chap5 R中其它特定场景的数据可视化包

作者：带土；Daitu; Adam 邮箱：2505131775@qq.com

在程序开头设置所有绘图风格

library(ggplot2)
theme_set(theme_bw(base_family = "STKaiti"))

5.1 igraph包可视化

社交网络图，邻接矩阵可视化

library(igraph)

## 
## Attaching package: 'igraph'

## The following objects are masked from 'package:stats':
## 
##     decompose, spectrum

## The following object is masked from 'package:base':
## 
##     union

library(igraphdata)

## 1:通过数据表格获得网络图，并可视化不同网络布局的网络图
## 读取数据
edge <- read.csv("data/edge.csv")
vertex <- read.csv("data/vertex.csv")

## Country:国家，airportnumber：机场数量，vtype：节点的类型
head(vertex)

##     Country airportnumber vtype
## 1   Algeria            43     2
## 2 Australia           296     3
## 3   Bahamas            33     2
## 4   Belgium            24     2
## 5    Brazil           234     3
## 6    Canada           417     3

## Country.x,Country.y :连线的两个点，connectnumber：连接的数量， etype：边的类型
head(edge)

##   Country.y     Country.x connectnumber etype
## 1   Algeria        France            67     1
## 2 Australia   New Zealand            64     1
## 3   Bahamas United States            72     1
## 4   Belgium         Spain            60     1
## 5    Brazil United States            57     1
## 6    Canada United States           364     3

## 从数据表中定义网络图
g <- graph_from_data_frame(edge, ## 边的数据
                           vertices = vertex, ## 节点数据
                           directed = FALSE ## 表示图像为无向图
                           )

## 对得到的图像进行进一步的修整
## 根据数据中的其它变量，定义边的粗细，为添加边的粗细特征
E(g)$width <- floor(log10(E(g)$connectnumber))
##  定义图像中会用到的颜色和形状向量，节点和边的颜色
colrs <- c("gray", "red", "blue")
shape <- c( "circle","csquare","sphere")
## 为图中的节点定义颜色,alpha.f为设置颜色的透明程度
V(g)$color <- adjustcolor(colrs[V(g)$vtype],alpha.f = 0.5)
## 为图中的边定义颜色
E(g)$color <- colrs[E(g)$etype]
# plot 4个图 - 2 rows, 2 columns，每个图使用不同的图像样式
par(mfrow=c(2,2), mar=c(0,0,1,0),family = "STKaiti") 
plot(g, 
     ## 使用圆形布局
     layout =  layout_in_circle(g),
     ## 节点的大小和形状
     vertex.size = floor(10*log10(V(g)$airportnumber)),
     vertex.shape = shape[V(g)$vtype],
     ## 节点标签的缩放比例
     vertex.label.cex = 0.6,vertex.label.color = "yellow",
     main = "圆形布局")
plot(g, layout =  layout_with_fr(g),
     ## 节点的大小和形状
     vertex.size = floor(10*log10(V(g)$airportnumber)),
     vertex.shape = shape[V(g)$vtype],
     ## 节点标签的缩放比例
     vertex.label.cex = 0.6,vertex.label.color = "yellow",
     main = "弹力模型布局")

plot(g, layout =  layout_on_sphere(g),
     ## 节点的大小和形状
     vertex.size = floor(10*log10(V(g)$airportnumber)),
     vertex.shape = shape[V(g)$vtype],
     ## 节点标签的缩放比例
     vertex.label.cex = 0.6,vertex.label.color = "yellow",
     main = "球体表面布局")
plot(g, layout =  layout_randomly(g),
     ## 节点的大小和形状
     vertex.size = floor(10*log10(V(g)$airportnumber)),
     vertex.shape = shape[V(g)$vtype],
     ## 节点标签的缩放比例
     vertex.label.cex = 0.6,vertex.label.color = "yellow",
     main = "随机布局")

##  直接可视化网络数据集
data("karate")
## karate 空手道俱乐部网络
head(karate)

## 6 x 34 sparse Matrix of class "dgCMatrix"

##    [[ suppressing 34 column names 'Mr Hi', 'Actor 2', 'Actor 3' ... ]]

##                                                                            
## Mr Hi   . 4 5 3 3 3 3 2 2 . 2 3 1 3 . . . 2 . 2 . 2 . . . . . . . . . 2 . .
## Actor 2 4 . 6 3 . . . 4 . . . . . 5 . . . 1 . 2 . 2 . . . . . . . . 2 . . .
## Actor 3 5 6 . 3 . . . 4 5 1 . . . 3 . . . . . . . . . . . . . 2 2 . . . 2 .
## Actor 4 3 3 3 . . . . 3 . . . . 3 3 . . . . . . . . . . . . . . . . . . . .
## Actor 5 3 . . . . . 2 . . . 3 . . . . . . . . . . . . . . . . . . . . . . .
## Actor 6 3 . . . . . 5 . . . 3 . . . . . 3 . . . . . . . . . . . . . . . . .

