33 results about "Cluster coefficient" patented technology

The invention relates to the technical field of platform management, and discloses an internet-based financial service platform management method, which comprises the following steps: collecting static identity data, dynamic behavior data, biological characteristics and environmental parameters of a user in real time, and generating a multi-dimensional dynamic identity chain after desensitization aggregation; the attack-defense engine simulates an attack strategy by using a diffusion model, analyzes causal contradictions and topology aggregation coefficients in combination with a space-time diagram convolutional network, marks high-risk nodes, and iteratively optimizes model parameters everyday through reinforcement learning; quantum signature and alliance chain verification are adopted in the transaction link, transactions involving high-risk countries or transactions with the fund abnormal transfer probability exceeding 95% are automatically frozen, and compliance channels are switched; the biological monitoring module analyzes the physiological signal in real time and triggers a multi-mode challenge; and aggregating edge node data through a federated learning framework, and automatically rolling back to a historical optimal version when the false alarm rate of the updated global risk control model exceeds the standard. According to the invention, efficient management of the financial service platform can be realized.

The invention requests to protect a supply chain network toughness evaluation method based on a complex network theory, and aims to evaluate and optimize the toughness of a supply chain network under different attack strategies (such as random attack and deliberate attack) by establishing an evolution model of the supply chain network. According to the method, a network model with enterprises as nodes and cooperation relations as edges is established, key parameters (such as node degree, shortest path length and clustering coefficient) in a network structure are analyzed, and the stability and recovery capability of a supply chain under external impact are evaluated. By simulating different risk scenes, the structural performance, the efficiency performance and the recovery capability of the network are calculated and evaluated. The toughness evaluation method provides a new theoretical basis and technical means for supply chain risk management, and is helpful for improving the toughness of the supply chain network.

The invention belongs to the technical field of offshore wind power networking sending-out scheme evaluation, and discloses a large-scale offshore wind power networking sending-out multi-dimensional comprehensive evaluation method, system and equipment based on grey cloud whitening weight clustering and a medium, so as to solve the problem that the evaluation result of the existing method is difficult to comprehensively reflect the actual performance of an offshore wind power networking sending-out scheme. The method comprises the following steps: constructing a multi-dimensional comprehensive evaluation system; performing quantitative evaluation on each evaluation index in the multi-dimensional comprehensive evaluation system by adopting a microscopic evaluation model; weighting each evaluation index in the multi-dimensional comprehensive evaluation system; constructing a grey cloud whitening weight clustering macroscopic comprehensive evaluation model; on the basis of the grey cloud whitening weight clustering macroscopic comprehensive evaluation model, grey cloud clustering coefficients, belonging to grey classes, of evaluation indexes are calculated, after normalization, the grey cloud clustering coefficients and comprehensive weights are subjected to weighted synthesis to obtain comprehensive clustering coefficients, and therefore the evaluation result of the offshore wind power networking sending-out scheme is calculated. According to the method, the evaluation accuracy and reliability of the offshore wind power networking sending scheme are improved.

The invention discloses a fracture seepage prediction method combining a yield criterion and network topology, and belongs to the field of rock mass engineering construction, and the method comprises the following steps: carrying out topology analysis on a fracture network based on a complex network theory, and determining the degree centrality of fracture network nodes and a fracture shear stress transfer coefficient; establishing a shear stress equation based on a Mohr-Coulomb yield criterion, and determining a criterion of fracture network damage leakage according to the yield criterion and a seepage stress coupling relationship; considering degree distribution and clustering coefficients of fracture network nodes, and establishing a damage evolution equation comprehensively considering node importance and fracture damage characteristics; and based on the established damage evolution equation, carrying out fracture seepage damage evolution to predict the percolation region. According to the method, the importance of the fracture network nodes is identified by using complex network indexes, and failure calculation is performed on fracture edges and the nodes in combination with the yield criterion and the seepage-stress coupling relationship condition, so that the fracture network damage evolution and seepage prediction efficiency is effectively improved.

