IMPORT DATA

getwd()

## [1] "/Users/anand/RProjects/KNN"

setwd("/Users/anand/RProjects/KNN")
bankloan <- read.csv("BANK LOAN.csv")
head(bankloan)

##   SN AGE EMPLOY ADDRESS DEBTINC CREDDEBT OTHDEBT DEFAULTER
## 1  1   3     17      12     9.3    11.36    5.01         1
## 2  2   1     10       6    17.3     1.36    4.00         0
## 3  3   2     15      14     5.5     0.86    2.17         0
## 4  4   3     15      14     2.9     2.66    0.82         0
## 5  5   1      2       0    17.3     1.79    3.06         1
## 6  6   3      5       5    10.2     0.39    2.16         0

library(caret)

## Loading required package: lattice

## Loading required package: ggplot2

## Registered S3 methods overwritten by 'ggplot2':
##   method         from 
##   [.quosures     rlang
##   c.quosures     rlang
##   print.quosures rlang

CREATE INDEX TO SPLIT DATA INTO TRAIN AND TEST

index <- createDataPartition(bankloan$SN, p = 0.7,list = FALSE)

**SUBSET THE DATA TO REMOVE IN-ESSENTIAL COLUMNS*8

bankloan2 <- subset(bankloan, select = c(-AGE, -SN, -DEFAULTER))

CREATE SCALED DATA TO FORMULATE TRAIN AND TEST DATA

bankloan3 <- scale(bankloan2)
head(bankloan3)

##       EMPLOY    ADDRESS      DEBTINC    CREDDEBT       OTHDEBT
## 1  1.2933881  0.5452741 -0.140697021  4.63182528  0.5936903584
## 2  0.2420275 -0.3338627  1.031080669 -0.09137366  0.2864682866
## 3  0.9929994  0.8383197 -0.697291424 -0.32753360 -0.2701816059
## 4  0.9929994  0.8383197 -1.078119173  0.52264220 -0.6808249692
## 5 -0.9595275 -1.2129994  1.031080669  0.11172390  0.0005388336
## 6 -0.5089444 -0.4803855 -0.008872031 -0.54952395 -0.2732234086

traindata <- bankloan3[index,]
testdata <- bankloan3[-index,]

CREATE CLASS VECTORS

Ytrain <- bankloan$DEFAULTER[index]
Ytest <- bankloan$DEFAULTER[-index]

FORMULATE MODEL

library(class)
knnmodel <- knn(traindata, testdata, k=round(sqrt(nrow(traindata)),0)-1, cl=Ytrain)
cMatrix <- t(table(Ytest, knnmodel)) ;cMatrix

##         Ytest
## knnmodel   0   1
##        0 138  49
##        1   3  18

LET’S CALCULATE CONFUSION MATRIX, SENSITIVITY, SPECIFICITY

library(caret)
sensitivity(cMatrix)

## [1] 0.9787234

specificity(cMatrix)

## [1] 0.2686567