http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113505244-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-214 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-367 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-36 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
filingDate | 2021-09-10-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2021-11-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2021-11-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-113505244-B |
titleOfInvention | Knowledge graph construction method, system, equipment and medium based on deep learning |
abstract | The present application discloses a knowledge graph construction method, system, device and medium based on deep learning. The method includes: inputting unlabeled medical document data into a relation extraction model to construct a first knowledge graph, and the relation extraction model is based on obtaining constructed from the labeled medical literature data; input the unlabeled medical literature data into the auxiliary labeling model to determine the classification result of each entity in the unlabeled medical literature data, and the auxiliary labeling model is constructed based on the medical database; The supervised learning algorithm updates the relation extraction model according to the classification result of each entity and the first knowledge graph, and obtains the second knowledge graph. This solution can label massive unlabeled medical literature data based on the auxiliary labeling model integrated with the medical database, and iteratively update the relation extraction model through a semi-supervised learning algorithm, which improves the generalization ability of the relation extraction model, so as to construct a comprehensive and rapid construction Generate high-quality knowledge graphs. |
priorityDate | 2021-09-10-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 33.