hr.Rmd

---
title: "Hr"
output: html_notebook
---

#Importing the data
```{r}
hr <- read.csv("HR_comma_sep.csv")
summary(hr)
```

#Checking for missing values
```{r}
(colMeans(is.na(hr)))*100
```
#Data Transformation
#1 Sales
```{r}
hr$sales[hr$sales == "accounting"] <- 0
hr$sales[hr$sales == "hr"] <- 1
hr$sales[hr$sales == "IT"] <- 2
hr$sales[hr$sales == "management"] <- 3
hr$sales[hr$sales == "marketing"] <- 4
hr$sales[hr$sales == "product_mng"] <- 5
hr$sales[hr$sales == "RandD"] <- 6
hr$sales[hr$sales == "sales"] <- 7
hr$sales[hr$sales == "support"] <- 8
hr$sales[hr$sales == "technical"] <- 9
```
#sales: 0-accounting 1-hr 2-IT 3-management 4-marketing 5-product_mng 6-Rand 7-sales 8-support 9-technical 

#2 Salary
```{r}
hr$salary[hr$salary == "low"] <- 0
hr$salary[hr$salary == "medium"] <- 1
hr$salary[hr$salary == "high"] <- 2
```
#salary: 0-low 1-medium 2-high

#Unbalanced Data
```{r}
hr$sales <- as.factor(hr$sales)
hr$salary <- as.factor(hr$salary)
summary(hr$sales)
summary(hr$salary)
```
#Train and Test set
```{r}
library(caret)
set.seed(3456)
trainIndex <- createDataPartition(hr$salary, p = .7,
                                  list = FALSE,
                                  times = 1)
Train <- hr[ trainIndex,]
Test <- hr[-trainIndex,]
```
#70%-Train 30%-Test