Assignment 3
Eklavya Attar
26 July 2017
Problem Definition :
The objective is to predict whether a patient has diabetes based on diagnostic measurements.
Setup
Adding Libraries
library(tidyr)
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(ggplot2)
library(corrgram)
library(gridExtra)##
## Attaching package: 'gridExtra'## The following object is masked from 'package:dplyr':
##
## combinelibrary(rms)## Loading required package: Hmisc## Loading required package: lattice## Loading required package: survival## Loading required package: Formula##
## Attaching package: 'Hmisc'## The following object is masked from 'package:gridExtra':
##
## combine## The following objects are masked from 'package:dplyr':
##
## combine, src, summarize## The following objects are masked from 'package:base':
##
## format.pval, round.POSIXt, trunc.POSIXt, units## Loading required package: SparseM##
## Attaching package: 'SparseM'## The following object is masked from 'package:base':
##
## backsolvelibrary(Deducer)## Loading required package: JGR## Loading required package: rJava## Loading required package: JavaGD## Loading required package: iplots##
## Please type JGR() to launch console. Platform specific launchers (.exe and .app) can also be obtained at http://www.rforge.net/JGR/files/.## Loading required package: car##
## Attaching package: 'car'## The following object is masked from 'package:rms':
##
## vif## The following object is masked from 'package:dplyr':
##
## recode## Loading required package: MASS##
## Attaching package: 'MASS'## The following object is masked from 'package:dplyr':
##
## select##
##
## Note Non-JGR console detected:
## Deducer is best used from within JGR (http://jgr.markushelbig.org/).
## To Bring up GUI dialogs, type deducer().library(caret)##
## Attaching package: 'caret'## The following object is masked from 'package:survival':
##
## clusterFunctions to determine outliers and NA vaues
Dataset
Reading the data from a file
dfrDiabModel <- read.csv("./data/diabetes-train.csv", header=T, stringsAsFactors=F)
head(dfrDiabModel)## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 6 148 72 35 0 33.6
