50 results about "Dendrogram" patented technology

A document correlation diagram drawing device includes extracting means (20, 30) for extracting content data and time data of document elements (E) each including one or more documents, dendrogram drawing means (50) for drawing a dendrogram showing a correlation between documents on the basis of the content data of the document elements, clustering means (70) for cutting the dendrogram in accordance with a predetermined rule and extracting clusters, and intra-cluster arranging means (90) for determining an intra-cluster arrangement of the document elements belonging to each cluster on the basis of the time data of the document elements. Accordingly, a dendrogram adequately showing the chronological development in each field can be automatically drawn.

A computer system and method is disclosed for mining current and archived address data in order to identify a preferred address for each service point in a territory. The data mining system may start in response to the presentation of a candidate address for matching. The set of mined data may be prioritized by clustering like characteristics, building similarity matrices, and by constructing dendrograms with nodes joined according to common characteristics. A computer system and method for maintaining a central database of preferred addresses is also disclosed. Selected address data gathered in a queue may be scored by characteristic, grouped by consignee location, and staged for processing. The scored queue of data may be prioritized by clustering like characteristics, building similarity matrices, and by constructing dendrograms.

Data from a data set is displayed as a tree map visualization by identifying data elements in the data set to be highlighted and generating a tree map visualization based on the data set where a location of bounding boxes corresponding to the identified data elements have greater color saturation in comparison to other bounding boxes.

A community is extracted by executing steps of: clustering relationship data; extracting a communication core of a relationship network; mapping the communication core to a dendrogram of relationship data; forming a community by using the dendrogram in accordance with a similarity degree of relationship data while the cluster is expanded; and aggregating communities. A community of a set of persons having high density relationships based on common topics and interests can be extracted from a set of human relationships and relationship data representative of the human relationships.

A group of cells are newly clustered on the basis of the result of implementing a SOM. A plurality of pieces of multivariate data are clustered via a SOM, and cells are displayed on a two-dimensional plane as rectangular or hexagonal shapes. The level of similarity between representative vectors from each adjacent cell is calculated, and a dendrogram is three-dimensionally depicted. Cells on a SOM map are colored differently in accordance with a plane for partitioning the dendrogram.

Source data is analyzed and visualized for a user. As the user adjusts graphical query devices, the displayed data is filtered and the visualization is updated accordingly. The invention is particularly useful where the data relates to chemical compounds in that a SAR table is preferably generated, presented in the visualization, and updated according to the query device settings. The displayed SAR table entries may also be clustered as a function of similarity with respect to a chosen property, such as molecular weight or some topological characteristic. A hierarchy of similarity may then be indicated in the SAR table using, for example, a dendrogram.

The present invention is directed to a data processing apparatus and a computer implemented method for modeling and analyzing relational data represented in a network that includes a plurality of nodes and a plurality of connections between the nodes. The method includes assigning at least one weight to a connection between two nodes in the network. A set of possible dendrograms is then generated for the network, and a likelihood of each dendrogram in the set is determined. The determination of the likelihood is based on at least the one weight of the connection. One of the dendrograms from the set is selected as an optimal dendrogram based on the determined likelihood. The selected dendrogram is then output via an output device. The dendrogram may be used to predict missing links or identify any possible false-positive (noisy) links within a relational dataset.

A computerized method of representing a dataset with an optimal layer of a taxonomy includes obtaining a taxonomy including a hierarchical arrangement of layers, wherein each layer represents a cluster set containing at least one cluster, wherein each cluster represents at least one record within a dataset; identifying a range of taxonomy layers based on a measure of intra-cluster homogeneity of each cluster within the taxonomy; selecting a taxonomy layer within the identified range as an optimal layer of the taxonomy, the selecting being based on a measure of inter-cluster heterogeneity between clusters of a taxonomy layer within the identified range; and labeling clusters within the optimal layer of the taxonomy.

