[color=rgba(0, 0, 0, 0.749019607843137)]#加载本次可视化所需包[color=rgba(0, 0, 0, 0.749019607843137)]library(readr) [color=rgba(0, 0, 0, 0.749019607843137)]library(sp) #地图可视化[color=rgba(0, 0, 0, 0.749019607843137)]library(maps) #地图可视化[color=rgba(0, 0, 0, 0.749019607843137)]library(forcats)[color=rgba(0, 0, 0, 0.749019607843137)]library(dplyr)[color=rgba(0, 0, 0, 0.749019607843137)]library(ggplot2)[color=rgba(0, 0, 0, 0.749019607843137)]library(reshape2) [color=rgba(0, 0, 0, 0.749019607843137)]library(ggthemes) #ggplot绘图样式包[color=rgba(0, 0, 0, 0.749019607843137)]library(tidyr)[color=rgba(0, 0, 0, 0.749019607843137)]library(gganimate) #动态图[color=rgba(0, 0, 0, 0.749019607843137)]
! f( W5 n- |# c7 M3 a5 ~[color=rgba(0, 0, 0, 0.749019607843137)]#一、国家名词整理[color=rgba(0, 0, 0, 0.749019607843137)]data<-read_csv('confirmed.csv')[color=rgba(0, 0, 0, 0.749019607843137)]data[data$`Country/Region`=='US',]$`Country/Region`='United States'[color=rgba(0, 0, 0, 0.749019607843137)]data[data$`Country/Region`=='Korea, South',]$`Country/Region`='Korea'[color=rgba(0, 0, 0, 0.749019607843137)]
, @0 p! ^8 B p. Z[color=rgba(0, 0, 0, 0.749019607843137)]information_data<-data[,1:4] #取出国家信息相关数据[color=rgba(0, 0, 0, 0.749019607843137)]inspect_data<-data[,-c(1:4)] #取出确诊人数相关数据[color=rgba(0, 0, 0, 0.749019607843137)]
! Y' g0 q" |1 l0 v- p( q[color=rgba(0, 0, 0, 0.749019607843137)]#二、日期转换[color=rgba(0, 0, 0, 0.749019607843137)]datetime<-colnames(inspect_data)[color=rgba(0, 0, 0, 0.749019607843137)]pastetime<-function(x){[color=rgba(0, 0, 0, 0.749019607843137)] date<-paste0(x,'20')[color=rgba(0, 0, 0, 0.749019607843137)] return(date)[color=rgba(0, 0, 0, 0.749019607843137)]}[color=rgba(0, 0, 0, 0.749019607843137)]datetime1<-as.Date(sapply(datetime,pastetime),format='%m/%d/%Y')[color=rgba(0, 0, 0, 0.749019607843137)]colnames(inspect_data)<-datetime1[color=rgba(0, 0, 0, 0.749019607843137)]2 G- m' V2 F- i- P$ z; ^
[color=rgba(0, 0, 0, 0.749019607843137)]#合并数据,data为累计确诊人数数据(预处理后)[color=rgba(0, 0, 0, 0.749019607843137)]data<-cbind(information_data,inspect_data)[color=rgba(0, 0, 0, 0.749019607843137)]二、新增确诊病例变化趋势#由累计确诊病例计算新增确诊病例- I% m$ Y1 e- \$ f- ~# n, W
inspect_lag_data<-cbind(0,inspect_data[,1
ncol(inspect_data)-1)])* j' U! `4 o* J! s/ k! B
increase_data<-inspect_data-inspect_lag_data. L+ V2 |5 @8 V0 k+ u( R
9 @6 i" x" e0 r. S#合并数据,new_data为新增确诊人数数据4 f f( l- b1 R- M: k
new_data<-cbind(information_data,increase_data)$ }7 u6 `" S4 z" ]
: k& c2 ]+ q& h& h! B1. 中国新增确诊病例变化趋势" t' Q' ]$ T9 h9 q% Q
