77 results about "Rough set" patented technology

The invention relates to the technical field of automatic driving automobiles, in particular to a scene extraction, generation and evaluation method for a vehicle safety test, which adopts a semantic segmentation network to keep scene element features in original natural driving data to the greatest extent. Based on a high-level decision deep reinforcement learning algorithm, a rough set is used to extract an association generation rule of a typical dangerous scene, the typical dangerous scene can be automatically generated in combination with simulation software, and a certain design thought is provided for automatically extracting and generating the characteristics of the scene by referring to natural driving data; meanwhile, the overall intelligence of the self-driving automobile is evaluated in combination with two dimensions of natural driving evaluation and test scene complexity evaluation, and key index advantages in different aspects are combined, so that the multi-dimensionality of evaluation of the self-driving automobile is improved; and the scene generation efficiency and the intelligent level are greatly improved.

The invention discloses an intelligent judgment method and system for comprehensive performance of an electric energy meter. The method comprises the following steps: preprocessing test data of comprehensive performance detection of the electric energy meter based on a rough set theory, and extracting key features from the preprocessed test data through a feature extraction technology of scale invariant feature transformation to obtain key features of the test data; and performing comprehensive analysis on the preprocessed test data and key features to obtain a comprehensive analysis result, and performing comprehensive judgment on the comprehensive performance of the electric energy meter according to the comprehensive analysis result to realize intelligent judgment on the comprehensive performance of the electric energy meter. The accuracy and the stability of the feature points are ensured, so that a reliable data basis is provided; information of different levels is effectively integrated, and comprehensiveness and accuracy of comprehensive evaluation are improved; samples with abnormal performance are effectively identified, and the accuracy of abnormal identification is improved.

The invention relates to the technical field of foundation pit detection, and discloses a deep foundation pit deformation prediction method and system based on neural network and rough set classification, and the method comprises the following steps: data collection and preprocessing; carrying out attribute reduction based on a rough set theory; constructing and training a model based on an improved attention mechanism recurrent neural network (A-RNN); carrying out credibility verification and dynamic feedback adjustment; and outputting a result and an early warning response. By integrating multi-parameter monitoring data of displacement, soil pressure, underground water level, support stress and the like, driving factors of deformation of the deep foundation pit are comprehensively captured, misjudgment caused by data deviation of a single sensor is reduced, the combination of bidirectional LSTM and an attention mechanism effectively captures local features of long-term dependence and key time points in time sequence data, and the accuracy of deep foundation pit deformation detection is improved. And the accuracy of deformation prediction is obviously improved.

The invention provides an image noise mark feature selection method and system, a storage medium and a computer. The method comprises the steps of obtaining a to-be-processed image noise mark data set; embedding a sample set in the image noise mark data set into a multi-granularity fuzzy cluster to construct a dynamic fuzzy membership evaluation matrix; dynamically evolving a multi-level high-precision granular ball cluster; obtaining mark distribution with high identification degree; constructing a rough perception feature evaluation framework based on granular ball topology driving, extracting decision equivalence classes by combining rough set upper and lower approximation and extended positive domain models, and determining and measuring the contribution degree of each feature to a decision system by fusing multi-granular-ball decision boundary information based on a dependency degree quantitative model; a particle and ball structure consistency verification mechanism is introduced, and multi-level evaluation is carried out on the importance of the features through dependency and consistency. According to the method, the optimal feature subset with strong anti-noise performance and high discrimination capability is obtained, and stable and efficient input support is provided for a subsequent image noise mark learning model.

The invention provides a network security attack threat analysis method and device based on a rough set, and relates to the technical field of network security. The method comprises the following steps: collecting real-time network flow data, analyzing the characteristics of the network flow data by applying a data attribute evaluation technology in a rough set theory, evaluating the fluctuation intensity and the fluctuation amplitude of the data, dynamically adjusting the size of a time window, and obtaining a time window size adjustment parameter. According to the invention, abnormal data are quickly positioned through statistical characteristics, the efficiency and response speed of network security monitoring are improved, resources are dynamically configured by evaluating the abnormal frequency and severity in the data stream, and the frequency of network monitoring is optimized, so that the network security management is more active and prospective, and the network security management efficiency is improved. The response and processing capabilities of the network are detected by dynamically adjusting the attack frequency, the network security configuration is further optimized, and the overall security and stability of the network are greatly improved.

