KNN

Data <- read.csv("D:\\DataScience\\Assignments\\KNN\\Zoo.csv")

View(Data)

table(Data$domestic)

## 
##  0  1 
## 88 13

# Replace 0 with Wild animal/Game animal and 1 with Domestic Animal. 

Data$domestic <- factor(Data$domestic, levels = c(0,1), labels = c("Wild Animal/Game Animal","Domestic Animal"))

# table or proportation of enteries in the datasets. What % of entry is Wild/Game animal and % of entry is Domestic Anima
round(prop.table(table(Data$domestic))*100,1)

## 
## Wild Animal/Game Animal         Domestic Animal 
##                    87.1                    12.9

Data1 <- Data[,-c(1,16)]

norm <- function(x){return((x-min(x))/(max(x)-min(x)))}

Data_n <-as.data.frame(lapply(Data1[1:16],norm))

View(Data_n)

#create training and test datasets

Data_Train <- Data_n[1:81,]
Data_Test <- Data_n[82:101,]

Data_Train_Lables <- Data[1:81,16]
Data_Test_Lables <- Data[82:101,16]

# Build a KNN model on taining dataset
library("class")
library("caret")

## Warning: package 'caret' was built under R version 3.5.1

## Loading required package: lattice

## Warning: package 'lattice' was built under R version 3.5.1

## Loading required package: ggplot2

# Building the KNN model on training dataset and also need labels which we are including c1. Once we build the preduction model
# we have to test on test dataset


Data_pred <- knn(train = Data_Train, test = Data_Test, cl = Data_Train_Lables,k=5)


class(Data_Train)

## [1] "data.frame"

class(Data_Test)

## [1] "data.frame"

x <- as.data.frame(Data_pred)

x

##                  Data_pred
## 1  Wild Animal/Game Animal
## 2  Wild Animal/Game Animal
## 3          Domestic Animal
## 4  Wild Animal/Game Animal
## 5  Wild Animal/Game Animal
## 6  Wild Animal/Game Animal
## 7  Wild Animal/Game Animal
## 8  Wild Animal/Game Animal
## 9  Wild Animal/Game Animal
## 10 Wild Animal/Game Animal
## 11 Wild Animal/Game Animal
## 12 Wild Animal/Game Animal
## 13 Wild Animal/Game Animal
## 14 Wild Animal/Game Animal
## 15 Wild Animal/Game Animal
## 16 Wild Animal/Game Animal
## 17 Wild Animal/Game Animal
## 18 Wild Animal/Game Animal
## 19 Wild Animal/Game Animal
## 20         Domestic Animal

x <- cbind(x,Data_Test_Lables)

x

##                  Data_pred        Data_Test_Lables
## 1  Wild Animal/Game Animal Wild Animal/Game Animal
## 2  Wild Animal/Game Animal Wild Animal/Game Animal
## 3          Domestic Animal Wild Animal/Game Animal
## 4  Wild Animal/Game Animal Wild Animal/Game Animal
## 5  Wild Animal/Game Animal Wild Animal/Game Animal
## 6  Wild Animal/Game Animal Wild Animal/Game Animal
## 7  Wild Animal/Game Animal Wild Animal/Game Animal
## 8  Wild Animal/Game Animal Wild Animal/Game Animal
## 9  Wild Animal/Game Animal Wild Animal/Game Animal
## 10 Wild Animal/Game Animal Wild Animal/Game Animal
## 11 Wild Animal/Game Animal Wild Animal/Game Animal
## 12 Wild Animal/Game Animal Wild Animal/Game Animal
## 13 Wild Animal/Game Animal Wild Animal/Game Animal
## 14 Wild Animal/Game Animal Wild Animal/Game Animal
## 15 Wild Animal/Game Animal Wild Animal/Game Animal
## 16 Wild Animal/Game Animal Wild Animal/Game Animal
## 17 Wild Animal/Game Animal Wild Animal/Game Animal
## 18 Wild Animal/Game Animal Wild Animal/Game Animal
## 19 Wild Animal/Game Animal Wild Animal/Game Animal
## 20         Domestic Animal Wild Animal/Game Animal

## Now evualuation the model performance

#install.packages("gmodels")
library("gmodels")

## Warning: package 'gmodels' was built under R version 3.5.1

# Create cross table of predicted and actual
CrossTable( x = Data_Test_Lables, y = Data_pred)

## 
##  
##    Cell Contents
## |-------------------------|
## |                       N |
## |         N / Table Total |
## |-------------------------|
## 
##  
## Total Observations in Table:  20 
## 
##  
##                         | Data_pred 
##        Data_Test_Lables | Wild Animal/Game Animal |         Domestic Animal |               Row Total | 
## ------------------------|-------------------------|-------------------------|-------------------------|
## Wild Animal/Game Animal |                      18 |                       2 |                      20 | 
##                         |                   0.900 |                   0.100 |                         | 
## ------------------------|-------------------------|-------------------------|-------------------------|
##            Column Total |                      18 |                       2 |                      20 | 
## ------------------------|-------------------------|-------------------------|-------------------------|
## 
##