## 可视化时，对网络数据的显示图像进行调整
# 生成节点的颜色和形状
V(karate)$color <- colrs[V(karate)$color]
## 单独设置主管和教练节点的颜色
V(karate)[V(karate)$label %in% c("H","A")]$color <- "green"
## 调整节点的形状
V(karate)$shape <- "circle"
V(karate)[V(karate)$label %in% c("H","A")]$shape <- "sphere"
## 设置节点的大小
V(karate)$size <- 15+degree(karate)
## 设置边的粗细
E(karate)$width <- E(karate)$weight
## 可视化调整后的网络图
set.seed(1234)
plot(karate,layout = layout.fruchterman.reingold)

## 通过连接创建小的网络图
plot(graph_from_literal( A--B, C--D, E--F--G--H, I, J--A--D--K),
     vertex.size = 20)

## 有向的网络图
g3 <- graph_from_literal( Alice +-+ Bob --+ Cecil +-- Daniel,
                     Eugene --+ Gordon:Helen )
plot(g3)

## 通过邻接矩阵可视化网络图
set.seed(123)
## 使用随机数生成一个邻接矩阵
adjm <- matrix(sample(0:1, 100, replace=TRUE,
                      prob=c(0.8,0.2)), nc=10)
colnames(adjm) <- letters[1:10]
rownames(adjm) <- letters[1:10]
set.seed(123)
# plot 4个图 - 2 rows, 2 columns，每个图使用网络声称模式
par(mfrow=c(2,2), mar=c(0,0,1,0),family = "STKaiti") 
plot(graph_from_adjacency_matrix(adjm,
                                 ## 有向网络
                                 mode = "directed"),
     main = 'mode = "directed"',
     layout =  layout_with_fr)

plot(graph_from_adjacency_matrix(adjm,
                                 ## 无向网络"undirected"或"max"
                                 mode = "undirected"),
     main = 'mode = "undirected"',
     layout =  layout_with_fr)
plot(graph_from_adjacency_matrix(adjm,
                                 ## 无向网络，只使用上三角的数据
                                 mode = "upper"),
     main = 'mode = "upper"',
     layout =  layout_with_fr)
plot(graph_from_adjacency_matrix(adjm,
                                 ## 无向网络，只使用下三角的数据
                                 mode = "lower"),
     main = 'mode = "lower"',
     layout =  layout_with_fr)

## 网络图生成邻接矩阵
karmat <- as_adjacency_matrix(karate,type = "both")
karmat

## 34 x 34 sparse Matrix of class "dgCMatrix"

##    [[ suppressing 34 column names 'Mr Hi', 'Actor 2', 'Actor 3' ... ]]

##                                                                             
## Mr Hi    . 1 1 1 1 1 1 1 1 . 1 1 1 1 . . . 1 . 1 . 1 . . . . . . . . . 1 . .
## Actor 2  1 . 1 1 . . . 1 . . . . . 1 . . . 1 . 1 . 1 . . . . . . . . 1 . . .
## Actor 3  1 1 . 1 . . . 1 1 1 . . . 1 . . . . . . . . . . . . . 1 1 . . . 1 .
## Actor 4  1 1 1 . . . . 1 . . . . 1 1 . . . . . . . . . . . . . . . . . . . .
## Actor 5  1 . . . . . 1 . . . 1 . . . . . . . . . . . . . . . . . . . . . . .
## Actor 6  1 . . . . . 1 . . . 1 . . . . . 1 . . . . . . . . . . . . . . . . .
## Actor 7  1 . . . 1 1 . . . . . . . . . . 1 . . . . . . . . . . . . . . . . .
## Actor 8  1 1 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
## Actor 9  1 . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 . 1 1
## Actor 10 . . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
## Actor 11 1 . . . 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . .
## Actor 12 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
## Actor 13 1 . . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
## Actor 14 1 1 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
## Actor 15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1
## Actor 16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1
## Actor 17 . . . . . 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . . .
## Actor 18 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
## Actor 19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1
## Actor 20 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
## Actor 21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1
## Actor 22 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
## Actor 23 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1
## Actor 24 . . . . . . . . . . . . . . . . . . . . . . . . . 1 . 1 . 1 . . 1 1
## Actor 25 . . . . . . . . . . . . . . . . . . . . . . . . . 1 . 1 . . . 1 . .
## Actor 26 . . . . . . . . . . . . . . . . . . . . . . . 1 1 . . . . . . 1 . .
## Actor 27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 . . . 1
## Actor 28 . . 1 . . . . . . . . . . . . . . . . . . . . 1 1 . . . . . . . . 1
## Actor 29 . . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 . 1
## Actor 30 . . . . . . . . . . . . . . . . . . . . . . . 1 . . 1 . . . . . 1 1
## Actor 31 . 1 . . . . . . 1 . . . . . . . . . . . . . . . . . . . . . . . 1 1
## Actor 32 1 . . . . . . . . . . . . . . . . . . . . . . . 1 1 . . 1 . . . 1 1
## Actor 33 . . 1 . . . . . 1 . . . . . 1 1 . . 1 . 1 . 1 1 . . . . . 1 1 1 . 1
## John A   . . . . . . . . 1 1 . . . 1 1 1 . . 1 1 1 . 1 1 . . 1 1 1 1 1 1 1 .

