33 results about "Co-occurrence networks" patented technology

The invention relates to the technical field of social networks, in particular to a micro-blog hot word and hot topic mining system and method. The method includes the following steps that content data released in a micro-blog are preprocessed to acquire a candidate hot word sequence; according to the frequency of occurrence and suddenness of candidate hot words in a candidate hot word set at the current moment and in a given historical time window, the vitality of each candidate hot word is worked out, and a hot word set is formed by screening the candidate hot words; according to the hot word set formed by screening the candidate hot words, hot word correlation is worked out, and a hot word co-occurrence network is constructed; according to the hot word co-occurrence network, the hot word set is partitioned through the hot word clustering algorithm based on multi-label propagation to acquire a hot topic set. By means of the micro-blog hot word and hot topic mining system and method, efficient micro-blog hot word and hot topic mining is achieved, and mining precision and processing efficiency are improved.

The invention belongs to the technical field of service computation, and relates to a service and tag recommendation method based on random walk. The method comprises a service recommendation step and a tag recommendation step, wherein the service recommendation step comprises the following process of: collecting the information of a Mashup service and an API (Application Program Interface) service, abstracting the relation between the Mashup service and the API service by association, sorting each tag after performing random walk on a co-occurrence network between the tags, and recommending the Mashup service through the association relation between the tag and the Mashup service and the Mashup semantic expansion method. The tag recommendation method comprises the following process of: searching the nearest neighbor according to the similarity of the API service description texts, and performing tag recommendation of the API service through the random walk and tag sorting method. The recommendation method can be used for improving the accuracy of the service searching result, and solving the problem of tag randomness.

The present application relates to a recommendation method, system and electronic device for learning based on network representation. The method comprises the following steps: step a: constructing auser-article co-occurrence network based on a bipartite graph network and a single projection image; step b: For the user-article co-occurrence network, defining search strategy to get neighbor nodesof each user node and item node. C, according to each user node, article node and respective neighbor node, obtaining vector representation of each user node and article node by using representation learning on network; Step d: according to the vector representation of each user node and the article node, obtaining the most relevant article node of each user node through vector calculation, and recommending the most relevant article to each user according to the calculation result. The present application alleviates the problem of sparsity of collaborative filtering, makes the recommendation system more interpretable, and greatly alleviates the problem of scalability in collaborative filtering.

The invention discloses a scientific literature key content potential association mining method based on a graph neural network, and the method comprises the following steps: S1, obtaining scientificliterature data related to a certain specific event, and carrying out the data cleaning and preprocessing; s2, extracting a literature content keyword by utilizing a TF-IDF method; s3, constructing aword co-occurrence network for the extracted keywords and references to which the keywords belong by taking sentences as units; s4, learning vector representation of the keywords by using a graph convolutional neural network; and S5, obtaining the relevancy between different keywords by utilizing a similarity calculation function, and mining the potential incidence relation of the different keywords. According to the method, modeling is carried out on the keyword relationship extracted from the article content, and the potential association of the main keywords of the literature is mined by utilizing the graph convolutional neural network technology, so that the analysis requirement on the scientific literature content is met, and the correlation of the scientific literature in different fields is analyzed; and an effective method is provided for systematic analysis of scientific literatures.

The invention discloses a student risk early warning model building technology based on a social network, and the technology comprises the steps: firstly obtaining the behavior and psychological dataof a student, and carrying out the data preprocessing and feature extraction; carrying out feature selection and division on the obtained data, carrying out pre-training on two thirds of the data, andestablishing a co-occurrence network of students; carrying out feature fusion after the co-occurrence network is obtained, and obtaining input of a student risk early warning model; and finally, fusing the co-occurrence network and the time recurrent neural network, and fusing an Attention mechanism to perform student risk prediction. According to the invention, student risk prediction is achieved, early warning information is sent to college teachers, college academic achievement early warning, study and life guidance and advice, psychological counseling and other services are provided for colleges, and teachers can be helped to better understand students.