## 2 1 85 66 29 0 26.6
## 3 8 183 64 0 0 23.3
## 4 1 89 66 23 94 28.1
## 5 0 137 40 35 168 43.1
## 6 5 116 74 0 0 25.6
## DiabetesPedigreeFunction Age Outcome
## 1 0.627 50 1
## 2 0.351 31 0
## 3 0.672 32 1
## 4 0.167 21 0
## 5 2.288 33 1
## 6 0.201 30 0Trying to find missing data after imputation
lapply(dfrDiabModel, FUN=detect_na)## $Pregnancies
## [1] 0
##
## $Glucose
## [1] 0
##
## $BloodPressure
## [1] 0
##
## $SkinThickness
## [1] 0
##
## $Insulin
## [1] 0
##
## $BMI
## [1] 0
##
## $DiabetesPedigreeFunction
## [1] 0
##
## $Age
## [1] 0
##
## $Outcome
## [1] 0Observation:
1. Hence we see that there are no NA values
Trying to find outliers Data after imputation
lapply(dfrDiabModel, FUN=detect_outliers)## $Pregnancies
## integer(0)
##
## $Glucose
## integer(0)
##
## $BloodPressure
## [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## $SkinThickness
## integer(0)
##
## $Insulin
## [1] 543 846 495 485 495 478 744 680 545 465 579 474 480 600 540 480
##
## $BMI
## [1] 0.0 0.0 0.0 0.0 0.0 67.1 0.0 0.0 0.0 0.0 0.0
##
## $DiabetesPedigreeFunction
## [1] 2.288 1.893 1.781 2.329 2.137 1.731 2.420 1.699 1.698
##
## $Age
## [1] 81
##
## $Outcome
## integer(0)Observation:
1. Hence we see that there are outliers in our data.
2. For this particular case we will impute the data by replacing the outliers value with the average values
***********************************Imputing Data***********************************************
1. BMI Values
Plotting the Graph
qplot(data = dfrDiabModel, x = dfrDiabModel$BMI) + ylab("Levels of BMI")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Observation:
1. Therefore we see that we have outliers value at 0 and at BMI values greater than 50
Replacing the Outliers Value with the average value
dfrDiabModel$BMI <- ifelse(dfrDiabModel$BMI==0 |dfrDiabModel$BMI> 50,NA,dfrDiabModel$BMI)
head(dfrDiabModel)## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 6 148 72 35 0 33.6
## 2 1 85 66 29 0 26.6
## 3 8 183 64 0 0 23.3
## 4 1 89 66 23 94 28.1
## 5 0 137 40 35 168 43.1
## 6 5 116 74 0 0 25.6
## DiabetesPedigreeFunction Age Outcome
## 1 0.627 50 1
## 2 0.351 31 0
## 3 0.672 32 1
## 4 0.167 21 0
## 5 2.288 33 1
## 6 0.201 30 0dfrDiabModel$BMI[is.na(dfrDiabModel$BMI)] <- round(mean(dfrDiabModel$BMI[!is.na(dfrDiabModel$BMI)]),digits=0)
dfrDiabModel$BMI <- as.integer(dfrDiabModel$BMI)
detect_na(dfrDiabModel$BMI)## [1] 0#head(dfrDiabModel)2. BloodPressure
Plotting the Graph
qplot(data = dfrDiabModel, x = dfrDiabModel$BloodPressure) + ylab("Levels of BloodPressure")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Observation: 1. Therefore we see that we have outliers values less than 40 and values greater than 110
Replacing the Outliers Value with the average value
dfrDiabModel$BloodPressure <- ifelse(dfrDiabModel$BloodPressure<40 |dfrDiabModel$BloodPressure>110,NA,dfrDiabModel$BloodPressure)
head(dfrDiabModel)## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 6 148 72 35 0 33
## 2 1 85 66 29 0 26
## 3 8 183 64 0 0 23
## 4 1 89 66 23 94 28
## 5 0 137 40 35 168 43
## 6 5 116 74 0 0 25
## DiabetesPedigreeFunction Age Outcome
## 1 0.627 50 1
## 2 0.351 31 0
## 3 0.672 32 1
## 4 0.167 21 0
## 5 2.288 33 1
## 6 0.201 30 0dfrDiabModel$BloodPressure[is.na(dfrDiabModel$BloodPressure)] <- round(mean(dfrDiabModel$BloodPressure[!is.na(dfrDiabModel$BloodPressure)]),digits=0)
dfrDiabModel$BloodPressure <- as.integer(dfrDiabModel$BloodPressure)
detect_na(dfrDiabModel$BloodPressure)## [1] 0#head(dfrDiabModel)3. Insulin
Plotting the Graph
qplot(data = dfrDiabModel, x = dfrDiabModel$Insulin) + ylab("Levels of Insulin")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Observation:
1. Therefore we see that we have outliers values greater than 400
Replacing the Outliers Value with the average value
dfrDiabModel$Insulin <- ifelse(dfrDiabModel$Insulin> 400,NA,dfrDiabModel$Insulin)
head(dfrDiabModel)## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 6 148 72 35 0 33
## 2 1 85 66 29 0 26
## 3 8 183 64 0 0 23
## 4 1 89 66 23 94 28
## 5 0 137 40 35 168 43
## 6 5 116 74 0 0 25
## DiabetesPedigreeFunction Age Outcome
## 1 0.627 50 1
## 2 0.351 31 0
## 3 0.672 32 1
## 4 0.167 21 0
## 5 2.288 33 1
## 6 0.201 30 0dfrDiabModel$Insulin[is.na(dfrDiabModel$Insulin)] <- round(mean(dfrDiabModel$Insulin[!is.na(dfrDiabModel$Insulin)]),digits=0)
dfrDiabModel$Insulin <- as.integer(dfrDiabModel$Insulin)