5.2 地图数据可视化

无坐标数据的地图数据可视化

library(ggplot2)
library(treemap)
library(rworldmap)

## Loading required package: sp

## ### Welcome to rworldmap ###

## For a short introduction type :   vignette('rworldmap')

## 2014年世界上所有国家的国民总收入和人口数据
data("GNI2014")
head(GNI2014)

##   iso3          country     continent population    GNI
## 3  BMU          Bermuda North America      67837 106140
## 4  NOR           Norway        Europe    4676305 103630
## 5  QAT            Qatar          Asia     833285  92200
## 6  CHE      Switzerland        Europe    7604467  88120
## 7  MAC Macao SAR, China          Asia     559846  76270
## 8  LUX       Luxembourg        Europe     491775  75990

GNI2014$log2GNI <- log2(GNI2014$GNI)
head(GNI2014)

##   iso3          country     continent population    GNI  log2GNI
## 3  BMU          Bermuda North America      67837 106140 16.69561
## 4  NOR           Norway        Europe    4676305 103630 16.66108
## 5  QAT            Qatar          Asia     833285  92200 16.49248
## 6  CHE      Switzerland        Europe    7604467  88120 16.42718
## 7  MAC Macao SAR, China          Asia     559846  76270 16.21883
## 8  LUX       Luxembourg        Europe     491775  75990 16.21352

## 可视化世界地图，根据人口分布来配色
## 1:将国家数据和地图根据ISO3联系再一起
Map1 <- joinCountryData2Map(GNI2014, 
                            ## 根据iso3数据中的编码来匹配区域
                            joinCode = "ISO3",
                            nameJoinColumn = "iso3")

## 184 codes from your data successfully matched countries in the map
## 4 codes from your data failed to match with a country code in the map
## 59 codes from the map weren't represented in your data

## 2:将国家数据和地图根据名称联系再一起 
Map2 <- joinCountryData2Map(GNI2014, 
                            ## 根据名称数据中的编码来匹配区域
                            joinCode = "NAME",
                            nameJoinColumn = "country")

## 184 codes from your data successfully matched countries in the map
## 4 codes from your data failed to match with a country code in the map
## 59 codes from the map weren't represented in your data

## 使用Map1数据可视化地图
p1 <- mapCountryData(mapToPlot = Map1,
               ## 想要可视化的数据，人口
               nameColumnToPlot = "population",
               addLegend=FALSE,## 先不显示图例
               ## 根据数据的多少设置颜色
               catMethod = "pretty",colourPalette="rainbow",
               ## 定义海洋的颜色，和没匹配数据国家的颜色
               oceanCol = "steelblue1",missingCountryCol = "white",
               mapTitle = "Population for Country") 
##为图像添加图例，并对齐进行调整
do.call(addMapLegend,c(p1,legendLabels="all", 
                       ## 颜色条的宽度和
                       legendWidth=0.5,
                       ## 根据数据划分图例的颜色条
                       legendIntervals="data",
                       ## 图例离图像下边界的距离
                       legendMar = 4))

## 可视化局部地区的图像，使用曲对数后的GNI数据
p2 <- mapCountryData(mapToPlot = Map2, ##  使用根据国家匹配的数据
               ## 想要可视化的数据，人口
               nameColumnToPlot = "log2GNI",
               addLegend=FALSE,## 先不显示图例
               ## 根据数据的多少设置颜色
               catMethod = "fixedWidth",colourPalette="heat",
               ## 定义海洋的颜色，和没匹配数据国家的颜色
               oceanCol = "steelblue1",missingCountryCol = "white",
               mapRegion="asia",
               mapTitle = "GNI for Country") 
##为图像添加图例，并对齐进行调整
do.call(addMapLegend,c(p2,legendLabels="all", 
                       ## 颜色条的宽度和
                       legendWidth=0.5,
                       ## 根据数据划分图例的颜色条
                       legendIntervals="data",
                       ## 图例离图像下边界的距离
                       legendMar = 4,
                       ## 垂直的图例
                       horizontal=FALSE))

## 使用点的大小来可视化国家的人口多少
par(mar = c(0,0,0,0))
p3 <- mapBubbles(Map1,
                 # 点的大小
                 nameZSize = "population",
                 ## 点的颜色
                 nameZColour = "continent",
                 colourPalette = "rainbow",
                 oceanCol = "steelblue1",
                 missingCountryCol = "white",
                 landCol="wheat",pch=21,
                 mapRegion = "world")

5.3 pheatmap包可视化热力图

详细介绍pheatmap包对数据进行可视化和可视化相关系数的包

library(pheatmap)
library(ggcorrplot)
library(corrplot)

## corrplot 0.84 loaded

library(readxl)
## 读取数据
heatdata <- read.csv("data/heatmapdata2.csv")
rownames(heatdata) <- heatdata$data
heatdata$data <- NULL
head(heatdata)