The invention relates to the field of artificial intelligence, and provides a behavior prediction method and device based on behavior co-occurrence network, equipment and a medium, which can obtain behavior information of a target user, and divide behaviors of the target user into at least one behavior segment according to occurrence time of each behavior so as to distinguish the behaviors of theuser. Different action subjects are effectively decoupled, behavior information of a user is better learned, a target behavior co-occurrence network is constructed based on behavior segments and inputinto a pre-constructed behavior prediction model, a prediction result is output, and the behavior prediction model is obtained based on Graph Pooling and co-occurrence network training. The behaviorco-occurrence relation is constructed through the divided behavior segments, the target behavior co-occurrence network is constructed, the distinction degree between the behavior segments is enhanced,model prediction is more accurate, and then behavior prediction of the user is achieved based on an artificial intelligence means. The invention also relates to a blockchain technology, and the behavior prediction model and the prediction result can be stored in the blockchain.

The embodiment of the invention discloses an information processing method and device and a storage medium. The method comprises the steps of performing word segmentation processing on a first corpusto obtain a first word set of the first corpus; wherein the first word set comprises at least two words; determining the relevancy between any two words in the first word set; utilizing the determinedrelevancy between any two words in the first word set to construct a first word co-occurrence network; wherein the first word co-occurrence network represents an association relationship between words in the first word set; determining a first feature matrix by utilizing the data of the first word co-occurrence network and combining a graph convolutional neural network (GCN) model; wherein the first feature matrix is a feature matrix corresponding to each node in a first word co-occurrence network; and performing dimension reduction processing on the first feature matrix to obtain a word embedding initialization result. In this way, a reasonable word embedding initialization result can be provided, and therefore the effect of shortening the word embedding training period is achieved through the word embedding initialization result.

The invention discloses a pipe gallery fault analysis method based on keyword co-occurrence. The pipe gallery fault analysis method comprises the following steps that 1, constructing a self-defined dictionary which is based on a general dictionary and covers special nouns and special terms of the pipe gallery industry; 2, after the fault information is obtained, carrying out fault information preprocessing on the fault information; 3, after a word set is obtained, conducting keyword selection on the word set; and step 4, after carrying out word cloud visualization on the keywords, forming a keyword co-occurrence matrix and visualizing the keyword co-occurrence matrix. According to the method, fault keywords are found by means of self-defined dictionaries and word frequency statistics, andare displayed through keyword word cloud visualization; furthermore, a keyword co-occurrence network is constructed, and layout optimization is carried out by adopting a graph layout algorithm, so that analysis and visual display of fault information are achieved.

The invention provides a method for mining relevance of species in micro-organisms, wherein the method comprises the steps of firstly, performing high-throughput sequencing on the DNA of the micro-organisms, calculating the sequencing data for obtaining species composition and abundance distribution; then, enlarging an inter-species relevance candidate range through a Loosen Definition method; andmining an indirect relation, a linear relation and a non-linear relation. Compared with the prior art, the method has advantages of well reproducing a micro-organism network for facilitating researching of a micro-organism co-occurrence network in a system level, and making a preparation for finding unknown relevance between key species.

The invention relates to construction of a hotspot mining system under a Hadoop framework, which comprises the following specific steps of acquiring data information A from a network by using a cloudcomputer Hadoop cluster module and preprocessing the data information A to obtain preprocessed data information B and sending the preprocessed data information B to the mining system; performing wordsegmentation on the preprocessed data information B to obtain a keyword set C; screening each keyword D in the keyword set C through a previous hotspot information word bank; sorting the keywords D from high to low, screening out hot words E, and constructing a hot word set F; constructing a word co-occurrence network according to the correlation among the hot words D in the hot word set F; and dividing the hot word set F by adopting a clustering algorithm according to the word co-occurrence network to obtain a hot topic set. According to the method, the hot topic set can be quickly obtained,and the accuracy of obtaining the hot topics from the network can be improved.

The invention discloses an improved TextRank keyword extraction method and device. The method comprises the following steps: constructing a word co-occurrence network of a text, and then based on the word co-occurrence network, introducing two complex network statistical characteristics of degree centrality and clustering coefficient of a node to obtain an initial weight of the node; according to the importance degree of the adjacent nodes to the nodes, allocating the initial weight to the connecting edge between the two nodes, and determining the weight of the connecting edge, so the weighting of the connecting edge is realized, and the importance score of each node is determined; introducing a position coefficient to adjust the importance score of the node, and determining the final weight of each node; and finally, sequencing the nodes according to the final weight of each node, and determining keywords of the text. According to the method and the device, the two features of the degree centrality and the clustering coefficient of the node are used for edge connection weighting, and the keyword extraction of the text is realized in combination with the position feature of the node, so that the keyword extraction accuracy can be effectively improved.

