摘要
CP-nets是一种简单、直观的图形化偏好表示工具,成为近几年人工智能的一个研究热点。然而对于CP-nets的基础性质——可满足性序列的研究却较少。通过构造CP-nets导出图,利用改进的图的深度优先遍历算法实现二值网的强占优测试,对强占优测试得到的可达矩阵进行分析,得出任意结构CP-nets的可满足性序列个数关系;给出了生成全部可满足性序列的算法;强化和扩充了CP-nets的基本概念,深化了CP-nets的基础理论研究。
CP-nets(condition preference networks) is a simple and intuitive graphical tool for representing conditional ceteris paribus preference statements over the values of a set of variables, and it has been a studying hotspot in artificial intelligence recently. The satisfiability ranking of CP-nets is studied. By constructing induced graph of CP-nets and solving the strong dominance testing with respect to binary-valued CP-nets, the number of satisfiability ranking was derived. And the algorithm of generating all satisfiability rankings was designed by analyzing the reachability matrix acquired from strong dominance testing. What's more, the syntax, semantics and some definitions were collated and introduced.
出处
《计算机科学》
CSCD
北大核心
2015年第5期270-273,285,共5页
Computer Science
基金
山东省高等学校科技计划项目(J14LN23)
山东省自然科学基金(ZR2014FL009
ZR2013FM011)
山东省自然科学基金青年项目(ZR2013FQ023)资助
作者简介
孙雪姣(1979-),女,硕士,讲师,主要研究方向为网络搜索技术、约束满足问题求解,E-mail:sunxuejia06@tom.com;
刘惊雷(1970-),男,博士生,副教授,主要研究方向为程序理论和计算方法。