##                Raw        Soak   Sonicate  Cellulase     Steam  Sonsteam
## sample1  1.3921110  0.70391718  0.5831566  0.5599809 0.3163506 0.5672277
## sample2  0.8527846 -0.43475544  0.5831566  0.4206362 0.3163506 0.5672277
## sample3  0.8527846  0.64561536 -1.7423985 -0.1312379 0.3163506 0.5672277
## sample4 -0.9271987 -0.58854932  0.5831566  0.4206362 0.3163506 0.5672277
## sample5 -0.3488467 -3.01213477  0.5831566  0.4206362 0.3818094 0.4333861
## sample6 -0.4101890 -0.02442258  1.4620044  1.7227105 0.3163506 0.5672277
##         Cellsteam     Roast   Sonroast  Cellroast
## sample1 0.2956562 0.8187462  0.6450521  0.5485523
## sample2 0.2956562 0.8187462  0.6450521  0.5485523
## sample3 0.2956562 0.8187462  0.6450521  0.5485523
## sample4 0.2956562 0.4555785  0.6450521  0.5485523
## sample5 1.5183751 0.7784026  0.6450521 -1.3296666
## sample6 0.2956562 0.6138959 -1.0411406  0.5485523

## 可视化相关系数热力图
heat_cor <- cor(heatdata)
## 使用corrplot包可视化相关系数热力图
par(family = "STKaiti")
corrplot(heat_cor,
         ## "circle", "square", "ellipse", "number", "pie"等
         method = "ellipse",
         ## "full", "lower", "upper"
         type = "full",
         title = "corrplot",
         diag = TRUE,
         mar = c(0, 0, 2, 0),
         ## 调整文本标签，变量的名称
         tl.cex = 0.8,tl.col = "black",
         tl.offset = 0.5, tl.srt = 60,
         ## lt:left,top;ld：left,diagonal;td;d;n:none;
         tl.pos = "lt",
         ## 调整颜色条,r:right;b:bttom;n:none;
         cl.pos = "r",cl.cex = 0.8)

## 使用corrplot.mixed函数可视化相关系数热力图
par(family = "STKaiti")
corrplot.mixed(heat_cor,
               ## 下三角使用数字，上三角使用圆
               lower = "number", upper = "circle",
               ## 调整文本标签，变量的名称
               tl.col="black",tl.pos = "d",
               tl.cex = 0.8,tl.srt = 60,
               ## 调整数字的显示大小
               number.cex = 0.8,
               mar = c(0, 0, 2, 0),
               main = "相关系数热力图",
               lower.col = "black")

使用pheatmap函数可视化热力图

## 对数据使用默认的参数可视化热力图
pheatmap(heatdata)

## 将数据反转可视化热力图
pheatmap(t(heatdata))

## 对热力图进行调整，不进行聚类
pheatmap(t(heatdata),cluster_rows = FALSE,cluster_cols = FALSE,
         main = "heatmap")

## 调整横纵坐标标签字体的大小
pheatmap(t(heatdata),cluster_rows = FALSE,cluster_cols = FALSE,
         main = "heatmap",
         fontsize_row = 10, fontsize_col = 8)

## 显示出数字
pheatmap(heatdata,cluster_rows = FALSE,cluster_cols = FALSE,
         main = "heatmap",
         fontsize_row = 7, fontsize_col = 10,
         ## 对数字进行调整
         display_numbers = TRUE, number_format = "%.2f",
         fontsize_number = 5)

## 固定每个单元的大小
pheatmap(heatdata,cluster_rows = FALSE,cluster_cols = FALSE,
         main = "heatmap",
         fontsize_row = 7, fontsize_col = 10,
         ## 对数字进行调整
         display_numbers = TRUE, number_format = "%.2f",
         fontsize_number = 6,
         cellwidth = 20, cellheight = 8)

##  在聚类热力图中对聚类结果进行划分
pheatmap(t(heatdata),cluster_rows = FALSE,
         main = "heatmap",
         fontsize_row = 10, fontsize_col = 8,
         cluster_cols = TRUE,cutree_cols = 4,
         gaps_col = TRUE)

## 为热力图的行或列生成注释，通常适合数据有分组数据的情况
ann_row <- data.frame(lab = as.factor(rep(c("lab1","lab2"),c(5,5))))
rownames(ann_row) <- colnames(heatdata)
head(ann_row)

##            lab
## Raw       lab1
## Soak      lab1
## Sonicate  lab1
## Cellulase lab1
## Steam     lab1
## Sonsteam  lab2

## 设置列标签的数据分组情况
ann_col <- data.frame(group = as.factor(rep(c("G1","G2","G3"),
                                          c(20,15,8))))
rownames(ann_col) <- rownames(heatdata)
head(ann_col)

##         group
## sample1    G1
## sample2    G1
## sample3    G1
## sample4    G1
## sample5    G1
## sample6    G1

pheatmap(t(heatdata),cluster_rows = FALSE,cluster_cols = FALSE,
         main = "heatmap",
         fontsize_row = 10, fontsize_col = 8,
         annotation_row = ann_row,annotation_col = ann_col)

## 调整热力图的颜色
pheatmap(t(heatdata),cluster_rows = FALSE,cluster_cols = FALSE,
         ## 颜色插值函数
         color = colorRampPalette(
           c("navy", "white","firebrick3"))(50),
         main = "heatmap",
         fontsize_row = 10, fontsize_col = 8,
         annotation_row = ann_row,annotation_col = ann_col)

pheatmap(t(heatdata),cluster_rows = FALSE,cluster_cols = TRUE,
         ## 颜色插值函数
         color = colorRampPalette(
           c("red", "white", "blue"))(100),
         main = "heatmap",
         fontsize_row = 10, fontsize_col = 8,
         annotation_row = ann_row,annotation_col = ann_col)

