<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[威廉的大数据实验室]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/bigdata_wang</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; bigdata_wang]]></copyright><item><title><![CDATA[如何构建一个死锁？]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/49805387</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/49805387</guid><author>bigdata_wang</author><pubDate>Thu, 12 Nov 2015 22:25:39 +0800</pubDate><description><![CDATA[死锁的构建及避免]]></description><category></category></item><item><title><![CDATA[Java中String为什么是Immutable的]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/49593415</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/49593415</guid><author>bigdata_wang</author><pubDate>Thu, 12 Nov 2015 17:03:19 +0800</pubDate><description><![CDATA[大家有没有想过为什么Java里的String是Immutable的呢？]]></description><category></category></item><item><title><![CDATA[MAC $JAVA_HOME设置]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/49706199</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/49706199</guid><author>bigdata_wang</author><pubDate>Sat, 07 Nov 2015 22:24:07 +0800</pubDate><description><![CDATA[Mac中JAVA_HOME环境变量的设置方式]]></description><category></category></item><item><title><![CDATA[Spark Machine Learning 总览]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/49429163</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/49429163</guid><author>bigdata_wang</author><pubDate>Mon, 26 Oct 2015 23:55:19 +0800</pubDate><description><![CDATA[Spark的ML(Machine Learning)库提供了主流数据统计/挖掘算法的实现，威廉将在本文中做一个总览]]></description><category></category></item><item><title><![CDATA[为什么Java不支持多继承？]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/49335371</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/49335371</guid><author>bigdata_wang</author><pubDate>Thu, 22 Oct 2015 19:46:46 +0800</pubDate><description><![CDATA[我们知道Java不支持多继承，但这样的设计是出于什么考虑呢？]]></description><category></category></item><item><title><![CDATA[Java Collection的Fail fast与Fail safe迭代器]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/49315261</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/49315261</guid><author>bigdata_wang</author><pubDate>Thu, 22 Oct 2015 11:27:09 +0800</pubDate><description><![CDATA[Java中Fail fast与Fail safe迭代器的区别]]></description><category></category></item><item><title><![CDATA[Spark TaskScheduler 功能及源码解析]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/48846129</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/48846129</guid><author>bigdata_wang</author><pubDate>Fri, 09 Oct 2015 19:09:07 +0800</pubDate><description><![CDATA[本文讲的是TaskScheduler的功能设计及具体实现]]></description><category></category></item><item><title><![CDATA[Spark DAGScheduler 功能及源码解析]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/48792317</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/48792317</guid><author>bigdata_wang</author><pubDate>Thu, 01 Oct 2015 11:02:03 +0800</pubDate><description><![CDATA[本文中，威廉将为大家展示DAGScheduler的功能及源码实现]]></description><category></category></item><item><title><![CDATA[Spark Partitioner源码]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/45876369</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/45876369</guid><author>bigdata_wang</author><pubDate>Tue, 22 Sep 2015 00:01:20 +0800</pubDate><description><![CDATA[Spark Partitioner 源码解析]]></description><category></category></item><item><title><![CDATA[Spark运行架构]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/48245581</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/48245581</guid><author>bigdata_wang</author><pubDate>Sun, 06 Sep 2015 21:21:47 +0800</pubDate><description><![CDATA[Spark运行架构及流程]]></description><category></category></item><item><title><![CDATA[Hadoop集群假设]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/48207325</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/48207325</guid><author>bigdata_wang</author><pubDate>Fri, 04 Sep 2015 20:51:37 +0800</pubDate><description><![CDATA[分布式的Hadoop集群构建步骤]]></description><category></category></item><item><title><![CDATA[SSH证书连接配置DEBUG步骤]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/48169165</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/48169165</guid><author>bigdata_wang</author><pubDate>Wed, 02 Sep 2015 00:54:23 +0800</pubDate><description><![CDATA[在设置SSH证书登陆的时候遇到了些问题，在此记录下细节及解决方法]]></description><category></category></item><item><title><![CDATA[第一个Spark程序]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/45768147</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/45768147</guid><author>bigdata_wang</author><pubDate>Sat, 16 May 2015 14:25:33 +0800</pubDate><description><![CDATA[在本文中，威廉将尝试构建我们的第一个Spark程序，并在之前文章中创建的Spark集群里运行起来]]></description><category></category></item><item><title><![CDATA[Spark任务运行详解]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/45691735</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/45691735</guid><author>bigdata_wang</author><pubDate>Wed, 13 May 2015 12:51:10 +0800</pubDate><description><![CDATA[威廉将向大家展示如何在Spark集群上部署运行我们自己的程序]]></description><category></category></item><item><title><![CDATA[Python easy_install 安装使用]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/45678051</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/45678051</guid><author>bigdata_wang</author><pubDate>Tue, 12 May 2015 23:32:20 +0800</pubDate><description><![CDATA[本文将介绍python easy_install工具的安装使用]]></description><category></category></item><item><title><![CDATA[Spark集群架设详解]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/45544097</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/45544097</guid><author>bigdata_wang</author><pubDate>Wed, 06 May 2015 23:46:06 +0800</pubDate><description><![CDATA[跟着威廉学Spark集群架设]]></description><category></category></item><item><title><![CDATA[Facebook好友分布规律R语言分析]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/44535619</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/44535619</guid><author>bigdata_wang</author><pubDate>Sun, 22 Mar 2015 09:47:15 +0800</pubDate><description><![CDATA[社交网络 R语言]]></description><category></category></item><item><title><![CDATA[PageRank算法解析及R语言实现]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/44187691</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/44187691</guid><author>bigdata_wang</author><pubDate>Thu, 12 Mar 2015 00:12:35 +0800</pubDate><description><![CDATA[在PageRank算法出现之前，早期的搜索引擎是根据关键词出现次数对网页进行排序，但这样的算法有严重的缺陷...]]></description><category></category></item><item><title><![CDATA[KNN算法及R语言实现]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/44139125</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/44139125</guid><author>bigdata_wang</author><pubDate>Sun, 08 Mar 2015 22:27:34 +0800</pubDate><description><![CDATA[KNN（k-Nearest Neighbor)分类算法是数据挖掘分类技术中较简单的方法之一。所谓k最近邻，就是k个最近的邻居的意思，说的是每个样本都可以用它最接近的k个邻居来代表。


        例如，上图中，绿色圆要被决定赋予哪个类，是红色三角形还是蓝色四方形？如果K=3，由于红色三角形所占比例为2/3，绿色圆将被赋予红色三角形那个类，如果K=5，由于蓝色四方形比例为3/5，因此绿色]]></description><category></category></item><item><title><![CDATA[朴素贝叶斯分类算法的R语言实现]]></title><link>https://blog.youkuaiyun.com/bigdata_wang/article/details/44115733</link><guid>https://blog.youkuaiyun.com/bigdata_wang/article/details/44115733</guid><author>bigdata_wang</author><pubDate>Sat, 07 Mar 2015 13:18:18 +0800</pubDate><description><![CDATA[贝叶斯分类常用来预测隶属关系，计算一个给定元组属于某一类的概率



首先我们来看下贝叶斯基本公式：




P(B|A)的意思是在A事件的情况下，发生B事件的概率，可以理解为概率论中的条件概率，而贝叶斯公式的巨大作用就是对因果关系进行了交换，通过上面的公式就可以计算P(A|B)的概率，只要通过上述的转换。






朴素贝叶斯分类的正式定义如下：]]></description><category></category></item></channel></rss>