The invention provides a highway pavement intelligent maintenance scientific decision-making system and a decision-making method thereof. A data collection and preprocessing module collects multi-source data in a public network and a highway management intranet, and performs cleaning, integration and dimensionality reduction by using a specific algorithm. And building an automatic road detection system based on a Python Django framework, determining an index weight by using a rough set theory, calculating PQI and MQI values, and generating a road technical condition evaluation report. And the pavement performance prediction module integrates multi-source data and predicts pavement performance by adopting a composite kernel function and a Gaussian process regression model. And the maintenance decision module determines the maintenance priority through ANP, generates a maintenance demand list in combination with a hidden Markov model, and generates a fund distribution scheme by using a cuckoo search algorithm. And the maintenance benefit evaluation module evaluates the maintenance benefit by adopting a gray ideal solution and a super-efficiency DEA model. And finally, utilizing a WebGL-based 3D visualization technology and an Echarts-based interactive query analysis function to realize road condition information display and auxiliary decision making.

The present invention relates to the technical field of wheel-rail tribology. Specifically disclosed is a wheel-rail material rolling contact fatigue damage prediction method based on different evaluation indices. The method comprises the following steps: on the basis of a wheel-rail shakedown map, carrying out wheel-rail rolling contact fatigue simulation tests, so as to obtain damage states and shakedown limit curve equations of wheel / rail materials with different shear yield strengths under different contact parameters; on the basis of a rough set theory, calculating the importance and relative weights of different evaluation indices; establishing a comprehensive evaluation index cloud model, respectively calculating comprehensive certainty degrees of an index attribute value to be evaluated belonging to different damage states, and using a maximum-value rule for the comprehensive certainty degrees to evaluate and predict rolling contact fatigue damage states; and using comprehensive evaluation index attribute values to perform validation, so as to finally determine a prediction result of a selected wheel-rail material rolling contact fatigue damage state under a condition to be subjected to prediction. The method in the present invention only depends on existing engineering data, such that the subjectivity of a damage state prediction process can be greatly reduced.

The invention relates to the technical field of concrete durability evaluation, in particular to a compound green concrete salt freezing resistance durability evaluation system under a frozen soil environment, which comprises a data input unit, a decision information table storage unit, a rough set analysis processing unit and an evaluation output unit, the data input unit receives a sample data set containing freeze-thaw cycle times, dry-wet cycle times, freeze-thaw-dry-wet cycle times, long-term soaking time, composite salt concentration, gas content and fly ash mixing amount, and the decision information table storage unit stores a decision information table constructed based on a rough set theory. The influence factor combination and the relative dynamic elastic modulus percentage are associated, the rough set analysis processing unit establishes a mapping relation through attribute reduction and rule extraction and quantifies the weight of each factor, and the evaluation output unit outputs a relative dynamic elastic modulus predicted value, durability grade classification and key factor contribution sorting. And the salt freezing resistance durability of the concrete under the multi-factor coupling effect is accurately evaluated.

The invention discloses a TBM cutterhead torque prediction method, system and device and a medium. The method comprises the steps that tunneling state parameter initial sample data influencing the cutterhead torque are obtained; the initial sample data are preprocessed, and abnormal tunneling state data in the TBM tunneling state parameters are removed; calculating the weight of each tunneling state parameter influence factor by adopting a rough set; selecting several characteristic parameters having the highest correlation with the cutterhead torque by using the rough set, forming a matrix by the characteristic parameters, and inputting the matrix into an LSTM neural network for training; and calculating a loss function according to TBM cutterhead torque prediction results output by the LSTM model in sequence, optimizing model parameters according to loss function calculation results, and finishing model training when training conditions are met to obtain a TBM cutterhead torque prediction model. According to the method and the system, the torque of the cutterhead can be quickly and accurately calculated in the tunneling process of the TBM equipment, so that reference is provided for an operator, the tunneling efficiency is improved, and equipment faults, cost increase and the like caused by improper selection of operation parameters are reduced.

