论文部分内容阅读
本文提出一种将粗集方法与SVM算法结合起来的模式分类方法。利用粗集理论在处理大数据量、消除冗余信息等方面优势,减少SVM训练数据,克服SVM算法因为数据量太大,处理速度慢等缺点;同时,借助SVM良好的分类性能,对粗集约简后的最小属性子集进行分类,实现模式分类算法的快速性能、高识别率和抗干扰性强等优点。本文以手写体汉字的识别为例,说明本算法的实用性。
In this paper, a method of pattern classification combining rough set method and SVM algorithm is proposed. The advantages of rough set theory in dealing with large amount of data and eliminating redundant information are reduced, and SVM training data are reduced, overcoming the disadvantages of SVM algorithm such as too large data volume and slow processing speed. Meanwhile, with the help of SVM’s good classification performance, The minimal subset of attributes is classified to realize the fast performance of pattern classification algorithm, high recognition rate and strong anti-interference. In this paper, the recognition of handwritten Chinese characters is taken as an example to illustrate the practicability of this algorithm.