20 results about "Co-occurrence networks" patented technology

This invention discloses a dynamic quantile filtering method for product co-occurrence networks, comprising: firstly, constructing a product co-occurrence network based on the co-occurrence relationships of products in orders; then, employing a dynamic quantile filtering strategy to denoise the network, specifically: obtaining the co-occurrence frequency weights of all product pairs and forming a co-occurrence matrix; extracting all non-zero co-occurrence frequency weights for each node; defining a local adaptive threshold based on the quantiles of these weights; retaining edges greater than or equal to the threshold; resetting the weights of edges less than the threshold to zero; and retaining isolated nodes; symmetricizing the adjacency matrix using an intersection strategy to obtain a weighted adjacency matrix; updating the network to obtain the denoised product co-occurrence network; and finally, using the Louvain algorithm to perform community partitioning on the denoised network to obtain a product community network. This invention can adaptively filter out weak connections and noisy edges in the product co-occurrence network, effectively improving the accuracy and robustness of community detection.

The invention discloses a carbon reduction and pollution reduction measure collaboration evaluation method, system and equipment based on text mining and network analysis and a medium, and belongs to the technical field of climate and environmental governance measure analysis and evaluation.The method comprises the steps that carbon reduction and pollution reduction related measure files are obtained, and a personalized external corpus is constructed; performing duplicate removal and word segmentation on the text data; initially selecting keywords based on a TF-IDF method, and checking and summarizing the keywords according to a keyword three-level classification system; calculating the co-occurrence frequency between every two keywords, and constructing a word co-occurrence network; and analyzing the structure characteristics of the co-word network from different levels by using indexes such as network average degree, edge density, average shortest path and the like, and summarizing the collaboration of carbon and pollution reduction measures. Based on real and reliable first-hand data and a more targeted personalized corpus, keywords are accurately extracted and classified for extraction and classification, and in combination with a network analysis method, internal collaboration of carbon and pollution reduction measures is analyzed.

The invention discloses a technical field subversive potential prediction method and device and a program product, and the method comprises the steps: firstly obtaining patent data of a target technical field, and constructing an IPC classification set for the patent data in each preset time window; constructing a patent technology classification bipartite graph according to the association relationship between the patents in the classification set and the IPC classification numbers; based on the co-occurrence strength of any two technology classification nodes in the bipartite graph, generating a technology classification co-occurrence network; performing community division on technical classification nodes in the technical classification co-occurrence network; performing cross-time window community similarity calculation on community division results in different time windows, and constructing a community evolution event list; and finally, obtaining a subversive potential index of the target technical field according to the community evolution event list, and obtaining the subversive potential index of the target technical field based on the subversive potential index. According to the embodiment of the invention, the accuracy of subversive potential prediction in the technical field is effectively improved.

This invention discloses a method and system for intelligent construction and dynamic optimization of supply chain data. It constructs an ontology based on OWL and defines a semantic space. It collects text and image data of business scope, performs layout detection and OCR to extract text from images, and encodes the text into visual embeddings. It then fuses TF-IDF vectors, semantic vectors, and visual embeddings to generate a feature matrix. A rule-based expression tree with AND / OR nodes is constructed to evaluate the feature matrix, outputting candidate companies and matching degrees. The expression tree is made differentiable and gated; within a sliding window, a loss function is constructed based on target number deviation and feedback from missed detections, and gradient descent is used to update keyword weights and thresholds. A relationship graph is constructed based on candidate companies to calculate PageRank centrality, and this is fused with matching degree and time decay to output a weighted ranking. Double buffering updates are used, and the keyword system is expanded with term co-occurrence networks, Louvain, and semantic drift detection to output a traceable evidence chain.

The invention provides a big data-based conference information processing method and system. The method comprises the following steps of: obtaining a to-be-processed conference document and a conference document corpus to which the to-be-processed conference document belongs; calculating the position of the lexical item and the regional weighted word frequency; calculating intra-class inverse document frequency of the lexical item in the class cluster to which the lexical item belongs based on a division result of the document theme class cluster; obtaining an initial weight of the lexical item based on a product of the position and region weighted word frequency and the intra-class inverse document frequency; screening out core lexical items of which the weight is greater than a first preset threshold value; constructing a weighted co-occurrence network graph of the core lexical items; and taking the initial weight of each core lexical item as the initial weight in the weighted co-occurrence network graph, iteratively updating the weight of each core lexical item node through a preset graph sorting iterative algorithm according to the connection relationship of the nodes in the graph and the weight of the edge, and when the weight of each node converges or reaches a preset number of iterations, carrying out iterative updating on the weight of each core lexical item node. And taking the obtained weight as the weight of the core lexical item.