## 5.4 可视化雷达图

## 数据准备
library(readr)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:igraph':
## 
##     as_data_frame, groups, union

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(climatol)

## Loading required package: maps

## Loading required package: mapdata

library(ggplot2)
## 读取上海一个月份的天气数据
weather <- read_csv("data/上海天气数据.csv")

## Parsed with column specification:
## cols(
##   province = col_character(),
##   city = col_character(),
##   town = col_character(),
##   temperature = col_double(),
##   relative_humidity = col_double(),
##   rainfall = col_double(),
##   wind_direction = col_double(),
##   wind_strong = col_double(),
##   day = col_date(format = ""),
##   hour = col_character(),
##   year = col_double(),
##   month = col_double(),
##   days = col_character()
## )

head(weather)

## # A tibble: 6 x 13
##   province city  town  temperature relative_humidi… rainfall wind_direction
##   <chr>    <chr> <chr>       <dbl>            <dbl>    <dbl>          <dbl>
## 1 上海     上海  嘉定           11               96        0             92
## 2 上海     上海  嘉定           11               96        0             81
## 3 上海     上海  嘉定            8               94        0            103
## 4 上海     上海  嘉定            8               95        0             79
## 5 上海     上海  嘉定           10               97        0             76
## 6 上海     上海  嘉定            8               95        0             92
## # … with 6 more variables: wind_strong <dbl>, day <date>, hour <chr>,
## #   year <dbl>, month <dbl>, days <chr>

summary(weather$wind_direction)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##     0.0    77.0   256.0   200.5   302.0   359.0

##  将风向划分为16个区间，代表16个方向
## 切分方向
dirbreak <- seq(-12.25,360,22.5)
## 代表的方向
direction <-c("N","NNE","NE","ENE","E","ESE","SE","SSE",
              "S","SSW","SW","WSW","W","WNW","NW","NNW")
## 调整角度
index <- weather$wind_direction > 347.25
weather$wind_direction[index] <- weather$wind_direction[index] - 360
summary(weather$wind_direction)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  -12.00   66.75  237.00  186.17  295.00  347.00

weather$wind_direction2 <- cut(weather$wind_direction,
                                breaks = dirbreak,
                                labels = direction,
                                include.lowest = TRUE)
summary(weather$wind_direction2)

##    N  NNE   NE  ENE    E  ESE   SE  SSE    S  SSW   SW  WSW    W  WNW   NW  NNW 
##  592  587  541  606  553  288  187  129  130  135  161  394  842 1157  933  685

## 将风力的强弱划分为4各等级
weather$wind_strong2 <- cut_interval(weather$wind_strong,4)

## 计算不同风向和等级对应下出现的次数
shwind <- as.data.frame.array(table(weather$wind_strong2,
                                    weather$wind_direction2))
shwind

##              N NNE  NE ENE   E ESE  SE SSE   S SSW  SW WSW   W WNW  NW NNW
## [0,3.25]   476 478 469 552 483 244 165 124 123 132 147 381 677 794 621 510
## (3.25,6.5] 102  89  62  49  69  41  22   5   7   3  13  12 146 312 271 151
## (6.5,9.75]  12  17  10   5   1   3   0   0   0   0   1   1  15  43  38  22
## (9.75,13]    2   3   0   0   0   0   0   0   0   0   0   0   4   8   3   2

## 可视化风向雷达图
par(family = "STKaiti")
rosavent(shwind,
         ## 参考圆的数量
         fnum=4, 
         ## 两个个参考员之间的百分比
         fint=4, ang=-3*pi/16, 
         margen=c(0,0,2,0),
         ## 填充使用的颜色
         col=rainbow(4,0.5,0.92,start=0.1,end=0.9),
         main="一个月实时风向雷达图")

## 针对其它数据可视化雷达图,使用fmsb包
library(fmsb)

## 数据准备
## 数据表的前两行，必须时每个变量取值的最大值和最小值
maxmin <- data.frame(A = c(10,0),B = c(10,0),C = c(10,0),
                     D = c(10,0),E = c(5,0),DE = c(10,0))
one <- data.frame(A = 8,B = 6,C = 10,D = 5,E = 3, DE = 7)
two <- data.frame(A = 4,B = 8,C = 7,D = 7,E = NA, DE = 10)
three <- data.frame(A = 6,B = 9,C = 3,D = 10,E = 4, DE = 8)
## 将数据连接到一起
plotdata <- rbind(maxmin,one,two,three)
plotdata

