http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112784017-B
Outgoing Links
| Predicate | Object |
|---|---|
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-35 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-23 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-45 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-3344 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-583 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-25 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-3346 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-55 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-432 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-35 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-55 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-33 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-45 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-583 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-432 |
| filingDate | 2021-01-28^^<http://www.w3.org/2001/XMLSchema#date> |
| grantDate | 2022-10-14^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationDate | 2022-10-14^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-112784017-B |
| titleOfInvention | A Feature Fusion Method for Archival Cross-modal Data Based on Principal Affinity Representation |
| abstract | The invention discloses an archive cross-modality data feature fusion method based on main affinity representation, which belongs to the field of cross-modality retrieval. The invention proposes a multi-feature fusion mutual information NBPCFMI algorithm to screen feature words of text corpus , realizes the feature representation of archive graphic data; on this basis, a principal affinity representation HKPAR algorithm based on mixed kernel function is invented. For the learning problem that a single kernel function cannot take into account the global and local features, the Gaussian kernel function and the The hybrid kernel function combined with the polynomial kernel function realizes the calculation of the main affinity, and finally realizes the semantic mapping between the above-mentioned representation and the label through multiple logistic regression, and realizes the unified representation of the archive graphic data. Compared with traditional file retrieval, it can greatly improve the retrieval efficiency and accuracy. |
| priorityDate | 2021-01-28^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
| Predicate | Subject |
|---|---|
| isDiscussedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID14058250 http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID426670387 |
Showing number of triples: 1 to 26 of 26.