The invention belongs to the technical field of natural language processing, particularly relates to a text mining-based air-taking service case semantic extraction and network construction method, and aims to solve the problem of how to intelligently extract effective information of air-taking service cases. The method comprises the following steps: acquiring an air-ride service case text, and preprocessing the text to form case text data; analyzing a syntactic structure relationship among words in the case text data by utilizing a HanLP algorithm, and screening out a semantic core structure relationship; taking part-of-speech and word frequency information into consideration, improving a feature extraction algorithm, and extracting keyword items; based on adjacent positions of words in the feature data, analyzing to obtain directed association features among the words, and obtaining directed association phrases; constructing a co-occurrence network based on the directed association features and the directed association phrases, and performing modular clustering analysis; and key node links and high-influence node groups are extracted from the co-occurrence network, so that efficient and accurate extraction of service case work key points and core elements is realized.

The invention discloses a phytoplankton community stability evaluation method based on cascade reservoir influence. The method comprises the following steps: step 1, river reach division and sample collection; step 2, community structure and diversity analysis; step 3, analyzing a community construction mechanism; step 4, co-occurrence network construction and stability evaluation; step 5, environment driving factor modeling; and step 6, performing multi-group comparative analysis. According to the method, multi-dimensional analysis of a community structure, a construction mechanism, network stability and environment driving factors is integrated, and a complete evaluation chain is formed; a neutral community model and a co-occurrence network analysis method are introduced, and the influence mechanism of the cascade reservoir is revealed from the angles of the ecological process and interaction; the method is suitable for river and reservoir systems of different scales and different types, and is especially suitable for ecological assessment of areas without data or with less data; the method has important theoretical and practical values for scientifically guiding ecological scheduling and biodiversity protection of the drainage basin.

The application discloses a construction method and a system of a classification co-occurrence network and a technical opportunity identification method, and belongs to the field of technical opportunity identification. Semantic information and co-occurrence information of a joint patent classification (CPC) are combined to form a CPC co-occurrence network, hidden connection modes between CPC nodes are mined through a graph neural network model, then CPC nodes that are likely to be connected with an artificial node representing a target field are predicted, and finally technical development points that are likely to appear in the future in the target field are identified, so that certain decision support is provided for the formulation of a technical development strategy. The application can fully combine the semantic information and the co-occurrence information of the CPC to mine potential correlations between technologies, rich node features provide a good basis for learning of the model, and the technical opportunity identification can be better performed.

The application belongs to the technical field of equipment information resources, and particularly relates to an information resource development and utilization method and device for an aircraft design stage. An information resource structure is formed by establishing a high-frequency word co-occurrence network; and a key intelligence target network is formed by open source data layering. Intelligence collection is performed on key targets and stored in a target database; a knowledge representation method is selected to perform cross-modal semantic association on the knowledge in the target database, and an information resource system is established; a core literature and technology theme identification method based on semantic analysis is designed to identify high-value literature and disruptive technology themes; and resource development and utilization intensity is determined to form a multi-dimensional information resource system. The application realizes information resource development and utilization for each stage of aircraft development, and converts disordered massive information into resource themes and products that are strongly related to user roles and task stages. Key targets can be quickly locked, information screening efficiency is improved, and the engineering properties of information resources are revealed.