The invention relates to an SDGs causal network toughness optimization method based on multi-modal data, and belongs to the technical field of sustainable development complex network data analysis, and the method comprises the following steps: obtaining multi-modal time sequence data representing SDGs, and carrying out missing value interpolation, standardization and time alignment processing on the data; predicting the future progress of each SDG based on a gated cycle unit model; identifying the causal relationship between the SDGs by using a panel vector autoregression model; calculating a potential interaction probability in combination with a local path index and a weighted random walk restart model, complementing missing links and forming a complete causal network; a multi-scene disturbance experiment is utilized, and a weighted clustering coefficient and weighted global efficiency are adopted to comprehensively calculate the overall toughness index of the network; and taking the toughness index as a fitness function, and performing global optimization by using a genetic algorithm to obtain an optimal network structure configuration and a priority management path. According to the method, the interaction relationship between the SDGs is quantitatively identified, the network structure and function toughness is improved, and technical support is provided for collaborative promotion and policy optimization of the SDGs.

The invention discloses an airport-high-speed rail traffic infrastructure double-layer network anti-disaster toughness recovery analysis method. The method comprises the following steps: establishing an airport-high-speed rail traffic infrastructure double-layer network model, and carrying out characteristic analysis on the model; calculating relative importance of a field and a high-speed rail network layer and intra-layer importance of nodes; calculating global importance of nodes in the model; selecting airport and high-speed rail nodes with large node degree values, aggregation coefficient values and global importance values for damage to obtain failure nodes in the airport-high-speed rail traffic infrastructure double-layer network model; the model toughness is evaluated by calculating the change of the double-layer network efficiency before and after node failure; and determining an optimal recovery sequence of the failure nodes based on a genetic algorithm. According to the method, the disaster resistance toughness of an airport-high-speed rail traffic infrastructure double-layer network is effectively improved, and an analysis method is provided for improvement of the operation efficiency of comprehensive three-dimensional traffic infrastructures in China and overall allocation of recovery resources under disasters.

The application discloses a kind of distributed mass node community division methods based on graph theory, belong to social network detection field, by the random sampling of complex network relation, cut is carried out in combination with weighted logarithm GN algorithm, interval calculation module degree and distributed finite iteration, can complete 100,000+ magnitude node relation complex network cutting in short time, and WS small world verification is carried out to the community after cutting, this method can effectively overcome the defects of existing method, can ensure accurate efficient real social network detection, the present application retains logarithm weighted relationship weight and finite random iteration distributed cutting, realize accurate efficient cutting of mass complex network, using small world theory ensures that the clustering coefficient of division community meets real social network.

This invention discloses a method for decomposing complex networks based on graph contrastive learning and multi-hop aggregation, comprising the following steps: collecting the number of neighbor nodes, connecting edges, and average clustering coefficients of a complex traffic network to construct a network decomposition model; inputting the original graph into a role graph generation module to obtain a role graph; inputting the original graph and the role graph into a multi-view representation learning module to obtain multi-angle graph representations of the role graph and the original graph; obtaining an importance score for each node; calculating and optimizing the joint loss function to train the network decomposition model; decomposing the traffic network to obtain the importance values ​​of traffic nodes in the traffic network, and setting stronger security measures for traffic nodes with high importance and weaker security measures for nodes with low importance. This invention uses intra-graph contrastive learning and cross-contrast learning to improve graph representation performance and proposes a multi-hop aggregation mechanism to predict node importance by combining multi-hop neighbor information, achieving high-performance network decomposition.

