文章目录

一、散点图二、散点图三、趋势分析四、相关性分析

---

一、散点图

代码如下（示例）：

#游戏数据可视化 # 绘制散点图 w <- read.csv("可视化数据.csv", header = F, as.is=T) str(w) head(w) data.names <- c("日期","活跃用户","新增用户","付费率", "付费用户","首次付费用户","ARPPU","ARPU") colnames(w) <- data.names str(w) head(w) #单指标数据可视化 summary(w) w$new.date <- paste(substr(w$日期,6,7),substr(w$日期,9,10),sep = "/") head(w) barplot(height = w$新增用户, names.arg = w$new.date) #还可以进行哪些单指标分析？通过分析单指标，得出相对应的结论。

二、散点图

#双指标数据可视化 plot(w$活跃用户~w$新增用户,col="violetred2",pch=16, main="活跃用户 vs 新增用户散点图") #添加线性拟合直线 abline(lm(w$活跃用户~w$新增用户),col="blue",lwd=2,lty=2) #还可以进行哪些双指标分析？通过分析双指标，得出相对应的结论。 #三指标数据可视化 plot(w$活跃用户~w$新增用户,col="violetred2",pch=16, cex=w$付费率, main="活跃用户 vs 新增用户气泡图")

三、趋势分析

#同比、环比 data <- data.frame(pre=c(113,134,123,145,137,196,187), now=c(129,122,134,149,146,215,208)); ylim.max <- 550 col = c("azure4","brown4") ##将主副标题放到barplot函数里，进行设置 ##font=3,表示字体为“斜体” ##cex=1.5,表示字体放大1.5倍 barplot(as.matrix(rbind( data$pre,data$now)), beside=TRUE,ylim=c(0,ylim.max),col=col,axes=F, main=list( "本周pv趋势分析图",cex=1.5,col="red",font=3), sub=paste("范围：2020.09.28--2020.10.04","\n","网站板块：军事科技"), ylab="网站日页面浏览量pv" ) axis(2) ##画图例 text.legend = c("上周pv","本周pv","pv同比增长","pv环比增长") col2 <- c("black","blue") legend("topleft",pch=c(15,15,16,16), legend=text.legend,cex=0.8,col=c(col,col2),bty="n", horiz=F) ##画刻度标记 text.x <- c("周一","周二","周三","周四","周五","周六","周日") axis(side=1, c(2,5,8,11,14,17,20), labels=text.x,tick=TRUE, cex.axis=0.75) axis(4,at=seq(from=250,length.out=7,by=40),labels=c("-60%","-40%","-20%","0","20%","40%","60%")) ##添加“同比增长曲线”和“环比增长曲线” ##同比增长 = (now[t]-pre[t])/pre[t] same.pre.growth <- (data$now-data$pre)/data$pre ##环比增长 = (now[t]-now[t-1])/now[t-1] ring.growth <- c(NA,diff(data$now)/data$now[1:(length(data$now)-1)]) a<-200; b<-370; lines(c(2,5,8,11,14,17,20), a*same.pre.growth+b,type="b",lwd=2) lines(c(2,5,8,11,14,17,20), a*ring.growth+b,type="b",lwd=2,col="blue") ##在同比和环比曲线上添加文字 j <- 1 for(i in 1:length(data[,1])){ #在bar上画数值 text(3*i-1, a*same.pre.growth[i]+b-5, paste(round(same.pre.growth[i]*10000)/100, "%",sep="")); j <- j+1; text(3*i-1, a*ring.growth[i]+b+5, paste(round(ring.growth[i]*10000)/100, "%",sep=""),col="blue"); j <- j+2; } ##在pv柱状图上添加文字 j <- 1 for(i in 1:length(data[,1])){ text(j+0.5,data$pre[i]+10, data$pre[i], col="deepskyblue4"); j < j+1 #组内间距为1 text(j+1.5,data$now[i]+10, data$now[i], col="deepskyblue4"); j <- j+3 #组外间距为3 }

四、相关性分析

logindata <- read.csv("logindatatest.csv", header = T, fileEncoding = "UTF-8") logindata <- read.csv("abalone.csv", header = T, fileEncoding = "UTF-8") str(logindata) nrow(logindata) dim(logindata) library(caret) dmy<-dummyVars(~.,data=logindata) dmyTsrf<-data.frame(predict(dmy,newdata=logindata)) dim(dmyTsrf) str(dmyTsrf) log.cor <- cor(dmyTsrf, use="pairwise.complete.obs") #install.packages("corrplot") library(corrplot) corrplot.mixed(log.cor[1:9,1:9], lower = "ellipse", upper = "number", tl.pos="lt",diag="u")

# 漏斗分析 # 导入新手教程关键路径用户数 funnel <- read.csv("新手教程路径留存人数统计.csv", header = T, fileEncoding = "UTF-8") funnel <- read.csv("glass.csv", header = T, fileEncoding = "UTF-8") # 查看funnel数据集 funnel # 绘制漏斗图 # 对数据进行降序排序 funnel_order <- funnel[order(funnel$用户数),] # 绘制漏斗图 barplot(funnel_order$用户数,horiz = T,axes=F,border=F,space=0.5, col="steelblue1",xlim=c(-700,700), main="新手教程关键路径漏斗图") barplot(-funnel_order$用户数,horiz=T,add=T,border=F,space=0.5, axes=F,col="steelblue1") # 增加步骤名称 text(x=rep(-600,6),y=seq(1,8.5,length.out = 6), labels=funnel_order$事件行为,cex=0.8,font=3,col="black") # 增加总体的转化率 text(x=rep(0,6),y=seq(1,8.5,length.out = 6), labels=paste0(round(funnel_order$用户数*100/max(funnel_order$用户数),2),"%"), cex=0.8,font=3,col="violetred3") # 增加上一步的转化率 text(x=rep(max(funnel$用户数),6),y=seq(2,7.5,length.out = 5), labels=paste0(round(funnel_order$用户数[1:5]/funnel_order$用户数[2:6]*100,2),"%"), cex=0.8,font=4,col="red3")