#合并所有省份新增确诊人数- u4 e6 C1 g0 F# r
china<-new_data[new_data$`Country/Region`=='China',]
- n" p3 F8 c) c! f" V" _8 C% gchina_increase<-data.frame(apply(china[,-c(1:4)],2,sum))
" C6 X& a5 U7 ~, `$ h( Lcolnames(china_increase)<-'increase_patient'8 Q6 ~8 r+ W# h8 r$ _8 J$ }
china_increase$date<-as.Date(rownames(china_increase),format="%Y-%m-%d")
0 n% n; { U" n$ l. {# a
; ^) _2 T" H. @) K& xggplot(china_increase,aes(x=date,y=increase_patient,color='新增确诊人数'))+geom_line(size=1)+
, p; \0 J* P; B5 j8 X9 c scale_x_date(date_breaks = "14 days")+ #设置横轴日期间隔为14天(注意:此时的date列必须为日期格式!)0 s: _ S; ~7 S( Y
labs(x='日期',y='新增确诊人数',title='2020年1月22日-2020年12月7日中国新增确诊人数变化趋势图')+
- T' T0 `2 f# ~8 c5 A, Z theme_economist()+ #使用经济学人绘图样(式ggthemes包)
& y# I: x7 n1 ]3 B7 n& o! e theme(plot.title = element_text(face="plain",size=15,hjust=0.5),
4 Y* E7 W; T! t& S$ P1 E axis.title.x = element_blank(),4 | b' G$ L; \( D- E) K
axis.title.y = element_text(size=15),
! y) G/ Q- A- c axis.text.x = element_text(angle = 90,size=15),; M) F9 n( B6 a. x1 k; f
axis.text.y = element_text(size=15),
3 B& H4 f1 @ a2 n) R, O legend.title=element_blank(),6 X/ c0 J e2 _( n/ p
legend.text=element_text(size=15))
5 s- ^& q* H' T( v( a
" I3 {! v3 q8 q; E- o+ ? 
8 {/ ]5 A5 Q: i% U" n, Y4 n' W, h2. 美国新增病例变化趋势% G O* H1 }8 p$ @0 T1 V! n
us<-new_data[new_data$`Country/Region`=='United States',]
0 m. g/ N0 F0 Hus_increase<-gather(us,key="date",value="increase_patient",'2020-01-22':'2020-12-07')
% d, u$ |$ k( P* P- N$ Pus_increase$date<-as.Date(us_increase$date)
4 Z$ l# a( _! c: n! ]ggplot(us_increase,aes(x=date,y=increase_patient,color='新增确诊人数'))+geom_line(size=1)+. {4 h% U* A% X3 K
scale_x_date(date_breaks = "14 days")+ #设置横轴日期间隔为14天% I# P, A4 j: A* ~+ `, ]
labs(x='日期',y='新增确诊人数',title='2020年1月22日-2020年12月7日美国新增确诊人数变化趋势图')+
0 {( U! Q) b0 T7 \. Z/ m theme_economist()+ #使用经济学人绘图样(式ggthemes包)) M6 U/ N) V7 I6 u1 `- ^8 n
theme(plot.title = element_text(face="plain",size=15,hjust=0.5),
/ c2 W L7 i( I3 p: |. E$ F axis.title.x = element_blank(),6 r) i' z" ?8 b( D- H! \, @, d0 q
axis.title.y = element_text(size=15),) e$ E7 J5 D* y: u7 ]
axis.text.x = element_text(angle = 90,size=15),6 L4 s( z f! E4 m2 |
axis.text.y = element_text(size=15),
' a- N$ W: m; z legend.title=element_blank(),
7 M: }% }& D8 Y; g8 v% h legend.text=element_text(size=15))2 Z9 T3 H% Y7 `2 n; f! W+ s
5 |. e/ ~; g3 M2 S# A) C+ d/ b
- U( b* m$ m) v3 E2 Y5 P
3. 全球新增病例变化趋势5 G' F7 V. O: O5 v. \, D5 x