The invention belongs to the field of unmanned ship navigation, and relates to an unmanned ship marine obstacle environment modeling method based on rough set data fusion. Comprising the following steps: (1) constructing a two-dimensional grid map by taking the position of an unmanned ship as a center; (2) the unmanned ship is equipped with m different types of obstacle sensors, and the two-dimensional grid map is divided into n areas according to different detection capabilities of the obstacle sensors; (3) establishing n information tables for n areas in the two-dimensional grid map; (4) performing condition attribute reduction on the information table of the n areas in the two-dimensional grid map by utilizing the dependency degree of the decision attribute in the information table on each condition attribute; (5) utilizing n simplified information tables in the two-dimensional grid map to classify and conclude a decision fusion rule for judging whether the grid is an obstacle area or not; and (6) judging whether each grid is an obstacle area or not, and completing modeling. According to the method, a new thought is provided for marine obstacle environment modeling of the unmanned ship, and the detection precision is improved.

The invention belongs to the technical field of nuclear power equipment detection, and particularly relates to a loading and unloading machine state monitoring and fault diagnosis method. Comprising the following steps: step 1, design function analysis; 2, building a hardware detection system; 3, establishing a software detection system; 4, testing software and hardware; 5, generating a data set; and step 6, fault diagnosis. The method has the beneficial effects that software and hardware logic detection of the loading and unloading machine system is realized, and a data set is established. On-line monitoring of the operation state of the loading and unloading machine equipment is achieved, a fault diagnosis model based on the rough set theory is used for evaluating the operation state of the system in real time and diagnosing and positioning faults of the loading and unloading machine, and maintenance suggestions are given.

The invention discloses a civil aircraft cockpit display interface availability evaluation index system construction method, which comprises the following steps: firstly, analyzing pilot human-computer interface interaction cognition through an EPIC model, extracting four-dimensional availability indexes, and preliminarily screening the availability indexes based on the experience and knowledge of a pilot; a civil aircraft cockpit display interface availability evaluation index system is constructed; secondly, in combination with a hierarchical clustering algorithm and a rough set information entropy algorithm, an HC-CEBARKNC algorithm is proposed to perform clustering reduction on the established index system, and a clustering reduction set of indexes of all dimensions is obtained; thirdly, verifying the reliability of each dimension index clustering reduction set by using the classification accuracy of the SVM so as to construct a reasonable civil aircraft cockpit display interface availability evaluation index system; and finally, calculating the weight of each index in the index system by using a rough set conditional information entropy improvement formula. The problem that an index system for civil aircraft cockpit display interface availability research is not comprehensive enough is solved.

The invention discloses a medical examination index reduction method based on rough set positive domain and information entropy synthesis, is suitable for high-dimensional data attribute reduction and knowledge discovery in scenes such as medical data analysis, and provides a new reduction strategy by fusing two perspectives of a rough set theory and an information theory method. The comprehensive importance degree of the attributes is obtained through a weighted average mode, and the weight is determined by the ratio of the positive domain to the boundary domain. Iterative reduction is carried out according to the comprehensive importance, deterministic and non-deterministic elements are reasonably considered, the problem of feasibility of attribute reduction of a large-scale fuzzy information decision system is solved, and a more accurate reduction set is obtained. Key diagnosis indexes are conveniently focused, the diagnosis efficiency and accuracy are improved, the clinical interpretability is enhanced, unnecessary detection items are reduced, and the effects of optimizing data, improving the diagnosis efficiency and saving the medical cost are achieved.

The invention provides a partial multi-label feature selection method and device, equipment and a medium, and relates to the technical field of data preprocessing. The method comprises the following steps: S1, acquiring an original training data set of partial multi-labels, initializing a label confidence coefficient matrix according to the original training data set, and constructing a distribution area of the labels; and S2, according to the distribution of the neighbors, updating the label confidence coefficient matrix, disambiguating the distribution area of the untrusted labels, and iteratively updating the distribution area of the uncertain labels until the maximum number of iterations is reached, thereby obtaining the label after the disambiguation. And S3, according to the disambiguated labels, combining granularity calculation and a graph theory, aggregating the labels from fine granularity to coarse granularity, and constructing virtual coarse granularity labels. And S4, constructing a neighborhood rough set model suitable for partial multi-label data according to the virtual coarseness label in combination with a plurality of features acquired from the sample data, and selecting the feature with the highest importance by adopting a forward greedy search algorithm.