detect_na(dfrDiabModel$Insulin)## [1] 0#head(dfrDiabModel)4. DiabetesPedigreeFunction
Plotting the Graph
qplot(data = dfrDiabModel, x = dfrDiabModel$DiabetesPedigreeFunction) + ylab("Levels of DiabetesPedigreeFunction")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Observation:
1. Therefore we see that we have outliers values greater than 1.5
Replacing the Outliers Value with the average value
dfrDiabModel$DiabetesPedigreeFunction <- ifelse(dfrDiabModel$DiabetesPedigreeFunction>1.5 ,NA,dfrDiabModel$DiabetesPedigreeFunction)
head(dfrDiabModel)## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 6 148 72 35 0 33
## 2 1 85 66 29 0 26
## 3 8 183 64 0 0 23
## 4 1 89 66 23 94 28
## 5 0 137 40 35 168 43
## 6 5 116 74 0 0 25
## DiabetesPedigreeFunction Age Outcome
## 1 0.627 50 1
## 2 0.351 31 0
## 3 0.672 32 1
## 4 0.167 21 0
## 5 NA 33 1
## 6 0.201 30 0Impute Data
dfrDiabModel$DiabetesPedigreeFunction[is.na(dfrDiabModel$DiabetesPedigreeFunction)] <- mean(dfrDiabModel$DiabetesPedigreeFunction[!is.na(dfrDiabModel$DiabetesPedigreeFunction)])
dfrDiabModel$DiabetesPedigreeFunction <- as.numeric(dfrDiabModel$DiabetesPedigreeFunction)
detect_na(dfrDiabModel$DiabetesPedigreeFunction)## [1] 0#head(dfrDiabModel)Trying to find missing Data
lapply(dfrDiabModel, FUN=detect_na)## $Pregnancies
## [1] 0
##
## $Glucose
## [1] 0
##
## $BloodPressure
## [1] 0
##
## $SkinThickness
## [1] 0
##
## $Insulin
## [1] 0
##
## $BMI
## [1] 0
##
## $DiabetesPedigreeFunction
## [1] 0
##
## $Age
## [1] 0
##
## $Outcome
## [1] 0Observation:
1. Hence we see that there are no NA values
Trying to find outliers Data
lapply(dfrDiabModel, FUN=detect_outliers)## $Pregnancies
## integer(0)
##
## $Glucose
## integer(0)
##
## $BloodPressure
## integer(0)
##
## $SkinThickness
## integer(0)
##
## $Insulin
## integer(0)
##
## $BMI
## integer(0)
##
## $DiabetesPedigreeFunction
## numeric(0)
##
## $Age
## [1] 81
##
## $Outcome
## integer(0)Observation:
1. Hence we see that there are no outliers in our data after the imputation operation.
Correlation
vctCorr = numeric(0)
for (i in names(dfrDiabModel)){
cor.result <- cor(as.numeric(dfrDiabModel$Outcome), as.numeric(dfrDiabModel[,i]))
vctCorr <- c(vctCorr, cor.result)
}
dfrCorr <- vctCorr
names(dfrCorr) <- names(dfrDiabModel)
dfrCorr## Pregnancies Glucose BloodPressure
## 0.2278834 0.4593730 0.1650356
## SkinThickness Insulin BMI
## 0.0862186 0.1203176 0.3055303
## DiabetesPedigreeFunction Age Outcome
## 0.1860203 0.2265074 1.0000000Data For Visualization
dfrDiabGraph <- gather(dfrDiabModel, variable, value, -Outcome)
head(dfrDiabGraph)## Outcome variable value
## 1 1 Pregnancies 6
## 2 0 Pregnancies 1
## 3 1 Pregnancies 8
## 4 0 Pregnancies 1
## 5 1 Pregnancies 0
## 6 0 Pregnancies 5Data Visualization
ggplot(dfrDiabGraph) +
geom_jitter(aes(value,Outcome, colour=variable)) +
facet_wrap(~variable, scales="free_x") +
labs(title="Relation Of Outcome With Other Features")
Summary
lapply(dfrDiabModel, FUN=summary)## $Pregnancies
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 1.000 3.000 3.827 6.000 17.000
##
## $Glucose
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0 99.0 116.0 120.5 140.5 199.0
##
## $BloodPressure
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 40.00 64.00 72.00 72.38 80.00 110.00
##
## $SkinThickness
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 0.00 23.00 20.41 32.00 99.00
##
## $Insulin
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 0.00 37.00 67.24 115.00 375.00
##
## $BMI
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 18.00 27.00 32.00 31.61 36.00 50.00
##
## $DiabetesPedigreeFunction
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0780 0.2400 0.3750 0.4539 0.6125 1.4760
##
## $Age
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 21.00 24.00 29.00 33.13 40.00 81.00
##
## $Outcome
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.3448 1.0000 1.0000Find Best Multi Logistic Model
Choose the best logistic model by using step().