The invention discloses an ultra-deep crack plugging layer mechanical structure evaluation method based on a topological structure, and relates to the technical field of leakage control. The method comprises the following steps that reservoir conditions of a well leakage position are obtained, and a plugging layer of a plugging formula to be evaluated is obtained in combination with an indoor plugging experiment; cT scanning is carried out on the plugging layer, and the geometric center and the particle size of the particle material are obtained through a digital image processing technology; judging whether particles are in contact or not based on the geometric center and the particle size of the particle material, and constructing a particle material contact network topology model according to the contact relationship; and calculating the matching property, the particle node betweenness, the network betweenness centrality and the clustering coefficient of the particle contact network based on the particle material contact network topology model, and evaluating the plugging layer based on the parameters. The method disclosed by the invention is relatively simple and rapid, the core characteristics of the mechanical structure of the plugging layer can be extracted by directly analyzing the particle contact network representing the geometric skeleton of the plugging layer, and the evaluation process is more direct and efficient.

The invention provides a protein structure prediction method based on a protein network, and the method comprises the steps: obtaining protein sequence data and three-dimensional structure data, and constructing the protein network; extracting topological characteristics according to the protein network, wherein the topological characteristics comprise edge weight, node degree, clustering coefficient, betweenness centrality, path length and modularization coefficient; based on the topological features and the three-dimensional structure data, training at least three machine learning models to construct a prediction model, the prediction model being capable of identifying a mapping relationship between the topological features and a protein structure; obtaining topological characteristics of to-be-detected protein, inputting the topological characteristics into the trained prediction model, and outputting to obtain a preliminary prediction three-dimensional structure of the to-be-detected protein; and optimizing the preliminary predicted three-dimensional structure through an optimization algorithm, and outputting to obtain a predicted three-dimensional structure of the to-be-detected protein. According to the method, the protein structure prediction efficiency and accuracy can be effectively improved, and the calculation cost is reduced.

The present invention belongs to the field of signal processing and analysis technology, and specifically relates to a method for identifying EEG-driven behaviors based on the dynamic evolution of coherent network features. The method comprises the following steps: Step 1: Standard preprocessing is performed on the original EEG signal, the beta wave frequency band is selected as the analysis frequency band, and the coherence value within the analysis frequency band is calculated to measure the synchronization strength of the two EEG signals. This value is used as the network connection weight to construct a weighted coherent network; Step 2: At each time point, the degree and clustering coefficient of each node in the weighted coherent network are extracted. For each pair of nodes, the instantaneous phase difference is calculated, and the spatial geometric distance between the two electrodes is introduced to obtain the behavioral characteristic response index. Step 3: The power spectrum parameters of the target behavior signals in all brain regions within the analysis frequency band are obtained, and the target behavior recognition value is compared with the behavior threshold to determine whether the target behavior is activated. The present invention significantly improves the accuracy, real-time nature, and physiological interpretability of behavior recognition.

The invention discloses a structure health assessment method based on stress root spectrum and ultrasonic cluster system superposition sound. The method comprises the following steps: stress root spectrum analysis of experimental structure characteristics; solving a superposition coefficient of the stress root spectrum and the ultrasonic cluster system; solving a self-adaptive clustering coefficient of the superposition coefficient; and solving the structural health coefficient. According to the method, recessive damage characteristics in stress distribution and ultrasonic cluster system response of the large-scale structural member can be effectively extracted, separation of irrelevant interference in a multi-source heterogeneous signal and accurate judgment of damage sensitive information are realized through stress root spectrum analysis and cluster system superposition sound fusion, and adaptive clustering is completed according to response similarity on the basis; according to the method, the health degree quantitative index directly corresponding to the structural damage state can be generated in real time, so that the reliability level of the component is dynamically and accurately evaluated, the early recognition and early warning capability of potential damage is remarkably improved, and reliable guarantee is provided for safe operation and service life management of large industrial equipment.