##    A  B  C  D  E DE
## 1 10 10 10 10  5 10
## 2  0  0  0  0  0  0
## 3  8  6 10  5  3  7
## 4  4  8  7  7 NA 10
## 5  6  9  3 10  4  8

radarchart(plotdata,
           axistype = 0, ## 坐标轴的样式可以0:5
           seg = 5, ## 每个坐标轴网格的切分情况
           ## 颜色可以直接指定，也可以使用数字
           pcol = c("red","blue","green"),
           plty = 1,
           ## 线的粗细
           plwd = 1,
           ## 设置多边形的填充
           pdensity = c(30,30,30),
           pangle = c(45,90,135),
           pfcol = c("red","blue","green"),
           ## 设置雷达图的网格
           cglty = 3,
           cglcol = "lightblue")

radarchart(plotdata,
           axistype = 2, ## 坐标轴的样式可以0:5
           seg = 5, ## 每个坐标轴网格的切分情况
           pcol = 1:3,
           ## 线的类型，支持1:8，重复循环使用
           plty = 3:5,
           plwd = 2,
           ## 设置坐标轴和数值的颜色
           axislabcol = "black",
           title = "plot data",
           ## 缺失值是否进行插值，默认为FALSE
           na.itp = TRUE,
           ## 中心是否从0开始
           centerzero = TRUE,
           ## 调整变量的名称，默认为数据表的变量名
           vlabels = c("apple","banana","watermelon",
                       "pear","orange\njuice"),
           vlcex = 0.7)

## 针对数据表的前两行不指定最大最和最小值的数据绘制雷达图
data("HairEyeColor")
HairEyeColor

## , , Sex = Male
## 
##        Eye
## Hair    Brown Blue Hazel Green
##   Black    32   11    10     3
##   Brown    53   50    25    15
##   Red      10   10     7     7
##   Blond     3   30     5     8
## 
## , , Sex = Female
## 
##        Eye
## Hair    Brown Blue Hazel Green
##   Black    36    9     5     2
##   Brown    66   34    29    14
##   Red      16    7     7     7
##   Blond     4   64     5     8

## 变量为性别和眼睛的颜色，行表示头发颜色
HairEyeColor <- as.data.frame.array(HairEyeColor)
HairEyeColor

##       Brown.Male Blue.Male Hazel.Male Green.Male Brown.Female Blue.Female
## Black         32        11         10          3           36           9
## Brown         53        50         25         15           66          34
## Red           10        10          7          7           16           7
## Blond          3        30          5          8            4          64
##       Hazel.Female Green.Female
## Black            5            2
## Brown           29           14
## Red              7            7
## Blond            5            8

## 可视化雷达图
rownames(HairEyeColor)

## [1] "Black" "Brown" "Red"   "Blond"

radarchart(HairEyeColor,
           axistype = 2, ## 坐标轴的样式可以0:5
           seg = 4, ## 每个坐标轴网格的切分情况
           ## 颜色可以直接指定，也可以使用数字
           pcol = c("black","brown","red","yellow"),
           ## 线的粗细
           plwd = 2,plty = 1,
           maxmin = FALSE)

5.5 可视化韦恩图，集合数据的可视化方式。

library(VennDiagram)

## Loading required package: grid

## Loading required package: futile.logger

library(UpSetR)
## 数据准备
one  <- 1:100
two <- seq(1,200,by = 2)
three <- seq(10,300,by = 5)
four <- seq(2,400,by = 4)
five <- seq(10,500,by = 10)

vcol <- c("red","blue","green","DeepPink","Gold")
## 将数据列表绘制为韦恩图,图片格式默认为tiff
fig1 <- venn.diagram(list(A = one,B = two,C = three,
                         D = four,E = five),
                    filename = NULL,
                    ## 设置标题
                    main.fontfamily = "STKaiti",
                    main = "韦恩图",sub = "sub title",
                    sub.pos = c(0.1,1),
                    lwd = 1,##线宽
                    fill = vcol,alpha = 0.5,## 填充颜色
                    margin = 0.1)

grid.newpage()
grid.draw(fig1)

grid.newpage()
## 可视化两个集合的韦恩图，
fig2 <- draw.pairwise.venn(area1 = 10,area2 = 10,cross.area = 5,
                          category = c("one","two"),
                          col = c("red","blue"),
                          fill = c("red","blue"),alpha = 0.6)
grid.newpage()
grid.draw(fig2)

## 韦恩图最多可以可视化5个集合的交集和并集的情况，而upset包可以分析更多个集合之间的关系

## 使用upset包可视化集合之间的交集、并集的情况
## 数据准备
six <- seq(3,400,by = 3)
## 1: 将5个集合的并集就算出来，
all <- unique(c(one,two,three,four,five,six))
## 建立一个数据表格
plotdata <- data.frame(matrix(nrow = length(all),ncol = 7))
colnames(plotdata) <-c("element","one","two","three","four",
                       "five","six")
## 2:数据表第一列是5个集合的并集的所有元素
plotdata[,1] <- all
## 3:其它列中的对应行，如果包含那一行的元素，则取值为1，否则取值为0
for (i in 1:length(all)) {
  plotdata[i,2] <- ifelse(all[i] %in% one,1,0)
  plotdata[i,3] <- ifelse(all[i] %in% two,1,0)
  plotdata[i,4] <- ifelse(all[i] %in% three,1,0)
  plotdata[i,5] <- ifelse(all[i] %in% four,1,0)
  plotdata[i,6] <- ifelse(all[i] %in% five,1,0)
  plotdata[i,7] <- ifelse(all[i] %in% six,1,0)
}
## 查看数据
head(plotdata)

