32 results about "Cluster graph" patented technology

A method of compressing a graph representation, and method of solving a problem represented by a graph using graph compression enable improved data searching techniques. The graph representation includes a plurality of vertices and a plurality of edges. A first clustered graph is generated by clustering the plurality of vertices of the graph representation. Each cluster of vertices is then replaced by a clustered vertex, wherein the clustered vertex inherits edges from the vertices of the cluster. Superfluous edges of the first clustered graph are then removed. A traversal probability for each of the removed edges is determined, and one or more of the edges that have been removed is selected for embedding in the first clustered graph, based upon a traversal probability for each of the edges. Information relating to the one or more edges is then embedded into the first clustered graph.

A method of detecting network worms include the following steps: (1) Profiling the TCP connection information collected from the protected network, quantifying the plurality of statuses contained in the TCP connection information; (2) Clustering the connection profiles to discover all the anomaly clusters that are specified by the condition composing of several adaptive thresholds; (3) Correlating the anomaly clusters to result in a new cluster graph or to extend an existing cluster graph; (4) Issuing a security incident about the worm propagation according to the propagation condition that also composes of several adaptive thresholds; and (5) Keeping and maintaining the status of the cluster graphs.

An approach for providing positioning services using multi-space clustering includes detecting sensor data across a plurality of sensor spaces for a device. The plurality of sensor spaces are associated with respective one or more sensor types. The approach also includes predicting a point propensity for the device to represent location data of the device based on the sensor data, and performing a clustering of the location data across the plurality of sensor spaces to generate a cluster graph of a physical space associated with the sensor data.

The invention discloses a semantic location privacy protection method based on an edge cluster graph, which includes the following steps: step 1, the user sends the current location, user privacy preference, and type of requested point of interest to an anonymous server; step 2, the anonymous server executes an anonymous algorithm to obtain Anonymous set; step 3, the anonymous server sends an anonymous request to the location server; step 4, the location server performs a query, and returns the query result to the anonymous server; step 5, the anonymous server refines the result, and returns the result to the user. The invention has the advantages of improving the success rate of the anonymous server in protecting the anonymous request, reducing the anonymous time, improving the query efficiency of the location server, and the like.

The invention discloses a graph theory-based optical network routing and wavelength allocation method, system and storage medium, wherein the method includes the following steps: establishing a cluster graph model according to the optical network; performing sparse processing on the cluster graph model; Initially partition and color the processed cluster graph to obtain an initial solution; improve the initial solution to obtain an optimal solution; output the optimal solution, generate a network construction plan according to the optimal solution, and complete the construction of the optical network. The scheme of the present invention utilizes graph division and coloring technology to model the optical network routing and wavelength allocation problems, forming a one-to-one correspondence, and then designs the routing and wavelength allocation scheme according to the graph division and coloring algorithm. Compared with the prior art, this invention The scheme is also efficient for larger networks, not only can guarantee the quality of the solution, but also can ensure the efficiency, thereby effectively saving the use of wavelength resources, providing a reasonable network construction scheme, and can be widely used in the field of optical network transmission technology.

The present application discloses an information processing method and electronic equipment. The method includes: obtaining preset first data, and extracting at least one entity word from the first data according to the vocabulary weight defined by the preset extraction rules; Associated words associated with words, establish association edges between entity words and associated words to construct a co-occurrence graph corresponding to the first data, wherein entity words, associated words, and associated edges have their own weights; based on preset clustering It is stipulated that the co-occurrence graph is clustered to form at least one graph cluster, and the graph cluster includes at least one entity word and the corresponding associated word; based on the meaning of the entity word and / or associated word in the graph cluster, determine the graph cluster Corresponding expressive intent to determine the expressive intent of the first data. The method can comprehensively summarize the actual intention of the first data from various angles for each entity word; realize the intelligent analysis of the first data, so that the current data can be processed accurately and effectively.

The invention provides a parallel gene splicing algorithm based on a cluster map structure. According to the parallel gene splicing algorithm based on the cluster map structure, a long sequence (scaffold) obtained by the splicing of a plurality of other gene splicing algorithms and a short-read long gene sequence (read-pair) generated by a double-end sequencer are used as input, and complementary scaffold are spliced to be the longer sequence by building index, mapping read-pair and scaffold clusters, building cluster map, searching path and other steps. The two steps of building index and reading length mapping aim at obtaining correlations and matching degrees of long sequences scaffold obtained by different algorithms by reading length, and clustering according to the correlations and matching degrees; all scaffold in the cluster are complementary and are potential splicing sequences. At last the cluster map is built to solve the overall longest path of the map, and thus the spliced long gene sequence is obtained.

The invention discloses a network security anomaly detection algorithm based on a cluster graph neural network. The algorithm includes the following steps: using a graph model to describe the network topology, using a graph neural network convolutional layer to optimize node features, using a graph clustering algorithm to divide the graph into multiple disjoint subgraphs, and treating each subgraph as a node, The adjacency relationship of the subgraph is regarded as an edge to form a subgraph, and the graph attention layer is used to learn a weight for each node, and the features of all nodes in each subgraph are weighted and summed to form the characteristics of the nodes in the subgraph, and finally Use the fully connected layer and the classifier layer to judge whether the network has received an attack. This method builds a hierarchical graph neural network, optimizes the node features in the graph through the graph convolution layer, captures local features on the graph through the pooling layer based on the graph clustering algorithm, and generates high-level semantic features. The situational features of the entire network are generated, and the network situation is classified by a classifier.