Agent普适机器学习分类器

Research on classifier of agent pervasive machine learning

从20世纪90年代提出普适计算的透明化和不可见性两个特点开始,到2003年国际普适计算会议提出使用自身的信息来提高系统性能以后,普适计算开始向机器学习的方向发展.本文借助普适计算的基本性质,结合机器学习的特点,提出Agent普适机器学习分类器方法.通过Agent的移动性和普适计算的随时随地性来实现机器学习为各行各业服务的普适功能,从而形成Agent普适机器学习基本内容,包括Agent普适机器学习基本概念和Agent普适机器学习分类器设计.最后通过实例分析进一步证明本文提出方法的有效性.

Pervasive Computing has developed towards machine learning since pervasive computing was put forward in the 1990s. Pervasive machine learning is a new machine learning algorithm to afford satisfaction to the demands of al walk a life, based on the generalization of machine learning combined with the characteristics of the pervasive computing. This thesis presents a method of Agent Pervasive machine learning classifier by means of the basic characteristic of pervasive computing and combing the characteristic of machine learning. For machine learning pervasive ability to serve every field by the mobility of Agent and pervasive computing characteristic of Agent pervasive machine learning, we give some basic concepts and the design of Agent pervasive machine learning classifier, And the validity of our method is further proved by a sample analysis.