11 results about "Network science" patented technology

The invention relates to the field of data mining and machine learning, and relates to a bridge group deformation mode evaluation method and system based on PS-InSAR and complex network theories. The method comprises the following steps of: 1, performing refined preprocessing on high-dimensional time series data of engineering structure monitoring points obtained by multiple sources; 2, constructing a behavior pattern distribution vector of each monomer structure by adopting a UMAP-KDPI algorithm; and step 3, constructing a group association network by calculating behavior similarity between structures, and performing community discovery, key node identification and multi-level association characteristic deep analysis by using a network science theory to realize bridge group deformation mode evaluation. The problem that in an existing monitoring data analysis method for an engineering structure group, potential and common behavior patterns in complex high-dimensional time series data are difficult to effectively recognize, and an effective means for accurately quantifying and comparing behavior similarity among structure individuals is lacked is solved.

The invention provides a network science-based method for evaluating a herbicide neurodevelopmental toxicity risk. The herbicide which is widely used and is detected at high frequency in the environment is collected; acquiring a herbicide target and risk genes of neurodevelopment disorder related diseases; constructing a herbicide target network and a disease module; analyzing the neuro-developmental toxicity risk of the herbicide by network proximity; constructing an overlapping network of a herbicide target network and a disease module; screening core pathogenic genes based on multiple network topology algorithms and carrying out gene enrichment analysis; the toxicity risk is evaluated from the molecular level, and the high-risk herbicide variety is effectively discriminated. And a neuro-developmental toxic molecular mechanism of the herbicide is analyzed based on a network analysis system, so that a new framework from risk assessment to mechanism analysis is constructed for toxicity research of environmental chemicals.

The invention discloses a circRNA (Ribonucleic Acid) and miRNA (Micro Ribonucleic Acid) interaction prediction system and a circRNA and miRNA interaction prediction method of a graph Fourier pulse neural network. The method comprises the following steps: on the basis of high-throughput sequencing omics data of complex diseases, constructing a heterogeneous biological information network containing drugs, diseases, proteins, circRNA, miRNA and lncRNA; converting the topological features of the entities into a unified feature space by using a graph convolutional network; designing a pulse graph neural network in combination with Fourier coding and a pulse neural network, and extracting a topological structure and high-order semantic features in the network; fusing sequences, topologies and semantic features of circRNA and miRNA through a gate multilayer perceptron to obtain embedding features of circRNA and miRNA; and finally, the interaction of circRNA and miRNA is predicted by adopting a Bayesian classifier. According to the method, heterogeneous biological information is modeled from the perspective of network science, Fourier coding, spiking neurons and graph embedding learning are utilized, the action mechanism of circRNA and miRNA in complex diseases can be disclosed, and the method has good practicability and application prospects in the fields of artificial intelligence, life science, clinical medicine and the like.

The utility model discloses a network technology router convenient to install and fix, which comprises an installation plate, the lower surface of the installation plate is fixedly connected with a supporting frame, the upper surface of the supporting frame is provided with an elastic supporting assembly, the upper end of the elastic supporting assembly is fixedly connected with a top plate, and the upper surface of the top plate is movably connected with a bottom plate. The upper surface of the bottom plate is fixedly connected with a router body, the upper surface of the mounting plate is fixedly connected with four connecting shells, the four connecting shells are symmetrically arranged, the inner surfaces of the connecting shells are slidably connected with sliding blocks, the side surfaces of the connecting shells are in threaded connection with lead screws, and one ends of the lead screws are rotatably connected to the side surfaces of the sliding blocks; by means of the components, complex tool operation and tedious installation procedures are not needed, the installation efficiency can be greatly improved, and compared with a traditional mode of fixing through screws and the like, the installation method can complete installation within a short time.

The application discloses a cognitive value evaluation method of medical record data technology based on network science, and particularly relates to the technical field of cognitive value evaluation, and the method comprises the following steps: constructing a multi-scale medical data description set and establishing a medical cognitive relationship network, extracting a node structure stability coefficient and a knowledge propagation influence coefficient to form a node comprehensive stability index, combining a node centrality change to obtain a node cognitive contribution fluctuation index, and jointly constructing a cognitive value index with the node time scale weight, so that stable quantitative evaluation of the cognitive value of the medical record data is realized, and the accuracy and reliability of the cognitive value evaluation result in the multi-time scale medical data environment are improved.