The invention discloses a structural reliability evaluation method based on clustering ultrasonic load interaction blind extraction. The method comprises the following steps: separating useless and useful data of a large-scale structure based on ultrasonic information; blind extraction of structure reliability correlation based on useful ultrasonic information; solving a structural reliability multi-index correlation coefficient; solving a clustering coefficient based on interactive blind extraction; and calculating a reliability numerical value of the large structural member. Separation of useless information and extraction of related reliability information in ultrasonic load spectrum information of the large structural component can be realized, clustering of similarity degree is carried out on the obtained extraction information, and then the reliability of the structure is determined in real time through the reliability clustering value, so that the damage condition of the to-be-analyzed structural component is accurately obtained, and the reliability of the to-be-analyzed structural component is improved. And the operation safety degree of a large-scale structure is effectively ensured.

The present application discloses a method, device, equipment and medium for analyzing the failure disaster chain of a gas transmission pipeline, which relates to the technical field of gas transmission pipeline analysis, and includes: when a failure disaster of a natural gas transmission pipeline occurs, all disaster events of the current natural gas transmission pipeline are used as nodes in the network, and the inducing factors between the disaster events are used as directed edges in the network, and a disaster chain network is constructed according to the nodes and directed edges; calculating topological structure parameters including in-degree, out-degree, total degree value, and clustering coefficient in the disaster chain network, and determining the evolution mechanism of the risk propagation of the failure disaster chain of the current natural gas transmission pipeline based on the topological structure parameters; calculating the risk degree of the failure disaster chain of the current natural gas transmission pipeline based on the disaster-causing rate, loss degree, and directed edge vulnerability of the disaster chain network; analyzing the current natural gas transmission pipeline according to the evolution mechanism of the risk propagation of the failure disaster chain of the current natural gas transmission pipeline and the risk degree of the failure disaster chain of the current natural gas transmission pipeline.

Embodiments of the present disclosure provide a keyword extraction method, comprising: first acquiring a text complex network corresponding to a to-be-processed text, the text complex network comprising nodes corresponding to a plurality of single characters in the to-be-processed text, weight values of connecting edges between the plurality of nodes, strengths of the plurality of nodes and in-out degrees of the plurality of nodes; then calculating a clustering coefficient of connections between the plurality of nodes based on the strengths of the plurality of nodes, the in-out degrees of the plurality of nodes and the weight values of the connecting edges in the text complex network; thereafter calculating a proportional mixing amount of the connections between the plurality of nodes based on the in-out degrees of the plurality of nodes, the in-out degrees of nodes connected by the plurality of nodes and the number of connecting edges in the text complex network; and calculating path lengths of the connections between the plurality of nodes based on the connecting edges between the plurality of nodes; and finally extracting keywords corresponding to the to-be-processed text based on the clustering coefficient of the connections between the plurality of nodes, the proportional mixing amount and the path lengths, thereby improving the accuracy and comprehensiveness of keyword extraction.

The present invention belongs to the field of knowledge management and application technology, and specifically relates to a reviewer recommendation method based on a graph neural network. The network of the present invention adds two networks to the output layer of the graph convolutional neural network, namely, learning networks for degree information and clustering coefficient information in the graph structure information. To accelerate learning convergence and make learning more in-depth, the present invention adds learnable parameters when performing the sum operation of the three loss functions. To enhance the graph neural network's understanding of network structure, the present invention adds network structure information of degree and clustering coefficient to the input layer of the entire NIGCN network.

The application discloses a kind of crack seepage prediction methods of combining yield criterion and network topology, belong to rock mass engineering construction field, including the following steps: topological analysis is carried out to crack network based on complex network theory, determine the degree centrality of crack network node and crack shear stress transmission coefficient;Based on Mohr-Coulomb yield criterion, establish shear stress equation, determine the criterion of crack network damage leakage according to yield criterion and seepage stress coupling relationship;Considering the degree distribution and clustering coefficient of crack network node, the damage evolution equation considering the importance of node and crack damage characteristics is established;Based on the established damage evolution equation, the percolation region of crack seepage damage evolution prediction is carried out.The application uses complex network index to identify the importance of crack network node, and carries out failure calculation on crack edge and node in combination with yield criterion and seepage-stress coupling relationship, which effectively improves the efficiency of crack network damage evolution and seepage prediction.