The invention discloses a theme identification method based on a theme co-occurrence network and external knowledge, and the method specifically comprises the steps: constructing a theme co-occurrence network based on annotation data: detecting theme sub-words in an existing domain knowledge text with theme annotation, and constructing the theme co-occurrence network according to the theme sub-words; constructing a switch module fusing external knowledge and a topic co-occurrence network: performing information richness sorting on the domain knowledge text with the topic annotation by using the co-occurrence network, and combining the sorting with the external knowledge to form the switch module; improving self-training of a domain knowledge text theme recognition model by introducing a switch module: training the domain knowledge text theme recognition model by using a self-training method, using information of the domain knowledge text without theme annotation as far as possible, and preventing generalization performance reduction caused by non-selective learning of the domain knowledge text without theme annotation by self-training; limited corpus information is fully and efficiently utilized, and the performance of a domain knowledge text theme recognition model is improved.

The invention provides a Yellow River basin evolution analysis method based on text mining. The Yellow River basin evolution analysis method comprises the steps of S1, obtaining an analysis sample corresponding to Yellow River basin evolution; s2, constructing a corresponding knowledge element co-occurrence network; s3, constructing a quotation coupling network, and constructing a knowledge element fusion network based on the quotation coupling network and the knowledge element co-occurrence network; s4, analyzing the constructed target knowledge element network corresponding to different time slices to obtain an evolution analysis path of the Yellow River basin; s5, obtaining a corresponding evolution analysis result based on the evolution analysis path and a co-occurrence cluster of each target knowledge element and each core knowledge element in the target knowledge element network; the method is used for meeting the analysis accuracy, instantaneity and intersectionality of literatures in each field in the evolution analysis process of the Yellow River field, rapidly and accurately tracking the development trend of the Yellow River basin in each field, and mastering the evolution rule of related knowledge, so that technical support is better provided for applying measures to the Yellow River basin according to local conditions and classified measures.

The invention discloses a keyword extraction method based on fusion of a network high-order structure and a topic model. The keyword extraction method comprises the following steps of 1, performing word segmentation on a news text D; 2, removing stop words from a word segmentation result to generate a word sequence; 3, constructing a word co-occurrence network G based on the word sequence; 4, endowing the connection edge of the word co-occurrence network G with a weight based on a network high-order structure to obtain a weighted adjacency matrix M; 5, calculating the topic expression capability of the words in the word co-occurrence network G under the target text; and 6, calculating final importance scores of words in the word co-occurrence network G based on the weighted adjacency matrix M obtained in the step 4 and the topic expression capability obtained in the step 5, and selecting the first k words as keywords of the news text D from large to small according to the final importance scores. According to the keyword extraction method implemented by the invention, on one hand, the calculation complexity is low; on the other hand, the topics of the words are fused, and the accuracy of news text keyword extraction is improved.

The invention discloses a method for constructing an accounting term co-occurrence network diagram, which comprises the following steps of: extracting semantic primitives of an accounting field, namely constructing a directed network diagram for vocabularies in an accounting dictionary, extracting the semantic primitives and describing field knowledge by utilizing an improved PageRank algorithm, and then combining based on synonym forest to finally obtain a candidate set of the semantic primitives of the accounting terms. The semantic primitive extraction method based on the graph theory is designed for the corpus of the accounting dictionary by utilizing the characteristics of knowledge in the accounting field. The accounting dictionary serves as an important professional corpus and an authoritative specification text in the accounting field, and systematically and comprehensively covers related terms and definitions thereof in the accounting field. If a computer can 'read' an accounting text by means of the semantic primitives extracted from the accounting dictionary, a large amount of information in the accounting field can be effectively utilized, so that term research based on the accounting dictionary effectively breaks through subjective analysis and small sample data limitation in semantic primitive extraction.