Data Mining Concepts and Techniques 3rd 读书笔记(2)

=============第二章:数据预处理*********第二节:描述性统计概述=================

  1. Measuring the Central Tendency
    • 算数平均 arithmetic mean / 加权平均 weighted arithmetic mean:对过大过小值敏感,属于algebraic measure
    • 裁剪后平均 trimmed mean:去掉一定比例的过值后进行平均,属于algebraic measure
    • 中位数 median:适用于非对称分布数据,属于holistic measure,可以通过分组近似估算中间组的中位数。
    • 众数 mode:可以根据含有众数的多少把数据分为 - 无峰值数据no mode,单峰值数据unimodal和多峰值数据multimodal,分布曲线平稳的单峰值数据的众数可以通过经验公式mean-mode = 3*(mean-median)来计算,属于algebraic measure
    • unimodal数据为均匀分布时mean=median=mode,为正倾斜分布(较小值出现的频率高)时mean>median>mode,为负倾斜分布(较大值出现的频率高)时mean<median<mode
    • 中列数 midrange:最大最小值求平均,属于algebraic measure
  2. Measuring the Dispersion of Data
    • 范围 range:最大值减最小值
    • 百分位数 kth percentile:median相当于 50th percentile,常用的还有Q1 - first quartiles 和 Q3 - third quartiles
    • 内距 IQR - interquartile range:Q3-Q1,属于holistic measure,通常1.5倍IQR以外的值可以被认为是outliers
    • 五数概括法 five-number summary:Minimum; Q1; Median; Q3; Maximum
    • 箱线图 Boxplots:大数据情况下暂时没有好的近似算法
    • 方差 Variance / 标准差 Standard Deviation:当把mean作为数据的中心时,标准差可用来计算数据的离散程度,属于algebraic measure
  3. Graphic Displays of Basic Descriptive Data Summaries
    • 直方图 frequency histograms:适用于单变量
    • 分位图 quantile plot:按0.5/n为偏移量,1/n为步长画出n个点的值
    • QQ图 quantile-quantile plot:2个quantile plot进行对比,点数取少的那个,某一个系列quantile值映射到x坐标
    • 散点图 scatter plot; 用来看2个属性之间的关联性
    • 局部回归光滑拟合曲线 loess curve:a(阿发)参数为平滑系数,越大越平滑,但数据失真大,一般取0.25到1之间的值;l(仁达)参数为曲线多项式的次数,通常为1(线性)或者2(2次曲线)

======对数据进行描述性统计可以提供整个数据的概览,方便鉴别数据噪声和过值,为数据清洗做准备========


=============第二章:数据预处理*********第三节:数据清洗=================

  1. Missing Values
    • 直接忽略 Ignore the tuple:通常用在分类算法时记录的某个类别的属性缺失,当缺失数据占比较大时肯定是不可取的做法
    • 手工补录数据 Fill in the missing value manually:oom
    • 全部用一个常量代替 Use a global constant to fill in the missing value:常用的常量是UNKNOWN,不是办法的办法。
    • 用在整体样本上该缺失属性的均值代替 Use the attribute mean to fill in the missing value:
    • 用在同类别样本上该缺失属性的均值代替 Use the attribute mean for all samples belonging to the same class as the given tuple:前提是已经有一个已经明确分好的类别,并且假设这个缺失属性和该类别强相关
    • 用样本空间里的记录的其他属性推断出该缺失属性的值 Use the most probable value to fill in the missing value:最常用或者相对比较合理的清洗方式,采用的推断方式一般是回归分析,决策树和贝叶斯推断

====== ========

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
The increasing volume of data in modern business and science calls for more complex and sophisticated tools. Although advances in data mining technology have made extensive data collection much easier, it's still always evolving and there is a constant need for new techniques and tools that can help us transform this data into useful information and knowledge. Since the previous edition's publication, great advances have been made in the field of data mining. Not only does the third of edition of Data Mining: Concepts and Techniques continue the tradition of equipping you with an understanding and application of the theory and practice of discovering patterns hidden in large data sets, it also focuses on new, important topics in the field: data warehouses and data cube technology, mining stream, mining social networks, and mining spatial, multimedia and other complex data. Each chapter is a stand-alone guide to a critical topic, presenting proven algorithms and sound implementations ready to be used directly or with strategic modification against live data. This is the resource you need if you want to apply today's most powerful data mining techniques to meet real business challenges. * Presents dozens of algorithms and implementation examples, all in pseudo-code and suitable for use in real-world, large-scale data mining projects. * Addresses advanced topics such as mining object-relational databases, spatial databases, multimedia databases, time-series databases, text databases, the World Wide Web, and applications in several fields. *Provides a comprehensive, practical look at the concepts and techniques you need to get the most out of your data

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值