The application discloses a brain function network training effect evaluation method based on a two-level threshold strategy and belongs to the technical field of brain science and artificial intelligence auxiliary diagnosis. In view of the problem that a single threshold method in traditional task state brain function network analysis is easy to make core connection be submerged by noise and low in analysis precision, the application proposes a two-level threshold strategy: firstly, a key sub-network is extracted through a fixed threshold to eliminate weak connection noise; then a series of sparseness is set on the basis of the key sub-network to perform multi-scale binary processing, so that a binary brain network under different sparseness levels is obtained. On this basis, clustering coefficients, characteristic path lengths and other graph theory topological properties are calculated, and significant differences between an untrained group and a trained group are analyzed by statistics comparison, so that neural electrophysiological indexes related to visual motion training are identified. The application can effectively improve the precision and stability of topological property calculation, accurately capture a dynamic change process of a training-induced brain function network from high local clustering to a sparse and efficient architecture, and provide reliable neurophysiological evaluation basis for the fields of motor skill acquisition, virtual reality training and neural rehabilitation.

The present application belongs to the technical field of uniformity analysis, and particularly relates to a perovskite film composition uniformity analysis method and system, to solve the technical problem that traditional technology is difficult to standardize quality monitoring of composition uniformity of large-area perovskite film. The analysis method comprises: S1, collecting photoelectric response signals returned by each to-be-tested point; S2, constructing a graph model based on the spatial positions of each to-be-tested point and the corresponding harmonic entropy; S3, calculating the clustering coefficient of each node in the graph model, and multiplying the clustering coefficient of the node by the harmonic entropy corresponding to the node to obtain a perturbation vector; S4, calculating the kurtosis of the perturbation vector, and linearly combining the kurtosis with the reference parameter to obtain an index representing the composition uniformity of the perovskite film. The analysis method provides an objective and reliable evaluation basis for quality control and process optimization of perovskite film.

The invention discloses a data processing method and device, a nonvolatile storage medium and electronic equipment. The method comprises the following steps: acquiring a network topological graph; according to a first parameter and a second parameter, a neighbor aggregation index of the node is determined, and the first parameter represents a summation result of degree centrality of the node and degree centrality of all adjacent nodes of the node; the second parameter represents a product result of an attenuation function of the clustering coefficients of the nodes and a target parameter, and the target parameter is the sum of the clustering coefficients of all adjacent nodes of the nodes; fusing the neighborhood gathering index with the information entropy to obtain a neighborhood gathering entropy index of the node; and according to the neighbor gathering entropy index, determining an importance sequence of the nodes. According to the node importance evaluation method and device, the technical problem that the node importance evaluation precision is insufficient due to the fact that the related node importance evaluation method cannot reflect the influence of the adjacent nodes on the target node and the effect of the target node in the whole network is solved.

The invention belongs to the technical field of signal processing and analysis, and particularly relates to an electroencephalogram driving behavior recognition method based on coherent network characteristic dynamic evolution, which comprises the following steps of: 1, performing standard preprocessing on an original electroencephalogram signal, selecting a beta-wave frequency band as an analysis frequency band, calculating a coherent value under the analysis frequency band, and calculating a coherent value under the analysis frequency band; the synchronous intensity of the two electroencephalogram signals is measured to serve as a network connection weight, and a weighted coherent network is constructed; and 2, for each time point, extracting the degree and clustering coefficient of each node in the weighted coherent network, for each pair of nodes, calculating the instantaneous phase difference of each pair of nodes, and meanwhile, introducing the space geometric distance between two electrodes to obtain a behavior characteristic response index. And step 3, obtaining power spectrum parameters of all brain region signals of the target behavior under the analysis frequency band, comparing the target behavior identification value with the behavior threshold value, and judging whether the target behavior is activated or not. The accuracy, the real-time performance and the physiological interpretation performance of behavior recognition are remarkably improved.