The invention relates to a method of clustering a set of proteins based on the sequence similarities of functional domain(s) such as pocket(s), functional site(s), allosteric site(s), and active site(s). The functional domain(s) of a protein sequence are identified based on the three-dimensional structure of the protein. Proteins are clustered based on the sequence similarity of the amino acid residues of the functional domain(s) and represented as a dendrogram. Proteins in a particular cluster show similar interaction patterns with specific drugs. Methods for identifying modulators for drug discovery based on the similarities of the functional domain(s) are provided.

Described is a system for automated community discovery in networks with multiple relational types. The system receives a network as input. The network comprises neighbors, edges connecting the neighbors, and vertices, where edges between two vertices represent a relation. A set of pair-wise similarity comparisons is computed for all pairs of relations. Two relations are considered similar if vertices connected to the two relations share similar relations to the same set of neighbors. A relation dendrogram is created based on the set of pair-wise similarity comparisons. Then, a cut in the relation dendrogram is selected to compute a community solution, resulting in a plurality of relation dendrogram partitions. Each relation dendrogram partition represents a community. A community density criterion is computed based on a density of each community calculated with respect to edge types contained within each community. Finally, a community solution is generated that maximizes the community density criterion.

The present invention represents a global state of branches in a whole dendrogram as well as detail of states of individual subtrees, to aid focusing of groupings and selection of a clustering method. The present invention selects a branch / subtree in a dendrogram displays the subtree on a separate window, replaces the subtree with an icon then restores the icon into the original subtree, and / or searches keywords contained in information of biopolymers in the subtree then highlight the locations of biopolymers containing the keywords in its information.

The invention provides a genetic analysis method used for rice. According to the method, capillary gel electrophoresis is utilized to detect SSR-PCR amplification products, and the high resolution characteristic of the capillary gel electrophoresis technology is utilized to distinguish rice varieties with near genetic relationship, so that analysis results are more accurate. By adoption of the genetic analysis method based on capillary gel electrophoresis and SSR labeling, a dendrogram established by UPGMA clustering analysis is used for genetic relationship analysis of rice varieties. The genetic analysis method has advantages of high resolution, rapidness, high efficiency, accurate and reliable results, and the like. The method has characteristics of easy automation and convenience for integration and analysis of a large amount of data, which are beyond the grasp of traditional detection methods.

The invention relates to a multidimensional assessment method of petroleum and natural gas pipeline warning priority. The method comprises the following steps: (1) accessing a database, and obtaining pipeline technology data; (2) expressing a technology process and a warning variable with symbols, establishing a warning dendrogram and finding a key warning; (3) calculating the technology time of a single warning according to technology conditions; (4) assessing warning response time according to the technology time, and determining a corresponding MTTR factor; (5) respectively assessing the severity of all consequences of the single warning, and determining the corresponding weighting; (6) calculating the point values of initial priorities, using the key warning in the step (2) as a node, successively adding 5 point values to the initial point values before the node, and successively subtracting 5 point values from the initial point values after the node so as to obtain the final point value; (7) determining the grade and the displaying and disposing manner of the warning priority, and inputting the grade and the displaying and disposing manner of the warning priority into the database. According to the invention, the multidimensional priority assessing method is formed for the warning in a pipeline warning system, so that more accurate and practical priority division is provided.

An urban low-voltage partitioning method based on a spectral clustering algorithm specifically comprises steps as follows: (1) acquiring multi-factor data of a power grid and constructing a sample space; (2) constructing a Laplacian matrix and automatically determining the number of low-voltage governance subareas according to a Laplacian matrix eigenvalue difference; (3) mapping first two eigenvectors and first three eigenvectors of the Laplacian matrix to two-dimensional and three-dimensional spaces, and partitioning governance subareas directly; (4) constructing an evaluation function and evaluating the low-voltage partitioning effect; (5) clustering the eigenvectors with a k-means algorithm and drawing up a subarea dendrogram to obtain a low-voltage subarea governance scheme with clear hierarchy. The low-voltage governance partitioning method considers four indexes, namely, the lowest voltage amplitude, the number of involved low-voltage households, the voltage lower limit crossing time and the annual power supply, analyzes the low-voltage state of a distribution network comprehensively, and can provide a decision basis for power grid workers and improve the low-voltage governance efficiency.

