基于制图专家案例库的知识归纳与推理-中国图象图形学报.docVIP

中图法分类号：TP301.6 文献标识码：A 文章编号：1006-8961（2013）- 论文引用格式： 采用案例归纳推理进行道路网智能选取 郭敏1,2，钱海忠1，黄智深1,3，刘海龙1，王骁1 1.信息工程大学 地理空间信息学院，河南 郑州，450052；2.75719部队, 湖北 武汉，430074；3 南疆测勤队, 新疆 喀什，844200 摘 要： 自动制图综合集技术、艺术与制图人员经验于一体，长期以来其自动化、智能化研究进展缓慢。基于机器学习的智能化自动综合也成为了制图综合发展过程中必须解决而仍未得到很好解决的核心难题之一。本文提出基于案例归纳学习的道路网智能选取方法。该方法以制图专家道路网选取案例库为学习对象，以决策树算法为推理机，从专家案例库中自动归纳、推理来获取决策树，并转化为满足计算机自动执行的规则集，据此来进行道路网自动选取。从而解决了把难以形式化表达的制图专家经验自动转化为满足计算机自动综合要求的规则，并据此进行智能化自动综合这一难题。最后，采用实例对本方法进行了验证，实验结果表明，该方法能够从专家案例库中自动获取核心规则，并进行自动综合，综合结果能够有效地反映制图专家的制图综合经验，同时具有普适性，从而为智能化自动制图综合发展探索了新的途径。 关键词：道路网；制图综合案例；决策树；归纳推理；智能 Intelligent road network selection method based on cases inductive reasoning Guo Min1,2, Qian Haizhong1, Huang Zhishen1,3, Liu Hailong1, Wang Xiao1 1 Institute of Geographical Spatial Information, Information Engineering University, Zhengzhou 450052,China;2 75719 Troops, Wuhan 430074, China;3 Surveying and Mapping Support Unit of South Xinjiang, Kashi 844200, China Abstract：The intelligence of automated generalization developed slowly because of the integration of complex generalization technology, art and cartographers’ experience. And the intelligent generalization based on machine-learning has also been one of the puzzles in the progress of automated generalization. A new approach of road network intelligent selection based on cases inductive reasoning was put forward in this paper, which took the road network selection case lib of cartographers as leaning objects, decision tree algorithm as reasoning machine, and concluded rules from expert case lib to form decision tree. Then the decision tree was transformed into rules that satisfied the computer’s requirement. With these rules, computer could generalize road network selection automatically. Through this approach, the core problem of transforming cartographers’ experience into rules that satisfying computer generalization automatically, and generalizing road network intelligently based on the rules was solved. Examples illustrate that, the new appro