443 results about "Support vector machines svms" patented technology

The invention relates to the technical field of industrial valve control, in particular to a floater control-based remote control and monitoring system for a quick-switching blind plate valve, which comprises a data acquisition unit, a cross-dimension data analysis unit and an instruction generation and feedback unit, and is characterized in that the data acquisition unit constructs a three-dimensional motion matrix through a floater trajectory tracking algorithm; an effective pressure fluctuation period is extracted in combination with a pressure-temperature coupling denoising algorithm, a cross-dimension data analysis unit extracts time-frequency fusion features by using a residual convolutional neural network, and dual verification of a fluid phase state is realized through a support vector machine and a dynamic time warping algorithm; and the instruction generation and feedback unit triggers a standard control strategy or a self-learning compensation mechanism according to the confidence coefficient, and corrects control deviation through a negative feedback optimization coefficient matrix, so that the problems of insufficient multi-source data fusion and poor complex working condition adaptability in the traditional technology are solved, accurate remote control and intelligent monitoring of the quick-switching blind plate valve are realized, and the working efficiency of the quick-switching blind plate valve is improved. And the leakage early warning success rate and the valve action reliability are improved.

The invention discloses a multi-dimensional training method and device for a support vector machine, electronic equipment and a computer readable storage medium, and the method comprises the steps: carrying out the discretization of a training sample data set, and obtaining a discretized data set, the discretized data set comprises a plurality of different attributes, and each attribute corresponds to a plurality of feature vectors; calculating a classification contribution parameter of each attribute feature vector to obtain a plurality of classification contribution parameters; performing data mapping on the plurality of classification contribution degrees by using a kernel function to obtain a target function; and optimizing and training the objective function by using a gradient descent algorithm to obtain a support vector machine model. According to the method, different dimension data are mapped through the kernel function, the data gain weight can be determined, the linear separable effect of the mapped dimension data can be improved, and then the classification and analysis capability of the SVM can be improved.

The invention relates to a rolling bearing embedded lubrication state evaluation method and a computer device. The rolling bearing embedded lubrication state evaluation method comprises the steps that a temperature signal of a rolling bearing and a frequency domain spectrum amplitude sequence are spliced to generate a multi-source spectrum fusion feature vector; a random forest model is adopted to screen mean value features extracted from the temperature signals and time domain features and frequency domain features extracted from the vibration signals, the sound signals and the sound emission signals to obtain a sensitive feature set; inputting the sensitive feature set into a support vector machine model to output a first lubrication state evaluation result; inputting the multi-source spectrum fusion feature vector into a MobileNet V2 model to output a second lubrication state evaluation result; and fusing the first lubrication state evaluation result and the second lubrication state evaluation result through a D-S evidence theory to obtain a rolling bearing lubrication state evaluation result. The multi-source fusion evaluation method considering accuracy, robustness and engineering practicability is constructed, so that the problems of high limitation, insufficient fusion layers, high model complexity and the like of an existing single signal are solved.

The invention discloses a crane fault diagnosis method and system based on data driving, and the method comprises the steps: collecting the data of a PLC and a multi-source sensor of a crane, carrying out the preprocessing, and inputting a prediction model with the fusion of multi-scale causal convolution and an attention mechanism, so as to obtain a feature value prediction sequence; then calculating a residual error between a prediction sequence and an actual measurement sequence, modeling by using a first-class support vector machine, and triggering third-class early warning; based on the constructed Bayesian network, inputting the early warning evidence and updating the posterior probability, and outputting a Top-N fault reason; and finally, a risk score is calculated by integrating the posterior probability, the residual amplitude and the abnormal frequency, grading is carried out, and a diagnosis result and a disposal suggestion are pushed to a user terminal. According to the scheme, accurate diagnosis of complex coupling faults can be realized, 'beforehand 'early warning is realized, and unplanned shutdown and even safety accidents caused by fault expansion are effectively avoided.

The invention belongs to the technical field of machine learning models, particularly relates to a personnel performance evaluation method based on IWOA-SVM, and solves the problems that a traditional support vector machine (SVM) is low in precision, difficult in parameter selection and the like in performance intelligent evaluation. The method comprises the steps that Tent chaotic mapping and a pseudo-opposition learning strategy are utilized to increase the diversity and quality of an initial population, and the whale algorithm (WOA) is prevented from falling into local optimum; the global optimization capability of the WOA is improved by adopting a differential evolution mechanism; a penalty factor and kernel function parameters of the SVM are optimized through an improved whale algorithm (IWOA), and performance evaluation can be effectively carried out while optimal parameters are obtained. According to the method, the whale algorithm can be improved by using Tent chaotic mapping, pseudo-opposition learning and a differential evolution strategy, SVM parameters are searched in a global range, and better model performance is obtained.

