模式识别实验报告叶斯分类器.docVIP

模式识别理论与方法 课程作业实验报告 实验名称：Generating Pattern Classes 实验编号：Proj02-01 规定提交日期：2012年3月30日 实际提交日期：2012年3月24日 摘要： 在熟悉贝叶斯分类器基本原理基础上，通过对比分类特征向量维数差异而导致分类正确率发生的变化，验证了“增加特征向量维数，可以改善分类结果”。 对于类数的先验概率已知、且无须考虑代价函数的情况，贝叶斯分类器相当简单：“跟谁亲近些，就归属哪一类”。 技术论述： 贝叶斯分类器基本原理：多数占优，错误率最小，风险最低 在两类中，当先验概率相等时，贝叶斯分类器可以简化如下： 2，增加有效分类特征分量，可以有助于改善分类效果 实验结果讨论： 从实验的过程和结果来看，进一步熟悉了贝叶斯分类器的原理和使用，基本达到了预期目的。 实验结果： 图1 原始数据 图2 按第1 个特征分量分类结果 图3 按第2 个特征分量分类结果 图4 综合两个特征分量分类结果 附录：（程序清单，参见压缩包） %在Matlab 版本2009a 下运行通过 videoconference, Secretary of the provincial Committee of Xia Baolong stressed: this year is the five water treatment 357 timetables for three years to resolve outstanding issues, obviously worked, victory of the year. The Countys departments at all levels must firmly establish the guanghui is gold and silver green development concept, convinced that Castle is not relaxed, always maintaining strategic ability and strong determination to improve the environment, consistently playing well five water treatment of winning the war, ever-higher levels of comprehensive well-off society was built. (A) lay the sewage uphill battle. This year we will river water quality improved at the county level, up to five this basic objective grasp of pollution control that focus, with the focus on three things: a good job, clear rivers reaching the County created. Organization look back special inspection, the County garbage River, black and odorous River is Dragnet investigation, timely rectification of problems found in place firmly to prevent river pollution rebound. In accordance with the one-stop strategy, improving the inferior five water quality monitoring stations work this year to ensure the Elimination of 1 provincial control, five water quality section 3 municipalities controlled deterioration, ensure the creation this year of clear rivers standards. Second, pay special attention to pollution-cutting construction of nanotubes. Intercepting nanotubes are the basis