发明授权
- 专利标题: Iterative refinement of pathways correlated with outcomes
- 专利标题(中): 路径的迭代细化与结果相关
-
申请号: US14070092申请日: 2013-11-01
-
公开(公告)号: US09299035B2公开(公告)日: 2016-03-29
- 发明人: Geetika T. Lakshmanan , Szabolcs Rozsnyai , Fei Wang
- 申请人: INTERNATIONAL BUSINESS MACHINES CORPORATION
- 申请人地址: US NY Armonk
- 专利权人: INTERNATIONAL BUSINESS MACHINES CORPORATION
- 当前专利权人: INTERNATIONAL BUSINESS MACHINES CORPORATION
- 当前专利权人地址: US NY Armonk
- 代理机构: Tutunjian & Bitetto, P.C.
- 代理商 Nidhi Kissoon
- 主分类号: G06F15/18
- IPC分类号: G06F15/18 ; G06N99/00
摘要:
A method for refining a process model includes mining a process model from a set of execution traces; determining whether the process model is too dense or too sparse; learning a predictive model from the execution traces to predict an outcome; modifying the predictive model; and mining a refined process model from updated traces based on attributes present in the modified predictive model. Modifying the predictive model includes making the predictive model more specific if it is determined that the process model is too dense; and making the predictive model more general if it is determined that the process model is too sparse.
公开/授权文献
信息查询