
基本信息:
- 专利标题: 一种文献摘要提取方法、设备及可存储介质
- 专利标题(英):Literature abstract extraction method and device and storable medium
- 申请号:CN201910535327.1 申请日:2019-06-20
- 公开(公告)号:CN110399477A 公开(公告)日:2019-11-01
- 发明人: 高源伯 , 李向伟 , 高爽 , 柴博 , 宋博川
- 申请人: 全球能源互联网研究院有限公司 , 国家电网有限公司 , 国网浙江省电力有限公司电力科学研究院
- 申请人地址: 北京市昌平区未来科技城滨河大道18号
- 专利权人: 全球能源互联网研究院有限公司,国家电网有限公司,国网浙江省电力有限公司电力科学研究院
- 当前专利权人: 全球能源互联网研究院有限公司,国家电网有限公司,国网浙江省电力有限公司电力科学研究院
- 当前专利权人地址: 北京市昌平区未来科技城滨河大道18号
- 代理机构: 北京三聚阳光知识产权代理有限公司
- 代理人: 张琳琳
- 主分类号: G06F16/34
- IPC分类号: G06F16/34 ; G06F17/27
The invention discloses a literature abstract extraction method and device and a storable medium, and belongs to the technical field of computers. The method comprises the steps of obtaining a targetdocument, splitting the target document into at least one sentence, and performing feature extraction on each word in the sentence to obtain a word initial feature vector; determining a preset word inthe sentence based on the word initial feature vector to obtain a feature vector of the sentence; based on the feature vectors of the sentences, determining context features of the sentences; and according to the context features of the sentences, determining whether the sentences are sentences in the target document abstract or not to obtain the target document abstract. Therefore, the abstractextracted by comprehensively considering the characteristics of the words in each sentence in the target document and the characteristics of the context of the sentence can improve the accuracy, and can obtain a faster speed and richer context association, thereby obtaining a more accurate result.