The application discloses a kind of network equipment early warning evaluation method and system based on SNMP, belong to network equipment state monitoring early warning evaluation technical field, method includes: using SNMP protocol and equipment OID information acquisition network equipment's key state information and store to database;Real-time early warning is carried out to key state information based on early warning filter pool, and real-time early warning result is obtained;Current data, historical data are recalled and combined with real-time early warning result, and state evaluation is carried out using improved grey clustering algorithm, with early warning result as the determination reference of index abnormal degree and the basis for grey class correction, the state evaluation result associated with early warning result is output by calculating clustering coefficient and determining grey class.The application realizes the deep fusion of early warning and evaluation, improves the accuracy and timeliness of network equipment state evaluation, provides strong support for preventive maintenance, and can be widely applied to the operation and maintenance of various network equipment.

The invention discloses a power distribution area flexible capacity expansion regulation and control method based on a topological structure, and the method comprises the steps: firstly constructing a power grid topological graph with a power supply area as a node and a power transmission and distribution line as an edge, obtaining the attributes of the node and the edge, and calculating the normalization degree, the clustering coefficient and the betweenness centrality; combining the output power, the current, the voltage fluctuation, the power supply area and the population to obtain a first confidence coefficient and a first stability, and further forming a node importance index; the method comprises the following steps: periodically collecting various real-time power supply data, fusing node attributes and importance to construct first sample data, constructing label data according to historical outage times, outage frequency, outage duration and power supply availability, and training a capacity expansion strategy prediction model; and constructing a real-time sample by utilizing rated parameters of capacity expansion equipment and real-time power supply data during operation, inputting the real-time sample into a prediction model to obtain predicted power supply data of each transformer area, determining capacity expansion priority according to the predicted power supply data, and realizing graded flexible capacity expansion regulation and control of the power distribution transformer area.

The invention discloses a city open block evaluation analysis method and system, and the method comprises the steps: setting an initial open block influence region of a city, obtaining the traffic volume of each road in the influence region before and after the opening of the initial open block influence region, calculating the saturation degree of each road before and after the opening, determining an affected road according to the road saturation degree, and carrying out the evaluation analysis of the urban open block. Determining the influence range of the influence area of the open block; constructing an evaluation analysis index system; establishing an evaluation analysis method model: determining an index grey class level quantitative standard according to an index layer, calculating a whitening weight function characteristic value, and constructing a whitening weight function; weighting each index layer, and calculating a gray clustering coefficient and a clustering coefficient vector; and according to the obtained clustering coefficient vector, determining a grey class to which the block is opened and affiliated, and further outputting a result for evaluating whether the block is suitable for being opened or not. From qualitative and quantitative angles, the invention provides an effective and reliable evaluation analysis method for block opening.

This invention relates to the field of complex network technology. It provides a method for identifying key nodes in complex networks based on Shannon's theorem and the Matthew effect. The method includes: acquiring the topology of the complex network and modeling it as an undirected graph; calculating the structural features of each node; mapping the node's clustering coefficient and structural hole constraint to channel noise, and mapping the node degree and clustering coefficient to channel bandwidth; quantifying the node's information propagation capability as a relay using a signal-to-noise ratio formula; calculating the node's initial information based on its degree and normalized K-shell value; iteratively calculating to give nodes with high initial information a weight gain when aggregating neighbor contributions, simulating the cumulative process of network resources tilting towards core nodes, and obtaining a Matthew effect information aggregation value; obtaining the node's final influence score based on the Matthew effect aggregation value and information propagation capability; and sorting the nodes in descending order based on the final influence score to identify key nodes in the complex network.