stpDiabModel=step(glm(data=dfrDiabModel, formula=Outcome~., family=binomial), trace=0, steps=100)
summary(stpDiabModel)##
## Call:
## glm(formula = Outcome ~ Pregnancies + Glucose + BMI + DiabetesPedigreeFunction,
## family = binomial, data = dfrDiabModel)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.6934 -0.7366 -0.4020 0.7396 2.8487
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -9.041522 0.745678 -12.125 < 2e-16 ***
## Pregnancies 0.140305 0.028466 4.929 8.27e-07 ***
## Glucose 0.033625 0.003498 9.613 < 2e-16 ***
## BMI 0.093970 0.016265 5.777 7.59e-09 ***
## DiabetesPedigreeFunction 1.292168 0.336558 3.839 0.000123 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 900.53 on 698 degrees of freedom
## Residual deviance: 661.04 on 694 degrees of freedom
## AIC: 671.04
##
## Number of Fisher Scoring iterations: 5Observation:
1.Best results given by Outcome ~ Pregnancies + Glucose + BMI + DiabetesPedigreeFunction
Making the Final Multi Linear Model
# make model
mLRDModel <- glm(data=dfrDiabModel, formula=Outcome~Pregnancies+Glucose+BMI+DiabetesPedigreeFunction, family=binomial(link="logit"))
# print summary
summary(mLRDModel)##
## Call:
## glm(formula = Outcome ~ Pregnancies + Glucose + BMI + DiabetesPedigreeFunction,
## family = binomial(link = "logit"), data = dfrDiabModel)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.6934 -0.7366 -0.4020 0.7396 2.8487
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -9.041522 0.745678 -12.125 < 2e-16 ***
## Pregnancies 0.140305 0.028466 4.929 8.27e-07 ***
## Glucose 0.033625 0.003498 9.613 < 2e-16 ***
## BMI 0.093970 0.016265 5.777 7.59e-09 ***
## DiabetesPedigreeFunction 1.292168 0.336558 3.839 0.000123 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 900.53 on 698 degrees of freedom
## Residual deviance: 661.04 on 694 degrees of freedom
## AIC: 671.04
##
## Number of Fisher Scoring iterations: 5Checking for R2 value
Diablrm <- lrm(formula = Outcome ~ Pregnancies + Glucose + BMI + DiabetesPedigreeFunction,data=dfrDiabModel,x=TRUE, y=TRUE)
print(Diablrm)## Logistic Regression Model
##
## lrm(formula = Outcome ~ Pregnancies + Glucose + BMI + DiabetesPedigreeFunction,
## data = dfrDiabModel, x = TRUE, y = TRUE)
##
## Model Likelihood Discrimination Rank Discrim.
## Ratio Test Indexes Indexes
## Obs 699 LR chi2 239.49 R2 0.401 C 0.835
## 0 458 d.f. 4 g 1.776 Dxy 0.670
## 1 241 Pr(> chi2) <0.0001 gr 5.908 gamma 0.671
## max |deriv| 3e-08 gp 0.298 tau-a 0.303
## Brier 0.154
##
## Coef S.E. Wald Z Pr(>|Z|)
## Intercept -9.0415 0.7457 -12.13 <0.0001
## Pregnancies 0.1403 0.0285 4.93 <0.0001
## Glucose 0.0336 0.0035 9.61 <0.0001
## BMI 0.0940 0.0163 5.78 <0.0001
## DiabetesPedigreeFunction 1.2922 0.3366 3.84 0.0001
## Observation:
1.We find that R2 value is 0.401 which means this particular model has a moderate strength
Confusion Matrix
PrdProbValue <- predict(mLRDModel, type='response')
PrdOutLevels <- ifelse(PrdProbValue > 0.5, 1, 0)
cnfmtrx <- table(Predicted=PrdOutLevels, ActUAL=dfrDiabModel$Outcome)
confusionMatrix(cnfmtrx)## Confusion Matrix and Statistics
##
## ActUAL
## Predicted 0 1
## 0 403 103
## 1 55 138
##
## Accuracy : 0.774
## 95% CI : (0.7411, 0.8045)
## No Information Rate : 0.6552
## P-Value [Acc > NIR] : 5.519e-12
##
## Kappa : 0.4749
## Mcnemar's Test P-Value : 0.0001847
##
## Sensitivity : 0.8799
## Specificity : 0.5726
## Pos Pred Value : 0.7964
## Neg Pred Value : 0.7150
## Prevalence : 0.6552
## Detection Rate : 0.5765
## Detection Prevalence : 0.7239
## Balanced Accuracy : 0.7263
##
## 'Positive' Class : 0
## Observation:
1. We find that accuracy of this model comes out to be 0.774
