deviation. The method comprises the following steps: firstly, determining a theoretical stop position through simulation analysis and experimental data, and establishing a process reference according to a riveting process and material characteristics; secondly, in the production process, the riveting stop position, the environment temperature, the power supply voltage and vibration data of the workpiece are collected in real time; environmental interference is eliminated through a multi-factor correction method, deviation is calculated and compared with theoretical parameters, and dynamic compensation is formed. In order to cope with long-term wear and mounting environment change, the invention also calculates wear correction and geological correction through regression analysis and neural network methods in combination with wear data, load history, geological environment and other information, and dynamically adjusts the riveting stroke according to the corrected deviation. Finally, through real-time feedback and closed-loop control, the riveting process is optimized, accurate riveting positioning is achieved, and the production efficiency and the product quality are improved.

The invention particularly relates to a meteorological equipment health degree assessment method based on regression and fuzzy comprehensive analysis. The method comprises the following steps: constructing a meteorological equipment health degree assessment index system; the method comprises the steps of collecting multi-source data of meteorological equipment, and completing data preprocessing through missing value filling and abnormal value correction; performing feature extraction on the multi-level indexes by adopting regression analysis, screening out core parameters strongly related to the health degree, and quantifying priorities; combining an analytic hierarchy process, integrating expert experience and feature importance results, constructing a hierarchical judgment matrix, and calculating the dynamic weight of each hierarchy index; and based on a fuzzy comprehensive evaluation method, establishing a multi-dimensional comment set and a membership function for the quantitative index and the qualitative index, fusing the weight matrix and the membership data, and outputting a quantitative score of the health state of the meteorological equipment. According to the method, the accuracy of health assessment of the meteorological equipment in severe environments such as extremely cold environments can be realized, and reliable technical support is provided for intelligent operation and maintenance of the meteorological equipment.

The invention belongs to the technical field of blasting vibration prediction, and particularly discloses a blasting vibration speed attenuation law prediction method based on regression analysis, and the method comprises the steps: laying monitoring points in multiple directions around a blasting source, synchronously collecting the geological parameters and mass point peak vibration speed of each point, and constructing a segmented propagation path; a single dominant propagation section is identified based on adjacent monitoring point data, independent influence of specific geological conditions on vibration attenuation is quantified by combining single-factor control regression analysis, decoupling modeling of geological factors and attenuation behaviors is realized, and meanwhile, a double-factor dominant propagation section is identified on the basis of quantizing the independent influence of the single geological conditions. And deducting a known effect through residual analysis to separate out an independent contribution of another geological parameter, and finally integrating a multi-factor influence function to construct a comprehensive attenuation model. Progressive modeling from single-factor decoupling to multi-factor collaboration is achieved, and the precision and physical interpretability of blasting vibration prediction in a complex heterogeneous stratum are remarkably improved.

The invention discloses an abnormal behavior detection method for a smart terminal remote management platform, which comprises the following steps: collecting multi-dimensional data streams of hardware state, software behavior, network communication and user operation in real time, dividing dynamic factors into three layers and nine classes, generating multiplicative / additive dynamic thresholds by utilizing regression analysis, and realizing abnormal association detection in combination with a multi-dimensional model. And a threshold value and a strategy are optimized in a closed-loop manner through historical data. According to the method, the limitation of a traditional fixed threshold is broken through, the problems of false detection and missing detection in dynamic scenes such as hardware aging and environment change are solved through dynamic factor modeling, the complex anomaly positioning capability is improved by utilizing multi-dimensional feature fusion, and intelligent operation and maintenance are realized by virtue of confidence classification and a strategy self-optimization mechanism. The method is suitable for intelligent terminal equipment in education, medical treatment and other scenes, and the anomaly detection accuracy and the system adaptability of a remote management platform are remarkably improved.

The invention provides a building energy-saving potential assessment method and system based on an energy consumption quota, and the method comprises the steps: carrying out the energy consumption analysis of a to-be-assessed building through a clustering analysis algorithm and a regression analysis algorithm according to the energy consumption multi-factor data of the to-be-assessed building, and obtaining the energy consumption quota data and the carbon emission quota data; based on the energy consumption multi-factor data and the energy consumption quota data, performing power consumption prediction on the to-be-evaluated building by using a multiple linear regression algorithm to obtain power consumption prediction data; calculating target energy efficiency data according to the electricity consumption prediction data; according to the energy consumption quota data, the carbon emission quota data and the target energy efficiency data, performing energy-saving potential assessment on the to-be-assessed building to obtain a comprehensive potential score of the to-be-assessed building; according to the method, the energy-saving potential of the building is evaluated by combining the energy consumption quota data, the carbon emission quota data and the power consumption prediction data obtained by using the multiple linear regression algorithm, the actual energy consumption of the building can be comprehensively reflected, and thus the potential energy-saving opportunity of the building can be identified more accurately.