The invention relates to the technical field of environmental monitoring and ecological environmental protection, and discloses a reclaimed water river ecological health evaluation method based on environmental DNA and a co-occurrence network, and the method comprises the steps: setting sampling points, and collecting a water sample of a monitoring region; extracting environment DNA (Deoxyribonucleic Acid); carrying out PCR (Polymerase Chain Reaction) amplification and high-throughput sequencing, and carrying out species annotation on the three biological groups; calculating a water quality comprehensive pollution index, and determining a reference point; constructing a co-occurrence network of various group organisms in each point location and calculating network topology parameters; establishing candidate index libraries of four dimensions of community structures, diversity, functional groups and network topology parameters for the three biological groups; calculating the integrity index of each class group after optimization and the multi-species integrity index after optimization; and delimiting the river health level according to the score of the integrity evaluation system. Therefore, the topological parameters of the ecological co-occurrence networks of various groups are included in integrity index calculation, objective weights are determined through machine learning model training, and the structure and function states of a river ecosystem can be scientifically and comprehensively reflected.

The application discloses a kind of phytoplankton community stability evaluation methods based on cascade reservoir influence, the method includes the following steps: step 1, river section division and sample collection;Step 2, community structure and diversity analysis;Step 3, community building mechanism analysis;Step 4, co-occurrence network construction and stability evaluation;Step 5, environmental driving factor modeling;Step 6, multi-group comparative analysis.The method of the application integrates the multidimensional analysis of community structure, building mechanism, network stability and environmental driving factors, forms a complete evaluation chain;Introduce neutral community model and co-occurrence network analysis method, reveal the influence mechanism of cascade reservoir from the angle of ecological process and interaction;It is suitable for rivers and reservoir systems of different scales and different types, especially suitable for ecological evaluation in areas with little or no data;It has important theoretical and practical value for scientific guidance of watershed ecological regulation and biodiversity protection.

The application provides a power grid safety regulation semantic retrieval method and system, which comprises the following steps: obtaining a power grid safety regulation corpus, constructing a co-occurrence network of core technical terms, and calculating a term relationship correction matrix; after a user inputs a retrieval query, converting the query and regulation clauses into general / field semantic vectors by using general and power grid field pre-training models respectively; fusing the two types of semantic vectors to calculate the initial semantic similarity of the retrieval query and each regulation clause; if the number of core technical term pairs of the query and the clause exceeds a threshold value, calculating the term association correction score by using the term relationship correction matrix; weighting and summing the initial similarity and the correction score to obtain the final relevance score, and returning the regulation clauses in descending order of the score.

The application discloses a pipeline accident chain determination method, device, medium and product, relates to the oil and gas pipeline technical field, and aims to solve the problems of low efficiency and inaccurate analysis results of general technology in the aspect of oil and gas pipeline accident analysis. The pipeline accident chain determination method comprises the following steps: firstly, obtaining an accident text to be processed, wherein the accident text to be processed comprises text content of a pipeline accident. Then, a first accident element in the accident text to be processed and a co-occurrence frequency of at least two first accident elements are determined based on a graph neural network model. Then, a target co-occurrence network of the pipeline accident is constructed according to the first accident element and the co-occurrence frequency of the accident text to be processed. The target co-occurrence network is used to represent the co-occurrence relationship between the first accident elements, and the co-occurrence relationship is positively correlated with the co-occurrence frequency. Then, an accident chain of the pipeline accident is determined according to the target co-occurrence network.

This information analysis system comprises: a collection unit that collects a plurality of pieces of document data; a reception unit that receives an input of a keyword; an extraction unit that extracts text data including the keyword from the document data on the basis of the keyword; an analysis unit that analyzes the text data and weights each word and each pair of words included in the text data; and an output unit that, on the basis of the weighting, outputs analysis results including a co-occurrence network in which the number of occurrences of each word is represented by the size of a circle and the number of occurrences of each pair of words is represented by the thickness of a line between the words.

The application provides a conference information processing method and system based on big data, acquires a conference document to be processed and a conference document corpus to which the conference document belongs, calculates position and area weighted term frequency of a term, calculates within-class inverse document frequency of the term in a belonging class cluster based on a division result of a document theme class cluster, obtains initial weight of the term based on a product of the position and area weighted term frequency and the within-class inverse document frequency, screens out core terms with weight greater than a first preset threshold, constructs a weighted co-occurrence network graph of the core terms, takes the initial weight of each core term as initial weight in the weighted co-occurrence network graph, iteratively updates weight of each core term node according to connection relationship of nodes and weight of edges in the graph through a preset graph sorting iteration algorithm, and takes the obtained weight as weight of the core term when weight of each node converges or reaches a preset iteration number.