total_increase<-data.frame(apply(new_data[,-c(1:4)],2,sum)) x; B% E/ f: C( x9 D
colnames(total_increase)<-'increase_patient'
- z0 |2 {1 y: ~' ]# r8 Wtotal_increase$date<-as.Date(rownames(total_increase),format="%Y-%m-%d")
9 n8 E1 K7 H- P% J: j5 k, o' \& Xggplot(total_increase,aes(x=date,y=increase_patient,color='新增确诊人数'))+geom_line(size=1)+3 r- `4 O/ h* ^7 y6 ~
scale_x_date(date_breaks = "14 days")+1 J; w1 c( _* X( ]5 X1 I
labs(x='日期',y='新增确诊人数',title='2020年1月22日-2020年12月7日全球新增确诊人数变化趋势图')+
$ i: c9 \. Z& D( `3 f2 V2 h theme_economist()+3 f, [- c% R- n5 i- m
scale_y_continuous(limits=c(0,8*10^5), #考虑数字过大,以文本形式标注y轴标签
0 }7 i. m) X$ _& G5 n, @* X breaks=c(0,2*10^5,4*10^5,6*10^5,8*10^5),
/ F& C' X6 W; z5 a* b6 ` labels=c("0","20万","40万","60万","80万"))+7 y g b, r! p
theme(plot.title = element_text(face="plain",size=15,hjust=0.5),; U8 L' J% w( f+ G" k$ d
axis.title.x = element_blank(),
5 ]. K* h( [# ` axis.title.y = element_text(size=15),
6 k/ f# S' V! R axis.text.x = element_text(angle = 90,size=15),
! k9 B& Y+ S" B+ J- v0 ` axis.text.y = element_text(size=15),+ j W* L( G4 t/ w% u) D% y
legend.title=element_blank(),
( a$ c3 o9 L$ A- Q% K legend.text=element_text(size=15))( m) ` P' s) z" ~' w7 E" P' B
* Z0 Q, e& i, y. Q, r8 l& B

* l; _4 M- t2 V! [8 ~三、新增确诊病例全球地理分布
+ x5 f M' `1 f3 F' kmapworld<-borders("world",colour = "gray50",fill="white") % H: Q2 n0 I7 O
ggplot()+mapworld+ylim(-60,90)+
$ v) }0 X8 Z% U7 Y0 j2 V+ _5 A, b- } geom_point(aes(x=new_data$Long,y=new_data$Lat,size=new_data$`2020-01-22`),color="darkorange")+
0 a- j9 |9 G+ [. f2 A scale_size(range=c(2,9))+labs(title="2020年1月22日全球新增确诊人数分布")+
: T! \9 p1 o( G# Y% B- } theme_grey(base_size = 15)+7 Z# O' |3 c; `
theme(plot.title=element_text(face="plain",size=15,hjust=0.5),- L3 j6 F( y% Z' D
legend.title=element_blank())6 a2 F( F0 W" T( ?6 F$ T( l
, F1 h) _" A# b! H$ d( o5 X7 J$ Dggplot()+mapworld+ylim(-60,90)+6 Y: {; A8 W z) z" q
geom_point(aes(x=new_data$Long,y=new_data$Lat,size=new_data$`2020-11-22`),color="darkorange")+& t7 h' x& G6 w8 R; e9 W4 c4 b
scale_size(range=c(2,9))+labs(title="2020年11月22日全球新增确诊人数分布")+
6 ~+ V9 H+ a. Q! K% M. P# t6 l theme_grey(base_size = 15)+9 O- L! D8 W! `# n7 D% [/ p% k
theme(plot.title=element_text(face="plain",size=15,hjust=0.5),
+ h0 _) u6 |0 Q" _% W legend.title=element_blank())
4 ]. p3 M8 y) C: c% {9 h
# ]" {6 y G* t
7 K8 n& z# u' O: H8 Z. q+ W/ X
四、累计确诊病例动态变化图1. 至12月7日全球累计病例确诊人数前十国家
! s1 B, _# B* x& gcum_patient<-data[c("Country/Region","2020-12-07")]