The invention discloses a text-oriented domain classification relationship automatic learning method. The method comprises the steps of adopting MEDLINE as a corpus library; performing term extractionand concept extraction; performing syntax similarity and semantic similarity-based five dimension similarity calculation for extracted concepts; performing weighting on the similarity of each dimension to obtain a final similarity matrix; based on this, performing hierarchical clustering to obtain an initial tree diagram; and performing corresponding pruning and cluster marking on the tree diagram to finally obtain a tree diagram reflecting a classification relationship among the concepts. According to the method, a large amount of manual marking is not required, so that the manpower and timeoverhead is saved; extracted terms and a UMLS metathesaurus of an authoritative knowledge base are mapped to obtain accurate domain concepts; by adopting a distributed method of the hierarchical clustering, and in combination with domain background knowledge, the five dimension similarity calculation is provided; and an extremal distance estimation-based unsupervised hierarchical clustering dynamic pruning method is proposed, so that the domain-related classification relationship can be better obtained.

Disclosed herein are methods and systems for identifying targets for distinguishing cell types. In some embodiments, the method comprises: hierarchically clustering expression profiles of cells to generate a dendrogram with each leaf representing the expression profile of a different individual cell. The dendrogram can be pruned by eliminating invalid splits (and their children). The remaining leaves in the dendrogram can be merged, independent of their locations in the dendrogram, based on their distances to one another to generate clusters of expression profiles. The method can identify targets for distinguishing cell types based on targets that are expressed differentially in the clusters.

The invention discloses a multi-source network public opinion theme mining method based on improved hierarchical clustering, and relates to the field of theme mining. The method specifically comprisesthe following steps of 1, obtaining a word vector; 2, preprocessing all the data; 3, vectorizing the total sample data sentences preprocessed in the step 2; 4, carrying out sentence vector semi-supervised hierarchical topic mining; and 5, outputting a tree diagram Dendrogram. According to the method, by utilizing the advantage that the hierarchical clustering algorithm comprises the hierarchicalinformation, and on the basis, carrying out optimization at the aspects of priori knowledge use, model input vectorization, high-quality topic screening and the like, so that finally the method provided by the invention can be effectively applied to the topic mining of multi-source network platform short texts with the wide topics, high text noise and lack of grammar specifications.

The invention relates to a method for early screening of lung cancer by using a vacuum ultraviolet photoionization mass spectrometer, and belongs to the field of efficient and rapid non-invasive cancer screening. According to the method, lung cancer characteristic marking VOCs gas in human exhaled gas can be detected and analyzed directly. The method comprises the following steps: firstly, collecting gas exhaled by a person to be detected with a disposable aluminum foil gas sampling bag; then, enabling the gas to enter a vacuum ultraviolet photoionization VOCs mass spectrometer; processing sampling signals with a GHz threshold discrimination ion signal processer; identifying 20 types of components of the lung cancer characteristic marking VOCs gas in a data image on a computer, so as to compare and calculate with a lung cancer early screening algorithm dendrogram, and carry out risk assessment and early screening to judge whether a person suffers from lung cancer. The method can be applied to the aspects of screening of high-risk population, early detection of lung cancer, prognosis monitoring and the like, and has great significance in reducing the lethality of lung cancer.

The invention relates to a molecular marking method for carrying out genetic relationship and clustering analysis on citrus fruit flies and belongs to the technical field of molecular marking. The molecular marking method for carrying out genetic relationship and clustering analysis on the citrus fruit flies comprises the following steps: carrying out PCR (polymerase chain reaction) amplification on DNA of citrus fruit flies by utilizing seven RAPD (random amplification polymorphism DNA) primers and 10 pairs of SRAP (sequence-related amplified polymorphism) primer combinations, carrying out agarose gel electrophoresis on amplification products, taking pictures, recording results, and carrying out statistics on an amplification zone of a detected material; and analyzing with genetic analysis software to obtain genetic distance between individuals, and drawing a dendrogram of citrus fruit flies in different geographical populations for distinguishing different class groups. The molecular marking method for carrying out genetic relationship and clustering analysis on the citrus fruit flies has the advantages of stable reaction, clear amplified fragments and high polymorphism, a molecular reaction system for population genetic diversity of the citrus fruit flies is constructed, genetic diversity among populations can be shown, clustering analysis can be rapidly and accurately carried out on the citrus fruit flies, different populations can be distinguished, and the molecular marking method for carrying out genetic relationship and clustering analysis on the citrus fruit flies has an important significance on prevention and control of the citrus fruit flies.