The invention relates to the field of mine wireless communication security, and discloses a wireless instruction encryption method and system for mining intrinsic safety type equipment. The method comprises the following steps: collecting historical mine environment data and extracting features to obtain characterization environment description; constructing a nonlinear mapping function of the multipath propagation characteristic and the interference degree of the signal; collecting current environment data to calibrate the nonlinear mapping function to obtain a calibrated model; obtaining an encryption strength prediction value based on the current environment data in combination with the calibrated model; if the predicted value is lower than the encryption strength threshold value, key parameters are adjusted to generate optimized encryption configuration; based on the configuration, a support vector machine is adopted to classify the dynamic environment data, an interference area is divided, and an adjustment coefficient of the interference area is determined; and adjusting the encryption parameter in real time according to the coefficient to obtain an encryption strength adjustment scheme matched with the actual demand. According to the method, the problem that the wireless instruction encryption strength is not matched in a complex mine environment is solved, and the wireless instruction transmission safety and adaptability are improved.

The invention discloses a shield tail sealing leakage detection method and system based on multi-sensor fusion, and belongs to the field of intelligent detection.The method comprises the following steps that a real-time multi-physical-field data set is obtained according to seepage, stress and temperature multi-physical-field data collected by a sensor array deployed in a shield tail sealing area; carrying out fusion processing on the real-time multi-physical field data set by adopting a data fusion method, extracting correlation characteristics of a seepage effect, stress distribution and temperature change, and generating a coupling characteristic vector; performing classification processing on the coupling feature vector based on a mixed model of a graph convolutional network and a support vector machine to obtain an abnormal classification result; a geological environment dynamic change trend is extracted according to the anomaly classification result, and a quantitative index sequence is generated; performing mode grouping on the quantitative index sequence by adopting a clustering analysis method, and determining leakage risk level distribution; and triggering a real-time alarm mechanism based on the leakage risk level distribution to obtain a leakage detection confirmation signal.

The invention discloses an electronic film cutting control method and system based on digital twinning. The method comprises the steps that tool and material interaction data are collected and denoised, time domain and frequency domain characteristics of the tool and material interaction data are extracted, spectral distribution is calculated through Fourier transform, and material characteristic fluctuation mode data are determined; if the threshold value is exceeded, judging that the toughness of the material is changed, obtaining time sequence data, obtaining a material difference feature vector, inputting the material difference feature vector into a support vector machine model for classification, matching a material database, and determining an optimized control parameter set; a deep neural network is adopted to predict the optimized control parameter set, and environment variables are fused to obtain a final dynamic control instruction set; and if the final dynamic control instruction is not matched with the current cutter state, a stable machining path is obtained through closed-loop feedback cycle updating. According to the method, the digital twin of the cutter and the material is constructed, the physical machining process is mapped in real time, the machining precision and adaptability can be remarkably improved, and intelligent response to the material difference is achieved.

The invention discloses an elevator steel belt damage detection and quantitative analysis system based on eddy current and magnetic flux leakage dual-mode detection, and aims to solve the problems of single-mode information loss and serious industrial field strong noise interference in the existing steel belt detection. The system synchronously integrates an eddy current sensor, a magnetic flux leakage sensor and an encoder through a multi-probe array adapter; and multi-dimensional damage physical information in the steel strip is obtained. An improved wavelet packet transform-empirical mode decomposition (WPT-EMD) collaborative noise reduction algorithm is adopted, and in combination with a sub-band energy entropy and a self-attention mechanism, non-stationary mechanical noise is effectively filtered out. A multi-rule feature extraction engine is used for extracting smooth residual errors, derivative mutation and other features, a support vector machine (DE-SVM) model introducing physical priori knowledge weights is constructed, and the small sample recognition problem is solved in combination with a virtual sample generation technology. The method can realize high-precision positioning and quantitative evaluation of steel strip damage under complex working conditions, and has the characteristics of strong anti-interference capability, high identification accuracy and good generalization performance.