The invention belongs to the technical field of geochemical numerical simulation and optimization, and particularly relates to a pollution source fingerprint spectrum construction optimization method and system.The method comprises the steps that pollution source sample data sets are collected, the pollution source sample data sets are classified, and specific pollutants in a fingerprint spectrum are determined; the pollution source sample data set is screened, and a learning sample set and a training sample set for pollution source type matching are screened out; calculating pollution source characteristic factor vectors in the pollution source fingerprint spectrum and the proportion of each characteristic factor vector in combination with a rough set training model; and manufacturing a pollution source fingerprint spectrum. And then optimizing the heavy metal migration enrichment model and optimizing the pollution source fingerprint spectrum. According to the method, the respective advantages of a machine automatic learning screening model and a migration enrichment model are coupled through pollution source component optimization factors, and the accuracy and reliability of the pollution source fingerprint spectrum in source analysis calculation can be obviously improved.

The present invention discloses a medical test index simplification method based on the combination of rough set positive domain and information entropy, which is suitable for high-dimensional data attribute simplification and knowledge discovery in scenarios such as medical data analysis. By integrating the two perspectives of rough set theory and information theory methods, a new simplification strategy is proposed, which obtains the comprehensive importance of attributes by weighted average, and its weight is determined by the ratio of the positive domain to the boundary domain. It performs iterative simplification according to the comprehensive importance, reasonably takes into account the deterministic and non-deterministic elements, solves the feasibility problem of attribute simplification in large-scale fuzzy information decision-making systems, and obtains a more accurate simplified set. It is convenient to focus on key diagnostic indicators, improve diagnostic efficiency and accuracy, enhance clinical interpretability, reduce unnecessary test items, and play a role in optimizing data, improving diagnostic efficiency, and saving medical costs.

The invention discloses an interval particle neighborhood rough set three-way decision-making-based classification method, which comprises the following steps of 1, acquiring a data set, and converting the data set into a tetrad decision-making information system; 2, obtaining an interval particle set of the decision information system; 3, obtaining a threshold pair through a three-branch approximation method of the shadow set, and obtaining the average ambiguity of the interval grain shadow set according to the threshold pair; 4, obtaining a threshold pair corresponding to the minimum fuzzy loss between the average ambiguity of the interval grain set and the average ambiguity of the interval grain shadow set, and taking the threshold pair as an optimal threshold pair; and 5, according to the optimal threshold pair, identifying an interval particle neighborhood of the prediction object, and classifying the prediction object. According to the method, higher-quality decision information system processing and expression are achieved, and the accuracy and robustness of the model are further enhanced.

The invention relates to the technical field of medical data mining and mode recognition, in particular to a special child early disease prediction method based on double-layer granular ball knowledge representation BGBK and three-way role arbitration TRA. The method comprises the following steps: firstly, converting original special child data into multi-granularity granular ball representation through an unsupervised granular ball generation algorithm; a BGBK structure is constructed, and coarse-grained particle ball information and fine-grained sample information are fused; a three-boundary neighborhood rough set theory is utilized to endow semantic roles to the granular balls, wherein the semantic roles comprise a core region, an abnormal boundary and a transition boundary; designing a TRA strategy, and calculating a sample abnormal score in combination with the role influence factor; and finally, constructing an abnormal factor by fusing the abnormal scores under the multi-attribute subspace, and realizing accurate early prediction of the special child disease. The method can effectively process complex special child data, has the advantages of high detection precision, strong robustness, good interpretability and the like, and significantly improves the accuracy of early prediction of special child diseases.

The present invention relates to the field of coal and rock mining technology, and in particular, provides a method for evaluating the rock burst hazard of coal mines. The method includes constructing an index system for evaluating the rock burst hazard of the amount of coal and rock ejected, and obtaining a training data set and a test data set through data regularization; optimizing and analyzing the constructed index system using a fuzzy neighborhood decision rough set model and the training data set to obtain a simplified subset; inputting the obtained simplified subset into a multi-classifier model to obtain a classification result, and performing a classification accuracy evaluation on the classification result, completing the weighting of the classifier based on the classification accuracy evaluation, and obtaining the classifier weight and the trained multi-classifier model; inputting the test data set into the trained multi-classifier model for testing to obtain a test result, and weighting and fusing the test results to obtain a hazard evaluation level, and the method realizes the accurate division of the rock burst hazard level of coal mines.

