- 大数据分析有用的资源
- 大数据分析 - 快速指南
- 大数据分析 - 资源
- 大数据分析 - 讨论
大数据分析 - R 简介
本节专门向用户介绍 R 编程语言。R可以从cran网站下载。对于 Windows 用户,安装 rtools和rstudio IDE非常有用。
R背后的一般概念是充当用 C、C++ 和 Fortran 等编译语言开发的其他软件的接口,并为用户提供分析数据的交互式工具。
导航到书籍 zip 文件bda/part2/R_introduction的文件夹并打开R_introduction.Rproj文件。这将打开 RStudio 会话。然后打开 01_vectors.R 文件。逐行运行脚本并按照代码中的注释进行操作。学习的另一个有用选项是仅键入代码,这将帮助您习惯 R 语法。在 R 中,注释是用 # 符号编写的。
为了显示书中R代码的运行结果,在代码评估后,对R返回的结果进行了注释。这样,您就可以复制粘贴书中的代码,并直接在 R 中尝试其中的部分内容。
# Create a vector of numbers numbers = c(1, 2, 3, 4, 5) print(numbers) # [1] 1 2 3 4 5 # Create a vector of letters ltrs = c('a', 'b', 'c', 'd', 'e') # [1] "a" "b" "c" "d" "e" # Concatenate both mixed_vec = c(numbers, ltrs) print(mixed_vec) # [1] "1" "2" "3" "4" "5" "a" "b" "c" "d" "e"
我们来分析一下前面的代码发生了什么。我们可以看到可以用数字和字母创建向量。我们不需要事先告诉 R 我们想要什么类型的数据类型。最后,我们能够创建一个包含数字和字母的向量。向量 mix_vec 已将数字强制转换为字符,我们可以通过可视化值如何打印在引号内来看到这一点。
以下代码显示了函数类返回的不同向量的数据类型。通常使用 class 函数来“询问”一个对象,询问他的类是什么。
### Evaluate the data types using class ### One dimensional objects # Integer vector num = 1:10 class(num) # [1] "integer" # Numeric vector, it has a float, 10.5 num = c(1:10, 10.5) class(num) # [1] "numeric" # Character vector ltrs = letters[1:10] class(ltrs) # [1] "character" # Factor vector fac = as.factor(ltrs) class(fac) # [1] "factor"
R 还支持二维对象。在下面的代码中,有 R 中使用的两种最流行的数据结构的示例:矩阵和 data.frame。
# Matrix M = matrix(1:12, ncol = 4) # [,1] [,2] [,3] [,4] # [1,] 1 4 7 10 # [2,] 2 5 8 11 # [3,] 3 6 9 12 lM = matrix(letters[1:12], ncol = 4) # [,1] [,2] [,3] [,4] # [1,] "a" "d" "g" "j" # [2,] "b" "e" "h" "k" # [3,] "c" "f" "i" "l" # Coerces the numbers to character # cbind concatenates two matrices (or vectors) in one matrix cbind(M, lM) # [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] # [1,] "1" "4" "7" "10" "a" "d" "g" "j" # [2,] "2" "5" "8" "11" "b" "e" "h" "k" # [3,] "3" "6" "9" "12" "c" "f" "i" "l" class(M) # [1] "matrix" class(lM) # [1] "matrix" # data.frame # One of the main objects of R, handles different data types in the same object. # It is possible to have numeric, character and factor vectors in the same data.frame df = data.frame(n = 1:5, l = letters[1:5]) df # n l # 1 1 a # 2 2 b # 3 3 c # 4 4 d # 5 5 e
如前面的示例所示,可以在同一对象中使用不同的数据类型。一般来说,这就是数据在数据库中的呈现方式,API 部分数据是文本或字符向量和其他数字。分析师的工作是确定分配哪种统计数据类型,然后为其使用正确的 R 数据类型。在统计学中,我们通常认为变量有以下类型 -
- 数字
- 名义或绝对
- 序数
在 R 中,向量可以属于以下类别 -
- 数字 - 整数
- 因素
- 有序因子
R 为每种统计类型的变量提供了一种数据类型。然而,有序因子很少使用,但可以通过函数因子创建或有序。
以下部分讨论索引的概念。这是一个非常常见的操作,处理选择对象的部分并对它们进行转换的问题。
# Let's create a data.frame df = data.frame(numbers = 1:26, letters) head(df) # numbers letters # 1 1 a # 2 2 b # 3 3 c # 4 4 d # 5 5 e # 6 6 f # str gives the structure of a data.frame, it’s a good summary to inspect an object str(df) # 'data.frame': 26 obs. of 2 variables: # $ numbers: int 1 2 3 4 5 6 7 8 9 10 ... # $ letters: Factor w/ 26 levels "a","b","c","d",..: 1 2 3 4 5 6 7 8 9 10 ... # The latter shows the letters character vector was coerced as a factor. # This can be explained by the stringsAsFactors = TRUE argumnet in data.frame # read ?data.frame for more information class(df) # [1] "data.frame" ### Indexing # Get the first row df[1, ] # numbers letters # 1 1 a # Used for programming normally - returns the output as a list df[1, , drop = TRUE] # $numbers # [1] 1 # # $letters # [1] a # Levels: a b c d e f g h i j k l m n o p q r s t u v w x y z # Get several rows of the data.frame df[5:7, ] # numbers letters # 5 5 e # 6 6 f # 7 7 g ### Add one column that mixes the numeric column with the factor column df$mixed = paste(df$numbers, df$letters, sep = ’’) str(df) # 'data.frame': 26 obs. of 3 variables: # $ numbers: int 1 2 3 4 5 6 7 8 9 10 ... # $ letters: Factor w/ 26 levels "a","b","c","d",..: 1 2 3 4 5 6 7 8 9 10 ... # $ mixed : chr "1a" "2b" "3c" "4d" ... ### Get columns # Get the first column df[, 1] # It returns a one dimensional vector with that column # Get two columns df2 = df[, 1:2] head(df2) # numbers letters # 1 1 a # 2 2 b # 3 3 c # 4 4 d # 5 5 e # 6 6 f # Get the first and third columns df3 = df[, c(1, 3)] df3[1:3, ] # numbers mixed # 1 1 1a # 2 2 2b # 3 3 3c ### Index columns from their names names(df) # [1] "numbers" "letters" "mixed" # This is the best practice in programming, as many times indeces change, but variable names don’t # We create a variable with the names we want to subset keep_vars = c("numbers", "mixed") df4 = df[, keep_vars] head(df4) # numbers mixed # 1 1 1a # 2 2 2b # 3 3 3c # 4 4 4d # 5 5 5e # 6 6 6f ### subset rows and columns # Keep the first five rows df5 = df[1:5, keep_vars] df5 # numbers mixed # 1 1 1a # 2 2 2b # 3 3 3c # 4 4 4d # 5 5 5e # subset rows using a logical condition df6 = df[df$numbers < 10, keep_vars] df6 # numbers mixed # 1 1 1a # 2 2 2b # 3 3 3c # 4 4 4d # 5 5 5e # 6 6 6f # 7 7 7g # 8 8 8h # 9 9 9i