The invention provides a new energy automobile high-voltage accident grading response and emergency rescue operation guiding method and system, and is applied to the technical field of data processing. Multi-source data of voltage, current and battery states of a high-voltage system are collected, a standardized association sequence is generated after denoising, then the standardized association sequence is converted into a three-dimensional risk map, electric leakage, overvoltage and thermal runaway scenes are analyzed and subdivided through a fault tree, and a classified accident grading model is constructed. Extracting hazard weight factors to establish a dynamic evaluation matrix, and cooperatively calculating multi-source data to obtain key response indexes; parameters are extracted according to accident stages, response levels are divided, and a feature matrix is generated; real-time data and historical data are compared, and correction factors are generated by means of the accident curve slope. And in combination with vehicle types, architecture and environment grouping, key factors are screened by using a support vector machine, a personalized rescue model is established by fusing multiple constraints, and finally, a real-time graded response instruction and scene-divided rescue guidance are output.

The invention relates to the technical field of defect identification, in particular to an AI-based welding defect identification method and system, which comprises the following steps: acquiring regional image data to generate a welding seam region mask image and extracting a welding seam center line to calculate tangential and normal vector field data; adjusting the initial offset vector through projection and weighted synthesis to generate a multidirectional anisotropic feature map, calculating the difference between a defect topology Euler number and a predicted value to obtain a topology constraint loss value, and finally inputting the feature map into a support vector machine to screen confidence so as to judge the defect category. According to the method, anisotropic adjustment is conducted by establishing tangential and normal vector fields of a welding seam and guiding feature extraction offset vectors so as to adapt to the trend and structure of the welding seam, and in combination with topological relation constraints of defect connected components and cavities, the model learns a defect deep structure instead of a surface profile; therefore, the identification precision and robustness of complex and high-directivity welding defects are enhanced, and misjudgment caused by form diversity is avoided.

The invention discloses a method and a system for detecting internal defects of a bathroom glass door based on multispectral imaging. The method comprises the following steps: carrying out multi-view and multiband image acquisition on the bathroom glass door by adopting a combined spectrum light source; performing adaptive reflection correction on the acquired image to generate a multi-band basic image; optimizing the multi-band basic image through a hybrid denoising algorithm to generate a multi-band detection image; performing multi-modal feature extraction and fusion on the multi-band detection image to obtain a multi-modal fusion feature vector, inputting the multi-modal fusion feature vector into a multi-feature fusion model combining a convolutional neural network and a support vector machine, and identifying and grading defect types by adopting a dual-stage feature enhancement-classification architecture; and outputting a defect type and severity grading result to a terminal in real time. The method is used for solving the technical problems of visual angle and spectrum blind areas, insufficient interference suppression capability, wave band crosstalk, feature deficiency and poor small sample generalization in existing bathroom glass door defect detection.

The invention provides a fault diagnosis method and system for a 500kV bus protection device, and the method comprises the following steps: obtaining the operation data of the 500kV bus protection device, and carrying out the preprocessing of the obtained operation data; according to the method, fault feature extraction is carried out by adopting the convolutional neural network in combination with an attention mechanism, fault information can be accurately captured, fault diagnosis and classification are carried out in combination with a support vector machine and a fuzzy logic theory, and the diagnosis accuracy of complex hidden faults is effectively improved; the fault location can be quickly and accurately determined through a fault location algorithm based on the graph theory, a fault isolation strategy is automatically generated, equipment action is controlled, fault expansion is avoided, and the reliability and stability of a power system are improved; by establishing the prediction model, the future operation state of the protection device can be predicted, early warning and prevention of faults are realized, a scientific and reasonable basis is provided for maintenance and repair of equipment, and the maintenance cost and power failure loss of the equipment are reduced.

The invention relates to the technical field of die casting intelligent manufacturing, and discloses an online monitoring and process compensation method and system for residual stress and deformation of a die casting. The method comprises the following steps: embedding a distributed temperature-stress composite sensor array in a mold cavity, and synchronously acquiring temperature and stress signals; after the signal is purified, a space-time correlation matrix is constructed to quantify a thermal-mechanical coupling relation; identifying a stress distribution mode through a support vector machine model, and positioning a fluctuation abnormal region; analyzing a defect mechanism based on mutual information and Granger causality test, and calculating a pore formation probability in combination with fluid dynamics simulation data; matching the pore high-risk location with a historical crack correlation model to generate a crack prediction index; and performing inversion optimization on the mold filling speed and pressure parameters by using the potential quality hazard evaluation function. Real-time monitoring of residual stress and deformation in the die-casting process, defect dynamic traceability and process online optimization are achieved, and air hole and crack defects are effectively restrained.