The application discloses a kind of based on multiple label co-occurrence network discrimination electric larceny type detection method and system, comprising: obtaining historical electric larceny sample data;User electricity process characteristics are mined using GRU network modeling, and the hidden state representation of current time step is output as user electricity process evolution characteristics;Based on the historical electric larceny sample data, establish electric larceny label co-occurrence relationship network graph;Based on the electric larceny label co-occurrence relationship network graph, the associated structure information between electricity features is mined, and the relationship characteristics between electric larceny labels are obtained;Determine the attention score weighting of user electricity process evolution characteristics, the relationship characteristics between electric larceny labels and user attribute characteristics, and after linear network layer processing, obtain the comprehensive characteristics of user electric larceny, and based on softmax activation function the comprehensive characteristics of user electric larceny are converted into probability value by output, to determine electric larceny behavior type based on probability value.

The invention discloses a method and a system for identifying generality technology opportunities, and belongs to the technical field of data processing. Based on patent data of different stages, IPC main group information is extracted, and a time-divided technology co-occurrence network is constructed to reflect an evolution relationship between technologies; node features are extracted from three dimensions of semantics, number and structure, a GraphSAGE model is adopted for node representation learning, and an embedded vector containing multi-dimensional information is obtained; taking the node pair embedded representation after Hadamard product processing as an edge feature, respectively training a plurality of classification models for link prediction, and selecting an optimal model for identifying a common technology combination which is not co-occurred yet but has fusion potential in a co-occurrence network, so as to realize common technology opportunity identification at a microscopic level; and on the basis of a prediction result, by combining a maximum group detection and overlapping community discovery algorithm, a generality technology cluster with strong fusion and high cooperation potential is identified, and generality technology opportunity identification at a macroscopic level is realized.

The invention provides a power grid safety regulation semantic retrieval method and system, and the method comprises the steps: building a co-occurrence network of core technology terms through obtaining a power grid safety regulation corpus, and calculating a term relation correction matrix; after a user inputs a retrieval query, respectively converting query and regulation terms into universal / domain semantic vectors by using a universal pre-training model and a power grid domain pre-training model; fusing the two types of semantic vectors, and calculating initial semantic similarity between retrieval query and each regulation term; if the quantity of the core technical term pairs of the query clauses exceeds a threshold value, performing summation by utilizing a term relationship correction matrix to obtain term association correction scores; and performing weighted summation on the initial similarity and the correction score to obtain a final relevancy score, and returning regulation terms in a descending order according to the score.

The invention discloses a ceramic color design system based on data visualization and an application method thereof, and the method comprises the following steps: preprocessing ceramic image data through a computer vision technology, and constructing a ceramic-color bipartite graph network connecting a ceramic work and a color as a basic model; the bipartite graph network is projected into a weighted color co-occurrence network and is used for revealing and quantifying a matching rule between colors; a user can dynamically adjust a CIEDE2000 color difference threshold through a visual interaction interface to simplify the co-occurrence network, and center analysis is combined, so that a core design color plate is extracted from a network topology structure; the method also supports generation of visual display plates such as an age-color-classification-color system-style five-level color flow mulberry-based graph so as to assist in analysis of macroscopic association of colors and cultural attributes. The invention is mainly used for providing a data-driven innovative tool capable of systematically mining traditional color matching genes for modern design.

The application relates to a method for identifying technical opportunities by using classification number co-occurrence and link prediction, comprising the following steps: dividing patents in a certain technical field or application field into three periods, extracting classification number information of a patent set in each period, determining a classification number co-occurrence relationship, and constructing a co-occurrence network with classification numbers as nodes according to the co-occurrence relationship; extracting three kinds of features of a node pair in the co-occurrence network in the three periods, namely, a similarity index, a node attribute related to the generation of a technology link and a node pair relationship index, and splicing them into a feature vector; performing node embedding by using graph representation learning based on a graph convolutional neural network (GCN), splicing the similarity index, the node pair relationship index and the GCN-based node attribute index to form a feature vector for the next step of modeling; and constructing and training a machine learning model based on the feature index and the node connection state of the node pair in the co-occurrence network to obtain a link prediction model.