基于支持向量机的DNA序列分类系统的设计与实现

针对传统统计方法进行DNA序列分类时要求DNA序列样本的概率分布函数已知，但多数情况下概率分布函数未知这一问题，采用支持向量机这一新的机器学习方法对DNA序列进行分类；以VB和Matlab为主要工具开发了基于支持向量机的DNA序列分类系统。结果表明：该系统能够动态选择DNA训练样本、待测试样本．以及支持向量机模型中的参数，并根据用户的指定条件动态输出计算结果；对于预测一批已知正确分类答案的DNA序列，系统能够自动统计识别率，以观察参数变化对于算法执行结果的影响。支持向量机能够在概率分布函数未知的条件下对DNA序列进行分类。

Design and realization of a DNA sequence classification system based on support vector machines

The distribution of DNA sequence samples must be known when classifying by the traditional statistical methods, unfortunately, it is unknown in most application cases. This paper mainly developed a DNA sequence classification system based on support vector machines (SVM) by VB and Matlab and proposed an new approach to express the DNA sequence data. The test results showed that the system had the merits of dynamically selecting DNA training samples and the samples to be tested, as well as supporting the parameters in SVM model. The system can also dynamically output the calculating results on demand of users, automatically make a statistics of reorganization rate to investigate the effects of parameters variations on the the computing results for a prediction process of a set of correctly classified DNA sequences.