The invention relates to the field of radio interference intelligent identification and positioning, in particular to a multi-source data fusion water radio interference identification and positioning method and system, and the method comprises the following steps: collecting multi-source data; performing multi-source data fusion processing; constructing a rule base, and screening out suspected abnormal signals from the frequency spectrum situation map through threshold judgment and rule matching; a support vector machine model is called, multi-dimensional static signal features are used as input, a classification hyperplane is constructed through a radial basis kernel function, and normal signals and abnormal signals are distinguished from suspected abnormal signals; calling a decision tree model, and performing scene judgment on the suspected abnormal signal to identify the abnormal signal; calling a convolutional neural network model, converting the time domain signal into a spectrogram through short-time Fourier transform, and extracting texture features through multilayer convolution to identify an abnormal signal; through the method and the system, real-time identification and high-precision positioning of interference signals can be realized.

The invention relates to the technical field of coal processing and utilization, and discloses a clean coal yield prediction method based on a support vector machine, which comprises a data acquisition module used for analyzing factors influencing the clean coal yield, acquiring related data and integrating the data into a data set, and a data preprocessing module connected with the data acquisition module and used for preprocessing the data. The data preprocessing module is used for randomly dividing a data set according to a 70% training set and a 30% test set and carrying out standardization and normalization preprocessing, the parameter optimization module is connected with the data preprocessing module and optimizes hyper-parameters of a support vector machine through an improved grey wolf algorithm, and the model building module is connected with the parameter optimization module and is used for building a model. The model building module is used for building and training a support vector regression model based on the optimized hyper-parameters, and the model verification module is connected with the model building module and uses a test set to verify the performance of the model. According to the method, the hyper-parameters of the support vector machine are optimized through the improved grey wolf algorithm, the problem that a traditional optimization method is prone to falling into local optimum is effectively avoided, and the precision of clean coal yield prediction is remarkably improved.

The invention discloses an access permission adjustment method and device and a computer program product, and relates to the technical field of safety protection, and the method comprises the steps: collecting access behavior data of a source device to a target device; according to the access behavior data and a target reference value corresponding to the access behavior data, performing exception analysis on the access behavior data, and extracting exception features in the access behavior data; converting the abnormal features into numerical vectors; according to the numerical value vector, adopting an isolated forest to calculate an abnormal score, adopting a first-class support vector machine to generate a boundary distance value, and adopting a local abnormal factor algorithm to calculate a local density ratio; performing weighted fusion on the abnormal score, the boundary distance value and the local density ratio to obtain a risk assessment result of the source equipment; and adjusting the access authority of the source device for the target device according to the risk assessment result. The problem that the communication security is reduced due to misjudgment of the access permission of the source device to the target device is solved, and the communication security between the source device and the target device is improved.

The invention provides a multi-algorithm fusion-based water-based scratch-resistant plastic gloss oil formula optimization method, which comprises the following steps of: obtaining coating component data and environment variable data from a preset material database, and constructing an initial data set; performing crossover mutation operation on the initial formula parameters by adopting a genetic algorithm to generate a formula candidate set; according to the component proportion in the formula candidate set, obtaining a hardness index and a flexibility index through simulation, and performing grouping processing on the candidate set to obtain a grouped candidate set; for the grouped candidate set, adopting a support vector machine to carry out classified training on flexibility indexes of the high-hardness group, extracting performance deviation characteristics, and determining an adjustment vector; the formula parameters of the low-hardness group are corrected according to the adjustment vector, and the scratch resistance simulation data and the bending resistance simulation data are fused to obtain a corrected formula set; the coating performance under the environment variable change is sampled by adopting Monte Carlo simulation, performance fluctuation distribution is extracted, and an optimization formula and a process combination are determined.

The invention relates to an engineering machinery remote control method and device based on the Internet of Things, equipment and a medium. The method comprises the following steps: acquiring multi-source data of a working site, and after preprocessing, separating key features by using an optimized support vector machine model to obtain a key feature set; based on the set, utilizing a deep neural network to fuse multi-source data to generate an integration state model, and synchronously and dynamically adjusting data acquisition frequency according to network bandwidth fluctuation to obtain a final integration state model; the transmission delay of the data is evaluated, semantic classification and compression are carried out on the optimized data, and a visual interface is generated after transmission analysis; and acquiring interface interaction data, converting the interface interaction data into a control instruction, calculating a signal parameter modification value, verifying control delay compliance, and finally generating remote control response data containing execution confirmation information. By adopting the method, the intelligent level and the operation safety of remote control of the engineering machinery are remarkably improved, and the labor and operation and maintenance cost is reduced.

