In ntels-BI/ntelsBI: Analysis R package for ntels Business Intelligence Team

pacman::p_load(knitr, tidyverse, ggplot2, data.table, mlbench); library(ntelsBI)
opts_chunk$set(fig.path = "output/figure/", fig.align = "center", out.width = "80%", 
               warning = F, message = F)

Install

devtools::install_github("ntels-BI/ntelsBI")

Tutorial

Machine learning

ml() 함수를 이용하여 기계학습 훈련 및 평가를 수행

• CART 를 통한 iris 꽃잎분류

ml(iris, "Species", method = "rpart") %>% fitSummary(type = "cla")

• GLM 을 통한 Sonar 의 Class 분류 (Class = "R" 이 주관심사일 경우)

data(Sonar, package = "mlbench")
ml(Sonar, "Class", method = "glm") %>% fitSummary(type = "cla", positive = "R")

• Random forest 를 통한 mtcars mpg 예측

ml(mtcars, "mpg", method = "rf") %>% fitSummary(type = "reg")

Useful Preprocessing

• spCoordinateTrans() function : 위경도 좌표변환

exdata <- data.frame(long = c(1039197, 1041137, 1039216, 1037176), 
                     lat = c(1926417, 1927056, 1927526, 1924963))

exdata # UTM-K(GRS-80)

spCoordinateTrans(exdata$long, exdata$lat) # To WGS84

• addressToCoord() function : 주소에 대한 좌표변환 (Google Maps API 를 이용)

addressToCoord("강남구청역")

library(leaflet)
library(dplyr)

pd <- data.frame(add = c("강남구청역")) %>%
  mutate(long = addressToCoord(add)[1], lat = addressToCoord(add)[2])

leaflet(data = pd) %>% 
  addTiles() %>%
  addMarkers(~long, ~lat, popup = ~as.character(add), label = ~as.character(add))

License

GPL-3

ntels-BI/ntelsBI documentation built on May 24, 2019, 2:49 p.m.