The invention relates to the technical field of multi-source information processing, in particular to a multi-source information decision fusion method and system for a multi-granularity multi-view rough set, and the method comprises the following steps: obtaining multi-source heterogeneous data of an object to be decided, constructing a multi-source decision information table through cleaning and standardization processing, and dividing the information table into a plurality of view subsets; and constructing a multi-granularity hierarchical structure based on the view subset, calculating rough set approximation precision and attribute dependency under different granularity hierarchies, and generating a local decision rule set. And performing weighted aggregation on the local decision rule set by using a multi-source information fusion algorithm, and constructing decision evaluation parameters based on consistency measurement. According to the method, the problems of one-sided information under a single view angle and detail loss under a single granularity are effectively solved through the calculation of the multi-view and multi-granularity collaborative mechanism reverse driving decision cost, and the accuracy and robustness of decision analysis under a complex and uncertain environment are remarkably improved based on the feedback adjustment of the dynamic weight; and deep mining and efficient fusion of multi-source information are realized.

The invention belongs to the technical field of new energy automatic optimization, and discloses a fault diagnosis and optimization method based on a rough set and a random forest, and related equipment. The fault diagnosis and optimization method based on the rough set and the random forest comprises the following steps: inputting multi-source heterogeneous operation data into a constructed fault diagnosis model, and outputting early warning information, the early warning information comprising a fault category; wherein the fault diagnosis model training method comprises the steps of performing knowledge reduction on a multi-dimensional data set by adopting a rough set algorithm, inputting an equipment operation sample set after knowledge reduction into a random forest model, performing multi-target optimization analysis through a big data platform based on early warning information and real-time operation data, and generating decision support scheme data; according to the method, a safe and executable operation and maintenance strategy and an operation instruction can be generated in real time by utilizing a multi-objective optimization technology, and the closed-loop technical bottleneck from data fusion, intelligent diagnosis to dynamic optimization is effectively broken through.

The invention provides a multi-dimensional quality and efficiency evaluation method and device for cluster and field power grid projects, and the method comprises the steps: taking a project cluster and a project field as research objects, obtaining a project data set corresponding to a power grid project cluster / field, fully considering the relevance between projects, determining the key evaluation dimension and evaluation index of the project cluster / field, and achieving the quality and efficiency evaluation of the power grid project cluster / field. A comprehensive quality and efficiency evaluation index system is constructed by combining existing data setting specific indexes and calculation formulas thereof, the index weight of each evaluation index is determined according to a rough set theory, the objectivity of weight distribution of each evaluation index is improved, and comprehensive evaluation of project clusters and fields is achieved.

The invention relates to the technical field of data fusion, and discloses a multi-sensor data fusion method, device and equipment and a storage medium, and the method comprises the steps: carrying out the synchronization of laser radar data, depth image data and ultrasonic distance data, obtaining a first sensor data set, and constructing an obstacle behavior prediction model; performing Bayesian posterior probability calculation on a second sensor data set acquired in real time according to the obstacle behavior prediction model to obtain a fusion trajectory prediction result and an uncertainty value; performing neighborhood rough set rule reduction based on the fusion trajectory prediction result to obtain a target obstacle avoidance decision rule group; according to the method, the prediction uncertainty is quantified into the risk coefficient to dynamically adjust the conservative degree of the obstacle avoidance strategy, the coupling analysis of the prediction confidence and the control strategy is realized, and the dynamic obstacle avoidance execution instruction is generated. And the cleaning efficiency and the obstacle avoidance safety are effectively balanced.

The invention discloses a leg posture analysis and rating system for a football player in a ball kicking process, and the system comprises a data collection unit which collects four-stage kinematics parameters of kicking, kicking and stretching, backward swinging, forward swinging and ball touching through a three-dimensional motion capture system and a high-speed camera; the data processing unit is used for processing parameters and extracting characteristic parameters of each stage; the expert evaluation interface is used for receiving evaluation data of multiple experts on different ball kicking technology types on the multi-score influence factors; the weight calculation unit is used for calculating an influence factor attribute weight by using a Bayesian method based on expert data; the rating decision matrix construction unit is used for constructing a comprehensive matrix in combination with a weight rough projection method; the multi-angle analysis unit is used for performing multi-angle risk rating by fusing a rough set theory and an approximate ideal solution sorting method; and the result output unit is used for outputting the posture quality grade and the training suggestion. According to the method, the actual requirements of football training on precision, standardization and comprehensiveness of posture analysis can be met.