cum_patient<-cum_patient[order(cum_patient$`2020-12-07`,decreasing = TRUE),][1:10,]
colnames(cum_patient)<-c("country","count")
cum_patient<-mutate(cum_patient,country = fct_reorder(country, count))
cum_patient$labels<-paste0(as.character(round(cum_patient$count/10^4,0)),"万")
ggplot(cum_patient,aes(x=country,y=count))+
geom_bar(stat = "identity", width = 0.75,fill="#f68060")+
coord_flip()+ #横向
xlab("")+
geom_text(aes(label = labels, vjust = 0.5, hjust = -0.15))+
labs(title='至2020年12月7日累计确诊病例前十的国家')+
theme(plot.title = element_text(face="plain",size=15,hjust=0.5))+
scale_y_continuous(limits=c(0, 1.8*10^7))
& @% L4 ?- x7 [9 G
2. 五国(India、Brazil、Russia、Spain、Italy)累计确诊病例动态变化图
/ V8 q- X1 q0 g u2 mcum_patient_time<-gather(data,key="date",value="increase_patient",'2020-01-22':'2020-12-07')6 p* H% j$ |0 d( ?4 H
colnames(cum_patient_time)<-c("
rovince","Country","Lat","Long","date","increase_patient")
* k6 ] o; g0 E [" o4 nfive_country<-subset(cum_patient_time,Country %in% c("India","Brazil","Russia","Spain","Italy"))5 i% o! F; j! x g$ e9 b1 E
five_country$date<-as.Date(five_country$date)
0 ?1 v# A/ g7 K/ m
( \7 t* S8 f D/ e7 s7 [ggplot(five_country, L8 ]# B0 L3 S+ w- l& m* H* |% `( v# W
aes(x=reorder(Country,increase_patient),y=increase_patient, fill=Country,frame=date)) + , B8 {& d6 _" s3 Q# Q) S! V7 R
geom_bar(stat= 'identity', position = 'dodge',show.legend = FALSE) +
# W# W. s8 h8 m5 W9 H geom_text(aes(label=paste0(increase_patient)),col="black",hjust=-0.2)+
5 A7 y( R5 ~9 p* b3 M5 H scale_fill_brewer(palette='Set3')+ #使用Set3色系模板
& y$ u# X5 G( p, M theme(legend.position="none",
! x. w8 s9 l$ g" K# `' ? panel.background=element_rect(fill='transparent'),0 v: K' v1 q+ l8 R
axis.text.y=element_text(angle=0,colour="black",size=12,hjust=1),0 `- u) c6 z& @' y+ M
panel.grid =element_blank(), #删除网格线
$ ]7 Y0 e: R# z3 ?, a1 X7 o* D axis.text = element_blank(), #删除刻度标签; V* @' m$ S# F% M" @
axis.ticks = element_blank(), #删除刻度线
9 w B3 E! z* |* X. ? )+
" e1 O' p4 G* e3 ]8 G6 x' C coord_flip()+ % f2 p' B1 f D; W7 t
transition_manual(frames=date) + #动态呈现
0 w& p) U. W! O- l, n$ D/ n) U* { labs(title = paste('日期:', '{current_frame}'),x = '', y ='五国累计确诊病例增长')+ * p/ X4 y, B" Y% Z1 X
theme(axis.title.x = element_text(size=15))+* P9 @2 d1 b: G& T1 y$ A' U
ease_aes('linear')
8 A- B4 L, ?" @4 D
( D4 j6 O: D, d) {/ manim_save(filename = "五国累计确诊病例增长动态图.gif") y! H e. x9 P
% N( j. D+ p1 z; U$ k

1 W# Q% T% `- S$ Z) M+ w- W
9 e6 d$ K! I0 H+ v, X# j3 a) ]% ~! j