The invention belongs to the technical field of network science and control, and discloses a multi-agent network hierarchical influence node identification method based on triangular structure recursive compression, which comprises the following steps: firstly, quickly detecting a triangular structure by using an intersection matrix to reduce the triangular counting complexity; evaluating nodes by using CPI comprehensive energy, neighbor change, connectivity and load; a conflict resolution and coverage maximization repulsive force algorithm is designed for a shared node / side triangle and a non-triangular area, so that influence on overlapping is avoided, and distribution is controlled in a balanced manner; top k control nodes are screened through virtual node compression recursion, the number of nodes is further reduced, and network robustness is improved. According to the method, the key nodes with the minimum number and the highest control efficiency can be accurately and efficiently identified, and the method is suitable for dynamic multi-agent networks such as unmanned aerial vehicle clusters and the Internet of Things.

The invention discloses a key node identification and price prediction method and system for a cryptocurrency market, and belongs to the technical field of cryptocurrency market analysis. The method comprises the following steps: firstly, acquiring historical price data of various cryptocurrency, and constructing a cryptocurrency price association network based on symbol correlation; secondly, performing community detection on the network by adopting an EDGly algorithm, and calculating and identifying key nodes in each community based on a Shapley value; and finally, acquiring historical price data and social media emotion data of the key nodes, inputting the historical price data and the social media emotion data into the trained Tuned BiLSTM-Sentient model, and outputting a future price prediction result. The system comprises a corresponding data input layer, a network construction module, a community and key node identification module and a price prediction module. Through organic integration of network science, game theory and deep learning, the problems that in the prior art, a market structure is not deeply depicted, key nodes are not completely recognized, and prediction precision is limited are solved, and a systematic solution is provided for market risk early warning and investment decision making.

The invention relates to the technical field of network science, and provides an ultra-large network data partitioning method based on an equalization cutting theory, which comprises the following steps of: mining a compact structure which is not shown through a connecting edge but actually exists in a network community by adopting a DP algorithm and simulating a dynamic process of a network, and generating a new adjacency matrix; constructing a new objective function for a clustering algorithm, and realizing intra-community path minimization and inter-community size balance based on the new objective function; introducing a Lasso model thought into a clustering process, and controlling the balance between the density in the communities and the density between the communities through the combination of a norm 1 and a norm 2; and gradually adjusting community division through alternate optimization and a Lagrange multiplier method until algorithm convergence, and generating a final community division result. The method provided by the invention is not only suitable for a heterogeneous network, but also especially suitable for a homogeneous network, in addition, the method can also be applied to the fields of distributed computing, data mining, artificial intelligence and the like, and has a wide application prospect.

The application relates to the field of complex systems and network science, and provides a network-type infrastructure system management method based on multi-dynamics coupling, which comprises the following steps: modeling an infrastructure system existing in a preset region in a current period and subjected to disturbance to obtain a network to be processed; under the action of LLRM and GLRM, a plurality of multi-dynamics mechanism coupling models corresponding to the network to be processed are built; dynamics simulation is performed on each multi-dynamics mechanism coupling model to obtain a cascading failure characteristic simulation result under each multi-dynamics mechanism coupling model; according to an evaluation target of the infrastructure system in the current period, the cascading failure characteristic simulation result is used to determine a target coupling model of the infrastructure system from the plurality of multi-dynamics mechanism coupling models, and the infrastructure system is managed based on a dynamics mechanism in the target coupling model. The application can realize effective management of the infrastructure system existing in the current period and subjected to disturbance.

The present application relates to a kind of anti-malware propagation methods based on network resistance, specifically includes: using network science theory to construct energy internet topology, and define the concept of network resistance;Based on the real-time feature analysis of each node neighbor node, identify the node attacked by virus malicious in network;By network topology formed by the rest of unqualified nodes network resistance maximization, the network is divided and detected, and the block of malicious software or virus propagation is realized.The present application method can be distinguished from the related research in the aspect of existing virus propagation dynamics, the network resistance concept is defined by using network science theory, the virus propagation risk is reduced by changing network structure, and the anti-malware propagation method is studied at the theoretical level.

The present invention discloses a method and system for identifying a dominant path in a multi-source scenario based on a perturbation response model, and relates to the field of computer technology. The present invention dynamically analyzes the impact of link state changes on the steady state of the system based on a perturbation response method, rather than relying on static indicators, and is closer to the real interaction mechanism of the network, thereby improving the accuracy of mechanism modeling. By traversing the link state, the actual contribution of each link to the target node response is quantified, avoiding ignoring the complex interactions of weak connections and multi-link collaboration. Through the process of the system re-evolving to a steady state after the disturbance, the traffic distribution and competition at the shared node are naturally modeled, rather than relying on local optimal addressing, thereby improving the global rationality of dominant path identification and solving the problem of multi-source target resource competition. The method of the present invention is applicable to all types of networks, and plays a core supporting role in constructing a communication risk warning model based on network science, implementing key node intervention strategies, and optimizing the intelligent governance mechanism of social information systems.