##   element one two three four five six
## 1       1   1   1     0    0    0   0
## 2       2   1   0     0    1    0   0
## 3       3   1   1     0    0    0   1
## 4       4   1   0     0    0    0   0
## 5       5   1   1     0    0    0   0
## 6       6   1   0     0    1    0   1

## 可视化6个集合的交并情况
upset(plotdata,
      sets = c("one","two","three","four","five","six"),
      nintersects = 40, ## 默认显示前40个交集
      order.by = "freq", ## 根据频数排序
      ## 设置主条形图
      matrix.color  = "blue", ## 数据矩阵的颜色
      main.bar.color = "red",# 主要条形图的颜色
      mainbar.y.label = "Intersection Size",
      ##设置集合条形图
      sets.bar.color = "green",
      sets.x.label = "Set size",
      ## 设置矩阵点图
      point.size  = 3,line.size = 0.5,
      ## 矩阵点图和条形图的比例
      mb.ratio = c(0.65, 0.35))

## 条形图生序排放
upset(plotdata,
      sets = c("one","two","three","four","five","six"),
      nintersects = 40, ## 默认显示前40个交集
      order.by = "freq", ## 根据频数排序
      ## 矩阵点图和条形图的比例
      mb.ratio = c(0.65, 0.35),
      ## 条形图生序排放,并且不显示数字
      decreasing = FALSE,show.numbers = FALSE,
      ## 阴影的颜色
      shade.color= "lightblue",shade.alpha = 0.5,
      matrix.dot.alpha = 0.5,
      ## 对交集的变换,"identity", "log10", "log2"
      scale.intersections = "identity",
      ## 对集和数量的变换,"identity", "log10", "log2"
      scale.sets = "log2",
      ## 对文本进行缩放
      text.scale = 1.3)

5.6 可视化树图

树图和可交互树图

library(treemap)
library(d3treeR)
data(GNI2014)
head(GNI2014)

##   iso3          country     continent population    GNI
## 3  BMU          Bermuda North America      67837 106140
## 4  NOR           Norway        Europe    4676305 103630
## 5  QAT            Qatar          Asia     833285  92200
## 6  CHE      Switzerland        Europe    7604467  88120
## 7  MAC Macao SAR, China          Asia     559846  76270
## 8  LUX       Luxembourg        Europe     491775  75990

## 使用树图可视化每个国家GNI数值的大小
treemap(GNI2014,index = "country",vSize = "GNI")

## 通过人口来的多少进行颜色填充
treemap(GNI2014,index = "country",vSize = "GNI",
        ##  设置填充颜色，默认根据index的取值进行填充
        vColor = "population",type = "value",
        title = "World GNI and population")

## 改变填充颜色的调色盘
treemap(GNI2014,index = "country",vSize = "GNI",
        ##  设置填充颜色，默认根据index的取值进行填充
        vColor = "population",type = "value",
        palette = "RdYlBu",
        title = "World GNI and population")

## 对数据使用两个分组变量进行索引
treemap(GNI2014,index = c("continent","country"),
        vSize = "population", ## 大小
        ##  设置填充颜色，默认根据index的取值进行填充
        vColor = "GNI",type = "value",
        palette = "RdYlBu",
        title = "World GNI and population",
        n = 10, ##  连续数值切分的份数
        ## 图例位置"bottom", "right", or "none"
        position.legend = "right")

## 使用d3treeR包绘制可交互的树图
## 1:先使用treemap函数绘制树图
tmap <- treemap(GNI2014,index = c("continent","country"),
                vSize = "population", ## 大小
                vColor = "GNI",type = "value")
## 模式1
d3tree(tmap,rootname = "World GNI and population")

## 模式2
d3tree2(tmap,rootname = "World GNI and population")

## 模式3
d3tree3(tmap,rootname = "World GNI and population")

5.7 可视化马赛克图

2个变量的马赛克图、多个变量的马赛克图

data("HairEyeColor")
HairEyeColor[,,1]

##        Eye
## Hair    Brown Blue Hazel Green
##   Black    32   11    10     3
##   Brown    53   50    25    15
##   Red      10   10     7     7
##   Blond     3   30     5     8

data("Titanic")

## 2个变量的马赛克图
## 1 : 使用mosaicplot {graphics}函数
## 可视化二维列联表
mosaicplot(HairEyeColor[,,1],main = "mosaicplot",
           ## 使用灰度渐变色
           color = TRUE)

## 可视化高维列联表，多个变量的马赛克图
mosaicplot(HairEyeColor,main = "mosaicplot",
           shade = TRUE)

## 使用vcd包可视化马赛克图
HairEyeColor[,,1]

##        Eye
## Hair    Brown Blue Hazel Green
##   Black    32   11    10     3
##   Brown    53   50    25    15
##   Red      10   10     7     7
##   Blond     3   30     5     8

library(vcd)

## 
## Attaching package: 'vcd'

## The following object is masked from 'package:fmsb':
## 
##     oddsratio

vcd::mosaic(HairEyeColor[,,1],shade = TRUE,legend = TRUE,cex = 0.8)

5.7 文本数据可视化

可视化词云

library(wordcloud)