The application discloses a bonding encapsulation equipment health state evaluation method based on twin data, classifies and combs three major functional components of an ultrasonic welding head system, a motion control system and a machine vision system of the bonding encapsulation equipment, establishes a hierarchical structure model and a fault tree, constructs a three-level health evaluation model including a component layer, a functional component layer and a whole machine layer, divides equipment health states into four qualitative grades of excellent, good, general and observation, introduces a health degree into quantitative characterization in combination with a fuzzy mathematics theory, adopts a multi-channel neural network model based on an attention mechanism to recognize the health state of a vibration signal, adopts a grey clustering model for health indexes which are not vibration type and are difficult to directly quantitatively evaluate, calculates clustering coefficients of the health indexes belonging to various health states based on a whitening weight function, and determines the health state; and adopts an entropy weight method based on twin data driving to determine the weight of each health index, carries out grey class weighted fusion, and outputs the real-time health degree and the health state grade of the equipment. The application improves the reliability, accuracy and equipment operation efficiency of the equipment health state evaluation, shortens the maintenance time, and reduces the production cost.

The invention discloses an industrial control network flow abnormity automatic monitoring method and equipment thereof, and the method comprises the steps: obtaining current session data after the internal network flow collected in a bypass mirroring manner is preprocessed, and constructing an IP connection relation graph according to the current session data; determining whether an IP connection relation graph constructed by the current session data reaches a steady state or not by using a multi-dimensional judgment rule; and according to the clustering coefficient corresponding to the determined steady-state IP connection relation graph, whether the constructed new IP connection relation graph exceeds a sixth preset threshold value is judged in real time so as to judge whether the network flow corresponding to the currently constructed new IP connection relation graph is abnormal or not. Therefore, on the premise that no network interference is carried out, modeling is carried out through the network and the IP node / IP set to carry out network state evaluation, so that the method has rich expandability and relatively high applicability; and the change of the network structure and the anomaly detection can be better captured.

The invention discloses an SG-Lasso-based depression (DP) automatic detection method, which is characterized in that a super-brain function network (HBFN) is constructed and optimized through electroencephalogram (EEG) signals, and efficient DP detection is realized. The method comprises the following steps: inputting a multi-channel EEG signal, constructing an HBFN by using SG-Lasso, establishing a mathematical model based on sparsity and stability, optimizing the HBFN at different frequency bands, extracting different types of clustering coefficients in the HBFN, and checking and evaluating differences among clustering coefficient groups through nonparametric arrangement at different frequency bands. And selecting clustering coefficients with significant differences among the groups as potential markers for DP detection, and finally verifying the effectiveness of the potential markers on DP automatic detection by adopting a classifier. According to the method, the HBFN is optimized in combination with sparsity and stability, the accuracy of DP detection is remarkably improved, and a new technical support is provided for clinical DP auxiliary diagnosis.

The present invention relates to the field of drone technology, and in particular to a method, device, medium, and equipment for determining candidate priorities of drone take-off and landing points based on a complex network model. The method obtains candidate priorities corresponding to initial drone take-off and landing points based on take-off and landing point weights, route weights, node betweenness, clustering coefficient, and preset initial weights, screens out M candidate drone take-off and landing points, measures differences between candidate take-off and landing point prediction scores of the M candidate drone take-off and landing points based on route anomaly, route length, and second traffic flow, obtains screening scores of the M candidate drone take-off and landing points to characterize screening accuracy, and serves as a basis for updating an initial weight set and candidate priorities. The M scores are used to screen out target drone take-off and landing points, and provide locations for drone take-off and landing facilities, thereby achieving a balanced distribution efficiency and distribution intensity of drones corresponding to each target drone take-off and landing point, thereby improving the overall work efficiency of the drones.

This invention belongs to the field of air pollution online public opinion analysis technology, and relates to an air pollution public opinion analysis method based on topic algorithms and network theory. The method mines air pollution public opinion data to obtain hot topics, uses keyword co-occurrence networks to abstract the relationship between topics and keywords, and calculates the degree distribution, clustering coefficient, node betweenness number, and network robustness of the network to obtain the connection between topics and keywords, as well as the distribution of key topics, providing a scientific basis for decision-making related to public opinion governance.