The invention discloses a fatigue analysis and diagnosis method for key components of a circuit breaker operating mechanism. The fatigue analysis and diagnosis method comprises the following steps: (1) collecting vibration signals in the operating process of a circuit breaker; and (2) tracing the energy storage spring deformation, and finely dividing vibration signals accompanying the operation process into three sub-stages: a tripping trigger stage, an energy storage spring deformation stage and a closing buffer stage. And (3) carrying out adaptive variation mode extraction (AVME) by using the signal in the deformation stage of the energy storage spring, so as to obtain an expected mode. And (4) extracting typical time-frequency domain characteristics of an expected mode as condition attributes of a theoretical information table of key components of the operating mechanism, taking whether fatigue occurs as decision attributes, and establishing a fault original decision information table. And (5) reducing the fault original decision information table by using a rough set theory to obtain a fault reduction decision information table. And (6) training and learning a support vector machine (SVM), and constructing a fault diagnosis model for the key components of the circuit breaker operating mechanism. Experiments show that the method has high accuracy and practicability in the field of circuit breaker fault identification and evaluation.

The rule learning unit 81 performs rough sets analysis using training data that includes threat information including a plurality of explanatory variables representing a threat event and a discrimination result of discriminating the threat information, to learn a decision rule specifying the discrimination result depending on a combination of the explanatory variables. The input unit 82 inputs the threat information to be analyzed. The analysis unit 83 applies the input threat information to the decision rule to identify the discrimination result of the threat information, and the explanatory variable as a basis for the discrimination result.

The invention discloses an autonomous sensing method for tunnel full-section mechanical construction and related equipment, relates to the field of tunneling, and mainly aims to solve the problem that fault detection of tunneling equipment for tunnel full-section mechanical construction is not rapid and accurate enough. The method comprises the following steps: determining optimized operation data of tunneling equipment; dividing the optimized operation data based on a rough set theory to construct a fault diagnosis model which is an RBF neural network; and determining the fault type and position based on the fault diagnosis model so as to autonomously perceive the tunneling equipment. The method is used for the autonomous sensing process of tunnel full-section mechanical construction.

The application discloses a kind of based on adaptive high-dimensional data local differential privacy protection method, using rough set theory and mutual information, the data of different relevance and sensitivity are clustered grouping, different grouping is used different random response probability, permanent random response and temporary random response are carried out.Local differential privacy processing data, using SNE method for dimension reduction and adaptive sampling gradient optimization in server end, with low-dimensional data joint probability distribution approximates the joint distribution of all attributes, according to the sampling strategy and the approximate joint distribution constitutes release dataset.The application solves the problem of high-dimensional data with dimension disaster and local differential privacy combination and maximum degree maintaining data relevance, and more reasonably allocates privacy budget according to the different data sensitivity to some extent, improves data availability and reduces communication cost.

The present invention discloses a method and system for dividing operation modes of a high-proportion distributed new energy power grid. The method obtains and pre-processes the operation data of the power grid, then coordinates the factors affecting the operation mode of the power system, and obtains the characteristics of each factor through logical analysis; defines the characteristic quantity expression corresponding to the characteristics of each factor, calculates the characteristic quantity of each factor and normalizes it; clusters each factor according to the normalized characteristic quantity of each factor, obtains several typical scenes of the factor, and extracts the typical curve of each typical scene of each factor; combines the permutations and combinations of the typical curves of the typical scenes of various factors in actual daily operation, and obtains the normalized operation mode set of the factors actually existing in the power grid; clusters the normalized operation modes of each factor according to the weighted characteristic vector, and obtains a rough set of power grid operation modes, and obtains the final power grid operation mode set through aggregation, merging, calibration and supplementation.

The present invention provides a diabetes parallel attribute reduction method based on collaborative evolution discrete particle swarm optimization, which belongs to the field of medical electronic medical records technology. It solves the technical problem that the diabetic electronic medical record data is large in dimension and redundant, which leads to doctors' wrong judgment. Its technical solution is: including the following steps: S1: storing diabetes data in a distributed file system; S2: computer nodes read the data of blocks in HDFS; S3: after the master node obtains the summarized data key-value pairs; S4: the master node performs the operation of step S3<key,value> The key-value pair data is broadcast to each child node; S5: the master node models the obtained attribute evaluation function in combination with the CQBPSO algorithm. The beneficial effects of the present invention are: the present invention combines the rough set theory and the Spark distributed computing platform to screen out the most representative and critical attributes from the diabetes data set.