## Loading required package: RColorBrewer

library(wordcloud2)
## 使用wordcloud2包中的中文词频数据用于展示
data("demoFreqC")
head(demoFreqC)

##         V2   V1
## 1     数据 2304
## 3     统计 1413
## 4     用户  855
## 5     模型  846
## 7     分析  773
## 8 数据分析  750

## 使用wordcloud包可视化数据
## 1: textplot()函数在指定的位置可视化文本
set.seed(1234)
x <- 1:10
y <- 1:10
par(family = "STKaiti")
wordcloud::textplot(x,y,demoFreqC$V2[1:10])

## 2:wordcloud()函数可视化词云
par(family = "STKaiti")
wordcloud(words = demoFreqC$V2,freq = demoFreqC$V1,
          scale=c(4,.5), ## 可视化时词的大小范围
          min.freq=50, ## 词频小于min.freq的不可视化
          max.words=200 ## 最多可视化200个词
          )

## 设置可视化词云时的颜色
pal <- brewer.pal(10,"BrBG")
par(family = "STKaiti")
wordcloud(words = demoFreqC$V2,freq = demoFreqC$V1,
          scale=c(4,.5), ## 可视化时词的大小范围
          min.freq=50, ## 词频小于min.freq的不可视化
          max.words=200, ## 最多可视化200个词
          colors = pal)

## 可视化分组数据的对比词云
## 数据准备，生成数据框
groupdata <- data.frame(groupA = demoFreqC$V1,
                        groupB = rev(demoFreqC$V1))
rownames(groupdata) <- demoFreqC$V2
head(groupdata)

##          groupA groupB
## 数据       2304     38
## 统计       1413     38
## 用户        855     38
## 模型        846     38
## 分析        773     38
## 数据分析    750     38

par(family = "STKaiti")
comparison.cloud(groupdata,scale=c(4,.5),max.words=100,  
                 ## 设置颜色
                 colors=brewer.pal(2,"Dark2"),
                 ## 设置标题
                 title.size=2,title.colors="blue")

## Warning in brewer.pal(2, "Dark2"): minimal value for n is 3, returning requested palette with 3 different levels

## 增加数据的组数，可视化多组数据的对比图像
set.seed(123)
groupdata$groupC <- demoFreqC$V1[sample(nrow(groupdata))]
set.seed(1)
groupdata$groupD <- demoFreqC$V1[sample(nrow(groupdata))]
par(family = "STKaiti")
comparison.cloud(groupdata,scale=c(4,.5),max.words=50,  
                 ## 设置颜色
                 colors=brewer.pal(4,"Set1"),
                 ## 设置标题
                 title.size=1.5,title.colors="black")

## 使用wordcloud2包可视化可交互的词云

## 1:letterCloud()函数根据字母的形状可视化词云
data("demoFreq")
head(demoFreq)

##          word freq
## oil       oil   85
## said     said   73
## prices prices   48
## opec     opec   42
## mln       mln   31
## the       the   26

letterCloud(demoFreq,word = "R",wordSize = 1,
            ## 颜色
            color ='random-dark',backgroundColor = "snow")

## 2:letterCloud()函数根据单词可视化词云
letterCloud(demoFreq,word = "Data",wordSize = 4)

## 3:wordcloud2函数根据字母的形状可视化词云
## 圆形词云
wordcloud2(demoFreq,size = 1, ## 字体大小
           ## 设置字体
           fontFamily = "Segoe UI", fontWeight = "bold",
           ## 设置颜色,'random-dark' 或者 'random-light' 
           color = "random-dark", backgroundColor = "white",
           ## 设置形状, 默认圆形
           shape = "circle")

## 心形
wordcloud2(demoFreqC,size = 1, ## 字体大小
           ## 设置字体
           fontFamily = "STKaiti",
           ## 设置颜色,'random-dark' 或者 'random-light' 
           color = "random-light", backgroundColor = "grey",
           ## 设置形状, 心形
           shape = "cardioid")

## 菱形，设置文字的倾斜角度
wordcloud2(demoFreqC[1:200,],size = 1, ## 字体大小
           ## 设置字体
           fontFamily = "STKaiti",
           ## 设置颜色,'random-dark' 或者 'random-light' 
           color = "random-light", backgroundColor = "white",
           ## 设置形状, 心形
           shape = "diamond",
           ## 文字的倾斜角度
           minRotation = -pi/6, maxRotation = -pi/6,
           rotateRatio = 1)

## 五角星形
wordcloud2(demoFreqC[1:100,],size = 1, ## 字体大小
           ## 设置字体
           fontFamily = "STKaiti",
           ## 设置颜色,'random-dark' 或者 'random-light' 
           color = "random-light", backgroundColor = "white",
           ## 设置形状, 五角星形
           shape = "star")

## 根据指定的图片可视化词云五角星形
figpaths <- "data/带土2.jpg"
wordcloud2(demoFreqC,size = 1, ## 字体大小
           ## 设置字体
           fontFamily = "STKaiti",
           figPath = figpaths,color = "lightblue")

figpaths <- "data/带土.jpg"
wordcloud2(demoFreqC,size = 1, ## 字体大小
           ## 设置字体
           fontFamily = "STKaiti",
           figPath = figpaths)