| 基于共享最近邻的专利自动分类技术研究 |
| |
| 引用本文： | 季铎,蔡云雷,蔡东风,苗雪雷.基于共享最近邻的专利自动分类技术研究[J].沈阳航空工业学院学报,2010,27(4):41-46. |
| |
| 作者姓名： | 季铎 蔡云雷 蔡东风 苗雪雷 |
| |
| 作者单位： | 沈阳航空航天大学知识工程研究中心,辽宁沈阳110136 |
| |
| 摘 要： | 专利文献是一种具有法律效力的科技文献,其内容覆盖了几乎所有的技术领域的最新、最活跃的创新技术信息,因此对专利文献的分类和组织在专利管理中有着重要意义。本文通过对现有的专利自动分类技术的归纳和分析,融合BM25相似度计算方法和样本邻域信息提出一种基于共享最近邻的KNN专利自动分类方法。本方法的有效性在NTCIR-8专利分类评测任务中得到充分验证。
|
| 关 键 词： | 专利分类 BM25 KNN 共享最近邻 |
Patent automatic classification research based on shared nearest neighbor |
| |
| JI Duo,CAI Yun-lei,CAI Dong-feng,MIAO Xue-lei.Patent automatic classification research based on shared nearest neighbor[J].Journal of Shenyang Institute of Aeronautical Engineering,2010,27(4):41-46. |
| |
| Authors: | JI Duo CAI Yun-lei CAI Dong-feng MIAO Xue-lei |
| |
| Institution: | (Knowledge Engineering Research Center,Shenyang Aerospace University,Liaoning Shenyang 110136) |
| |
| Abstract: | Patent document is related law document,which almost covered all the latest and the most activity innovative information in technical fields,therefore patent classification is of important significance in patent management.Through induction and analysis of existing patent automatic classification technology,this paper presents a KNN patent categorization method based on shared nearest neighbor,effectively combining the BM25 similarity calculation method and the neighborhood information of samples.Validity of the method has been fully verified in the NTCIR-8 patent classification evaluation. |
| |
| Keywords: | patent classification BM25 KNN SNN |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