The invention relates to the technical field of data processing. The invention discloses a log optimization method and device, log optimization equipment and a storage medium. The log optimization method comprises: extracting attribute characteristics and attribute values of each log in a log file set; classifying same, obtaining sets of attributes, constructing a a dendrogram, selecting a targetattribute feature, performing abstract processing on an attribute value under the target attribute feature according to a preset abstract algorithm to obtain a key abstract of the log file set, replacing the log file set stored in the current system with the key abstract, and finally analyzing the key abstract to improve the exception diagnosis efficiency.

The invention relates to the technique of development and application of spirulina and aims at providing a method for detecting production trait goodness and badness of spirulina strains. According to the method, a Tris-HC1 extracting solution is utilized for extraction so as to obtain water-soluble protein of spirulina cells; protein samples are separated by using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, subsequently the optical density value of a protein zone at 102kD of a protein electrophoresis diagram is detected, and a dendrogram of the spirulina plants is established according to the value; if the optical density value of a candidate strain at the protein zone at the 102kD is larger than 140 and the candidate strain is gathered together with a known stain with good production nature, the stain is good in production nature and is applicable to large-scale cultivation and production; and if the optical intensity value of the protein zone at 102kD is less than 40 and the candidate strain is gathered together with known strains with poor production nature, the stain is poor in production nature and is not applicable to large-scale cultivation and production. The method is simple, efficient, reliable and low in cost, and complex tests and analysis on nucleic acid sequencing and bioinformatics comparison do not need to be carried out, therefore, the method is applicable to large-scale and high flux separation.

The invention provides a method for presenting visual search results on a mobile terminal. The method comprises the following steps: receiving a search condition, searching a matched target application according to the search condition, and obtaining associated applications having an associated relationship with the target application; generating a visual dendrogram comprising a root node icon and child node icons connected with the root node icon, wherein the root node icon is mapped to the target application, and the child node icons are mapped to the associated applications; representing the visual dendrogram on the mobile terminal. Correspondingly, the invention further provides a system for presenting visual search results on the mobile terminal. Through the implementation of the method and the device provided by the invention, the intuitive and demonstrative properties of the search result of the mobile application are improved, and meanwhile, the operability of the search results and the user experiences are also improved.

A document correlation diagram drawing device includes extracting means (20, 30) for extracting content data and time data of document elements (E) each including one or more documents, dendrogram drawing means (50) for drawing a dendrogram showing a correlation between documents on the basis of the content data of the document elements, clustering means (70) for cutting the dendrogram in accordance with a predetermined rule and extracting clusters, and intra-cluster arranging means (90) for determining an intra-cluster arrangement of the document elements belonging to each cluster on the basis of the time data of the document elements. Accordingly, a dendrogram adequately showing the chronological development in each field can be automatically drawn.

The invention discloses a gastroesophageal reflux disease danger factor determining method based on machine learning and a system thereof. The method and the system settle a problem of low accuracy indetermining the gastroesophageal reflux disease danger factor according to statistics. The method comprises the steps of constructing a user information set which comprises the gastroesophageal reflux disease danger factor, performing quantification processing on the factor in the user information set, and obtaining a quantification data matrix; secondly, performing standardization on the quantification data matrix, performing dimension reducing processing on the standardized matrix according to a main component analysis algorithm; then clustering the data in the processed data set by means of a hierarchy clustering algorithm, and obtaining a hierarchy clustering dendrogram; then according to a clustering number which is determined by the hierarchy clustering dendrogram, performing clustering dividing on the data in the processed data set according to the clustering number, and obtaining a plurality of class clusters; and finally calculating a correlation index among elements in eachclass cluster, and determining the element with highest correlation index as the gastroesophageal reflux disease danger factor.

