http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-103077404-B
Outgoing Links
Predicate | Object |
---|---|
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
filingDate | 2012-12-31-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2016-07-06-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2016-07-06-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-103077404-B |
titleOfInvention | Local regularity formula based on markov random file visualizes clustering method |
abstract | The invention discloses a kind of local regularity formula based on Markov random field and visualize clustering method, mainly include iteration and process two processes, wherein: totally unknown target data is calculated its dissimilarity matrix, using the input as iterative process of this matrix, obtain k rank neighborhood system and informativeness matrix;Using the output of iterative process as the input of processing procedure, obtain the information such as image and cluster confidence level that reorders.The method character according to Markov random field, adopts k rank neighborhood system to calculate overall situation probability measure;According to k rank neighborhood system, reduce computation complexity by dimensionality reduction;According to informativeness matrix, select optimum division classification, and provide the confidence level of cluster result further.The method provides prior information for cluster, understands the distribution situation of data structure and data category intuitively;Simultaneously, moreover it is possible to once can provide cluster result and Cluster Evaluation;And provide the implicit labyrinth of data and cluster confidence level. |
priorityDate | 2012-12-31-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 13.