The invention discloses a method for analyzing the uplift bearing capacity of a PHC spiral pipe pile based on numerical simulation. The method comprises the following steps: vertically screwing a pile body into a designed elevation by adopting intelligent rotary excavating equipment provided with a special drill bit, and monitoring the press-in speed, the rotating speed, the torque and the perpendicularity in real time; recording a press-in resistance-depth curve, a torque-depth curve and pile body perpendicularity data; establishing a three-dimensional refined numerical model containing a real spiral blade, and calibrating soil key parameters through a parameter inversion algorithm to obtain a calibrated numerical model; an anti-pulling load is simulated on the calibrated numerical model, and the ultimate anti-pulling bearing capacity and the failure mode are analyzed; based on a multi-working-condition simulation result, a practical calculation formula of the ultimate uplift bearing capacity is obtained through regression analysis; and model inversion is driven through construction data, refined numerical simulation is combined, the uplift bearing capacity prediction precision is improved, and a reliable basis is provided for the uplift design of the PHC spiral pipe pile.

The invention discloses a temperature control method for low-temperature high-NOx-concentration flue gas denitration in the nuclear industry, and relates to the technical field of flue gas denitration control. According to the method, flue gas temperature time sequence distribution, NOx concentration gradient and environment radiation intensity data are integrated through a space-time alignment algorithm, a synchronous data matrix representing a system coupling state is generated, then a temperature and efficiency response curved surface is established, a high-efficiency temperature area and an inactivation risk area are dynamically divided through fuzzy logic control, and an optimal temperature regulation and control threshold value is recognized. The drift effect of quantitative radiation on the surface temperature of the catalyst is analyzed through regression, a quantitative relation model of radiation dose and temperature zone deviation is established, and real-time correction of anti-radiation compensation parameters is achieved. And finally, optimizing the multi-objective cost function by adopting a non-dominated sorting genetic algorithm, and obtaining dynamic balance between the maximum NOx conversion rate and the minimum temperature fluctuation. And multi-parameter coupling control of a temperature field and a radiation field is realized, and the denitration efficiency stability under a low-temperature condition is remarkably improved.

The invention discloses a supply chain deployment system and method based on big data, and particularly relates to the technical field of supply chain deployment, and the system comprises a data collection module which is used for collecting the data of each link of a supply chain; a data storage and processing module; the demand prediction module is used for carrying out market demand prediction by using a time sequence analysis model and a regression analysis model; the intelligent allocation module is used for formulating an optimal supply chain allocation scheme by using an optimization algorithm; the logistics optimization module is used for carrying out optimization scheduling on logistics transportation routes and vehicles; the monitoring and feedback module is used for monitoring the operation state of each link of the supply chain in real time; according to the supply chain deployment system and method based on big data, the demand prediction accuracy is improved, through big data analysis and an advanced prediction algorithm, multiple influence factors are fully considered, the market demand can be predicted more accurately, the inventory overstock and stockout phenomena are reduced, and the inventory cost and opportunity cost of an enterprise are reduced.

The invention discloses a diagnosis method for predicting tuberculosis risk by using blood routine indexes, which comprises the following steps: (1) collecting open-source blood routine data, preprocessing, and dividing into a training set, a test set and a verification set; (2) screening blood routine examination and 25 indexes derived from the blood routine examination, and obtaining a preliminary screening result through LASSO regression analysis; (3) inputting the preliminary screening result into seven machine learning models, including a logistic regression model, a random forest model, a naive Bayes model, a K proximity model, a support vector machine model, an XGBoost model and a GBM model, for analysis to obtain a final variable combination and an optimal model; (4) inputting the verification set into the optimal model to obtain a DCA decision curve and a calibration curve; early discovery and precise diagnosis and treatment of tuberculosis are promoted, and meanwhile the pressure of medical resources is effectively relieved.

The invention provides a liver and gall disease prediction model construction method, system and device combined with multi-modal data and a medium, and belongs to the technical field of data prediction model construction. Medical data of a patient is collected; extracting the medical data in the medical data set, and evaluating the feature weight of each piece of medical data by using a regression analysis statistical algorithm; a comprehensive feature vector set is generated through fusion, and a preliminary liver and gall disease prediction model is constructed; and deploying the liver and gall disease prediction model to a medical terminal, and performing periodic updating through a cloud. A regression analysis statistical algorithm is utilized to evaluate a feature weight, features, such as key features such as glutamic-pyruvic transaminase and liver ultrasound image texture features, which have important influences on liver and gall disease prediction can be screened out, the data dimension is reduced, the model complexity is reduced, the key feature effect is highlighted, and the interpretability and prediction accuracy of the model are improved.