Described is a system for feature selection that extends supervised hierarchical clustering to neural activity signals. The system generates, using a hierarchical clustering process, a hierarchical dendrogram representing a set of neural activity data comprising individual neural data elements having neural activity patterns. The hierarchical dendrogram is searched for an optimal cluster parcellation using a stochastic supervised search process. An optimal cluster parcellation of the hierarchical dendrogram is determined that provides a classification of the set of neural activity data with respect to a supervised classifier, resulting in a reduced neural activity feature set. The set of neural activity data is classified using the reduced neural activity feature set, and the classified set of neural activity data is decoded.

The invention discloses a wind power plant abandoned wind power assessment method based on a hierarchical clustering method. The method comprises the following steps: step S10, performing standardization moment flattening on actually measured wind speed data of a wind power plant; step S20, subjecting a wind speed matrix after the standardization moment flattening to empirical orthogonal function decomposition; step S30, based on the empirical orthogonal function decomposition result, calculating the variance contribution rate and the cumulative variance contribution rate of each spatial type; step S40, performing clustering analysis on the spatial matrix obtained by decomposing the empirical orthogonal function using the hierarchical clustering method; step S50, drawing a clustering dendrogram based on the result of hierarchical clustering analysis; step S60, dividing wind turbine generators based on the clustering dendrogram, selecting a mark post draught fan and determining distribution coefficients; and step S70, using the mark post draught fan method to calculate the amount of the abandoned wind power. The method of the invention can reduced errors in a mark post draught fan method, and relatively accurately reflect abandoned wind conditions in the wind power plant.

A method for verifying whether a queried text of less than 500 characters has been compiled by an author, comprising the following steps:multivariate statistical analysis of the queried text, for example, PCA or PCoA, in order to generate a matrix of coordinates in a space with N dimensions;hierarchical clustering of the points of this space that can be represented by a dendrogram;verification of the author of the queried text on the basis of this clustering.

A method of comparing dendrograms created by different references or algorithms displays individual dendrograms (201, 202) on a screen in an overlapped yet distinguishable manner. The overlapped portions of the dendrograms are displaced vertically or horizontally from one another and indicated by different colors for individual dendrograms. A specified branch and elements hanging therefrom can be highlighted when displayed, such that differences in the manner in which individual elements of the overlapped dendrograms are clustered can be easily recognized.

The invention provides an adversarial training method and device based on a neural network model. In adversarial training of the neural network model, defining a plurality of mixed features in the neural network model, and performing feature purification on the plurality of mixed features; forming a correlation matrix based on features in the neural network model, and performing hierarchical clustering on the correlation matrix to construct a tree diagram; forming different clustering groups according to the tree diagram, and allocating each clustering group to a corresponding neuron; associating each neuron with a cluster group, introducing Gaussian random variables and spherical Gaussian variables to guide adversarial training, at the moment, performing adversarial training on a neural network model so as to facilitate adversarial training on the neural network model, performing feature purification on a plurality of mixed features, and allocating each cluster group to the corresponding neuron so as to obtain an adversarial training result; gaussian random variables and spherical Gaussian variables are introduced to guide adversarial training, so that the adversarial robustness of the neural network model is enhanced.

The invention relates to the technical field of water quality information monitoring, in particular to a water quality monitoring method, system and device for environmental protection. According to the method, an aggregation hierarchical clustering algorithm is utilized, monitoring nodes are clustered based on pollution degree differences among the monitoring nodes, and after a tree diagram is obtained, the tree diagram is segmented to obtain a combination of clustering clusters segmented each time. For each time of segmentation, representative nodes are screened out based on the pollution degree difference between the nodes, the feature significance of the representative nodes obtained through current segmentation is evaluated in combination with distribution of the representative nodes in the drainage pipe network and the influence degree of key areas, the reasonable degree is obtained, the optimal segmentation process is screened out, and key monitoring nodes are determined. Based on the hierarchical clustering algorithm, the key monitoring node position with remarkable characteristics can be determined based on the dynamic characteristics of the water area of the current drainage pipe network, and effective pollution information in the current drainage pipe network can be extracted based on the key monitoring node position.