http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112214992-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_e9eabbd5f0a2678151e329094ecfdd66 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F40-211 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F40-216 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F40-216 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F40-211 |
filingDate | 2020-10-14-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_66a341f0f0ab1bbf016341368ad99ec4 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_4a26d8e6f591238715e33718697d4d95 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_2df74890c48b5f91e533671c4f385c3b http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_7bea564584795c02bda9f07739f2e4a2 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_396f927bd9c86917756da59c933999bd |
publicationDate | 2021-01-12-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-112214992-A |
titleOfInvention | Deep learning and rule combination based narrative structure analysis method |
abstract | The invention discloses a narrative structure analysis method based on deep learning and rule combination, and belongs to the technical field of composition analysis. S1, selecting a structural tag group analysis system according to the composition type, and segmenting the narrative into segments and sentences; s2, sentence-level structure label recognition is carried out on each sentence in each paragraph, and the effect of each sentence in the whole text is obtained; s3, identifying the partial subsection structure label segment by segment to obtain the function of the label in the whole text; and S4, identifying the structural labels needing the related information of other paragraphs, adjusting the existing structural labels, and obtaining the result of the structural analysis of the narrative. According to the method, the narrative is divided into four types of scenes, writers, narratives and forms according to different description objects, a set of complete narrative structure system is designed, and the narrative structure can be automatically analyzed by using a method of combining deep learning and rules without manual processing, so that the speed of analyzing the narrative structure is greatly increased, and the labor cost is saved. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113836306-A |
priorityDate | 2020-10-14-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 30.