The invention discloses a network security monitoring method, which comprises the following steps of: acquiring full-life-cycle data and attributes of network processes in real time, constructing a process chain table and an associated network, and sorting a relationship between the processes; secondly, matching a network event with a process timeline, dynamically dividing a time window according to a process life cycle, calculating a process time characteristic and a network activity characteristic, comparing a historical normal mode, quantifying a difference by utilizing an algorithm, and identifying an abnormal process; then, inputting the comparison process into a support vector machine to establish a prediction model, and predicting an abnormal process in real time; and finally, abnormal process information is fed back to the security equipment, and abnormal files are positioned and isolation / deletion operation is executed in combination with network behavior data association analysis. According to the method, the abnormal process can be accurately identified, misjudgment is reduced, the sensitivity requirements of different service scenes are met, the abnormal file is positioned by means of network behaviors, and the accuracy and timeliness of network security protection are effectively improved.

The invention provides an enterprise supply chain financial risk prediction method fusing a dynamic knowledge graph and a graph neural network. The method comprises the steps of multi-source data loading and preprocessing, supply chain knowledge graph construction, financial feature extraction, risk label generation and the like. Aiming at the problems that multi-subject information of core enterprises, suppliers, customers and the like in supply chain data is dispersed and is inconsistent with independent financial data structures, the method solves the problems that manual integration is low in efficiency and prone to errors through automatic data cleaning, entity matching and relation mapping, and a unified and structured enterprise supply chain relation knowledge graph is constructed; the limitation of manually designing network features is overcome by utilizing a graph attention network; an effective feature fusion strategy is designed, and joint feature representation with higher discriminative force is formed; and constructing an end-to-end evaluation framework, training classifiers such as a support vector machine to learn the fusion features, realizing accurate and stable classification of high-risk and low-risk enterprises, and providing probability output to enhance result interpretability and decision support capability.

The invention discloses a big data-based supply management method and system and a storage medium, and the method comprises the steps: obtaining supply chain order big data, carrying out the data classification of the supply chain order big data based on a support vector machine, carrying out the data fluctuation analysis of the obtained food supply data, obtaining a data fluctuation curve, and carrying out the calculation of the data fluctuation curve based on a linear regression algorithm. Performing data prediction on the basis of the data fluctuation curve, and obtaining price prediction fluctuation data and supply quantity prediction fluctuation data of each supply unit; obtaining supply demand information of a next preset supply period, and analyzing the matching degree of each supply area by taking the supply demand information as a supply target and combining the price prediction fluctuation data and the supply quantity prediction fluctuation data of each supply unit; and in the map model, based on the matching degree and the geographic position of each supply area, carrying out joint supply analysis and logistics aging matching optimization on the plurality of supply areas, and obtaining an optimal supply scheme in the next preset supply period.

The application provides an image recognition-based online detection method and system for surface defects of ultra-wide and thick plates, and relates to the technical field of image recognition. The method comprises the following steps: synchronously collecting surface images and process state parameters containing final rolling temperature, rolling force distribution, laminar cooling intensity and real-time speed, firstly completing image splicing according to the final rolling temperature and the rolling force distribution to obtain a topological image; then separating a texture channel image by feature decoupling with the laminar cooling intensity as a constraint; subsequently extracting edge features of the texture channel image by Hough transformation, and introducing the rolling force distribution and the real-time speed as constraints for curve fitting to generate topological structure features; finally dynamically adjusting a kernel function by a support vector machine, and analyzing the topological structure features and a preset feature library to obtain a defect detection result. The application can effectively improve the accuracy and adaptability of defect detection under complex working conditions through deep fusion and dynamic guidance of process parameters.

The invention discloses an intelligent micro-grid network attack detection method and system based on a block chain, wavelet transform and a support vector machine, a medium and a processor, and relates to the field of power grid network attack detection. The method aims at solving the problems that normal and attack exception are difficult to distinguish, the missing and false detection rate is high, and data are tampered easily in a traditional method. The method comprises the following steps: collecting and preprocessing DC micro-grid data; decomposing into high and low frequency components through wavelet transform, and extracting amplitude, frequency and energy related characteristic parameters; constructing and training a support vector machine model for real-time attack detection; and after detecting full-process data hash processing, uploading the data to the block chain, and storing evidence based on a PoA consensus mechanism. According to the method, attack features are accurately captured through wavelet transformation, high-accuracy classification is realized in combination with the SVM, the block chain guarantees data credibility, a micro-grid dynamic scene can be quickly responded, and the network security protection capability is improved.

