平行坐标法在分类算法中的可视化应用  

Visual Appfication on Classification Methods Based on Parallel Coordinates

作　　者：董军凯[1] 

机构地区：[1]燕山大学,河北省秦皇岛市066004

出　　处：《电脑应用技术》2009年第1期1-6,共6页Microcomputer Application Technology

摘　　要：平行坐标技术是可视化数据挖掘的重要方法之一，通过此方法实现分类中多维数据在二维平面上的显示，设计平行坐标与分类算法结合的方法，通过程序接口把具体算法与平行坐标实现分离开来，完成了最近邻分类算法与平行坐标可视化方法的紧密结合的程序实现，体现出平行坐标可视化方法对数据挖掘技术所带来的重大意义。Parallel coordinates technology was one of the important visual data mining methods. Parallel coordinates made multi-dimensional data present on two dimensions plane in the classification of visualization. Designed the intcgrity process of parallel coordinates combining with classification algorithm, separated throngh the standard interface the concrete algorithm and the parallel coordinates realization. Completed programs of K-nearest neighbor classification algorithms, in close integration with parallel coordinates, manifested parallel coordinates visualization the great significance which brought to the data mining technology .

关 键 词：平行坐标 可视化数据挖掘 最近邻分类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP311.13[自动化与计算机技术—计算机科学与技术]