发明授权
US07177852B2 Method and apparatus for extracting knowledge from software code or other structured data
有权
从软件代码或其他结构化数据中提取知识的方法和装置
- 专利标题: Method and apparatus for extracting knowledge from software code or other structured data
- 专利标题(中): 从软件代码或其他结构化数据中提取知识的方法和装置
-
申请号: US10312768申请日: 2001-06-28
-
公开(公告)号: US07177852B2公开(公告)日: 2007-02-13
- 发明人: Zhan Cui , Yang Li , Hongji Yang
- 申请人: Zhan Cui , Yang Li , Hongji Yang
- 申请人地址: GB London
- 专利权人: British Telecommunications public limited company
- 当前专利权人: British Telecommunications public limited company
- 当前专利权人地址: GB London
- 代理机构: Nixon & Vanderhye P.C.
- 优先权: GB0016639.7 20000706; EP00307431 20000830
- 国际申请: PCT/GB01/02907 WO 20010628
- 国际公布: WO02/05088 WO 20020117
- 主分类号: G06F17/00
- IPC分类号: G06F17/00 ; G06N5/02
摘要:
A method for knowledge representation and management is disclosed which has particular but not exclusive application in knowledge recovery techniques. To help with program understanding, software maintainers use computer aided reverse engineering (CARE) tools which utilize techniques such as syntax analysis, structural analysis, and domain knowledge base analysis (DKBA). Known DKBA methods are complex since the domain knowledge recovery relies on the use of knowledge at different layers while some elements of the knowledge at domain level is not used at all. In the disclosed method for representing knowledge in a knowledge base, the knowledge is represented by a set of one or more concepts and by one or more first type of relationships between two or more or said concepts. The method comprises dividing the set of concepts into a plurality of sub-sets, each sub-set comprising a set of concepts related to a sub-set of the knowledge; dividing the relationships among the sub-sets of concepts; and providing a further set comprising all the concepts along with a second type of relationship that provides an association between each concept in the further set of concepts and one or more of the sub-sets of concepts and relationships.