The invention discloses an on-demand data cleaning method and system oriented to a support vector machine, relates to the technical field of data cleaning, and provides a new on-demand data cleaning normal form for solving the problem that an existing data cleaning method is low in reliability of a cleaning result. The data and the downstream task model are regarded as a whole, and the data quality oriented to the specific model is further optimized on the basis that the downstream task effect and the generalization requirement are met. The normal form breaks through the limitation that a traditional method only pays attention to the quality of data, and the reliability of a data cleaning result is greatly improved.

This invention discloses a method, system, and device for biochar identification based on support vector machines, belonging to the field of biochar detection technology. This invention collects the main physicochemical properties of waste biomass from different sources and biochar prepared at different carbonization temperatures, and uses these collected physicochemical properties to construct a biochar identification model based on support vector machines. This model can effectively perform the task of identifying biochar with many different properties and improve the accuracy of biochar identification.

The application discloses a white granite type high-purity quartz purity prediction method and application thereof, and belongs to the technical field of rock purity prediction. The prediction method encodes and assigns one or more prediction parameters, such as main mica types, rock structures, rock structures, feldspar types, quartz types, solid inclusions, liquid inclusions, solid inclusion quantity, aluminum impurity element content grading, titanium and lithium element content grading, calcium and phosphorus element content grading, potassium and sodium element content grading and grain factor grading, and then predicts the quartz purity in the ore based on a random forest classifier prediction model or a support vector machine prediction model. The comparison between the actual measurement and the prediction result proves that the prediction method has accurate and reliable prediction results, realizes the replacement of manual detection by artificial intelligence prediction, greatly reduces the manpower and material resources, and provides an efficient and accurate prediction method for quartz purity prediction.

The invention discloses a method for predicting the erosion rate of a control manifold of an offshore oil and gas exploration well. The method comprises the following steps: constructing a support vector machine model; wherein the input parameters of the support vector machine model are the manifold diameter, the manifold bending angle, the fluid speed, the fluid density and the fluid viscosity, and the output parameter of the support vector machine is the maximum erosion rate of the well control manifold; optimizing penalty function coefficients and kernel function parameters of the support vector machine model by adopting a genetic algorithm to obtain an optimized support vector machine model; training the optimized support vector machine model to obtain a well control manifold erosion rate prediction model; and acquiring the manifold diameter, manifold bending angle, fluid velocity, fluid density and fluid viscosity of the to-be-predicted well control manifold, and inputting into the well control manifold erosion rate prediction model to obtain the maximum erosion rate of the to-be-predicted well control manifold. According to the well control manifold erosion rate prediction method provided by the invention, the maximum erosion rate of the well control manifold can be accurately predicted.

The application discloses a kind of student abnormal behavior analysis and early warning method based on deep learning, specifically related to data analysis technical field;By real-time acquisition and student safety related multi-source behavior data, after standardization processing, input deep learning model extracts high-level behavior characteristics, and further calculates behavior fluctuation degree and situation behavior inconsistency index, generates normalized risk score;Combined with the output result of classification model, the suspected pseudo-normal sample that risk is high but is misjudged as normal is identified, and the model classification boundary is optimized in pertinence by local weighted incremental support vector machine, to significantly improve the identification ability and early warning accuracy of the system to implicit abnormal behavior, enhance the robustness and practicality of model in complex campus behavior scene.

The application relates to an enterprise recommendation method, device, equipment and medium based on big data, which comprises the following steps: obtaining public Internet data of a target enterprise, obtaining various policy files and related public files of an industry to which the target enterprise belongs, analyzing each enterprise in the industry to form an enterprise list, obtaining data of declaration conditions met by each policy project declared by each enterprise in the enterprise list, performing data normalization processing on all the obtained original data, forming low-dimensional feature vector data, mapping the low-dimensional feature vector data to a high-dimensional feature space based on a support vector machine by using a kernel function, directly calculating the inner product difference between the centralized high-dimensional new data and the support vector of the high-dimensional feature space in the high-dimensional feature space, eliminating the enterprise corresponding to the high-dimensional new data with a negative inner product difference, and outputting similar enterprise recommendation data of the target enterprise after sorting, so that enterprise recommendation processing about the target enterprise is realized, and the enterprise recommendation accuracy is greatly improved.