Regression Data
dfrDiabPlot <- mutate(dfrDiabModel, PrdProbValue=PrdProbValue, PrdOutLevels=PrdOutLevels)
head(dfrDiabPlot)## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 6 148 72 35 0 33
## 2 1 85 66 29 0 26
## 3 8 183 64 0 0 23
## 4 1 89 66 23 94 28
## 5 0 137 40 35 168 43
## 6 5 116 74 0 0 25
## DiabetesPedigreeFunction Age Outcome PrdProbValue PrdOutLevels
## 1 0.627000 50 1 0.66551782 1
## 2 0.351000 31 0 0.04123596 0
## 3 0.672000 32 1 0.77969800 1
## 4 0.167000 21 0 0.04471700 0
## 5 0.453894 33 1 0.54792737 1
## 6 0.201000 30 0 0.13817377 0Regression Visulaization
#dfrPlot
ggplot(dfrDiabPlot, aes(x=PrdProbValue, y=PrdOutLevels)) +
geom_point(shape=19, colour="blue", fill="blue") +
geom_smooth(method="gam", formula=y~s(log(x)), se=FALSE) +
labs(title="Binomial Regression Curve") +
labs(x="") +
labs(y="")
ROC Visulaization
#rocplot(logistic.model,diag=TRUE,pred.prob.labels=FALSE,prob.label.digits=3,AUC=TRUE)
rocplot(mLRDModel)
Test Data
dfrDiabTest <- read.csv("./data/diabetes-test.csv", header=T, stringsAsFactors=F)
head(dfrDiabTest)## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 4 118 70 0 0 44.5
## 2 2 122 76 27 200 35.9
## 3 6 125 78 31 0 27.6
## 4 1 168 88 29 0 35.0
## 5 2 129 0 0 0 38.5
## 6 4 110 76 20 100 28.4
## DiabetesPedigreeFunction Age
## 1 0.904 26
## 2 0.483 26
## 3 0.565 49
## 4 0.905 52
## 5 0.304 41
## 6 0.118 27Predict
resVal <- predict(mLRDModel, dfrDiabTest, type="response")
prdSur <- ifelse(resVal > 0.5, 1, 0)
prdSur <- as.factor(prdSur)
levels(prdSur) <- c("NoDiabetes", "Diabetes")
dfrDiabTest <- mutate(dfrDiabTest, Result=resVal, Outcome=prdSur)
dfrDiabTest## Pregnancies Glucose BloodPressure SkinThickness Insulin BMI
## 1 4 118 70 0 0 44.5
## 2 2 122 76 27 200 35.9
## 3 6 125 78 31 0 27.6
## 4 1 168 88 29 0 35.0
## 5 2 129 0 0 0 38.5
## 6 4 110 76 20 100 28.4
## 7 6 80 80 36 0 39.8
## 8 10 115 0 0 0 0.0
## 9 2 127 46 21 335 34.4
## 10 9 164 78 0 0 32.8
## 11 2 93 64 32 160 38.0
## 12 3 158 64 13 387 31.2
## 13 5 126 78 27 22 29.6
## 14 10 129 62 36 0 41.2
## 15 0 134 58 20 291 26.4
## 16 3 102 74 0 0 29.5
## 17 7 187 50 33 392 33.9
## 18 3 173 78 39 185 33.8
## 19 10 94 72 18 0 23.1
## 20 1 108 60 46 178 35.5
## 21 5 97 76 27 0 35.6
## 22 4 83 86 19 0 29.3
## 23 1 114 66 36 200 38.1
## 24 1 149 68 29 127 29.3
## 25 5 117 86 30 105 39.1
## 26 1 111 94 0 0 32.8
## 27 4 112 78 40 0 39.4
## 28 1 116 78 29 180 36.1
## 29 0 141 84 26 0 32.4
## 30 2 175 88 0 0 22.9
## 31 2 92 52 0 0 30.1
## 32 3 130 78 23 79 28.4
## 33 8 120 86 0 0 28.4
## 34 2 174 88 37 120 44.5
## 35 2 106 56 27 165 29.0
## 36 2 105 75 0 0 23.3
## 37 4 95 60 32 0 35.4
## 38 0 126 86 27 120 27.4
## 39 8 65 72 23 0 32.0
## 40 2 99 60 17 160 36.6
## 41 1 102 74 0 0 39.5
## 42 11 120 80 37 150 42.3
## 43 3 102 44 20 94 30.8
## 44 1 109 58 18 116 28.5
## 45 9 140 94 0 0 32.7
## 46 13 153 88 37 140 40.6
## 47 12 100 84 33 105 30.0
## 48 1 147 94 41 0 49.3
## 49 1 81 74 41 57 46.3
## 50 3 187 70 22 200 36.4
## 51 6 162 62 0 0 24.3
## 52 4 136 70 0 0 31.2
## 53 1 121 78 39 74 39.0
## 54 3 108 62 24 0 26.0
## 55 0 181 88 44 510 43.3
## 56 8 154 78 32 0 32.4
## 57 1 128 88 39 110 36.5
## 58 7 137 90 41 0 32.0
## 59 0 123 72 0 0 36.3
## 60 1 106 76 0 0 37.5
## 61 6 190 92 0 0 35.5
## 62 2 88 58 26 16 28.4
## 63 9 170 74 31 0 44.0
## 64 9 89 62 0 0 22.5
## 65 10 101 76 48 180 32.9
## 66 2 122 70 27 0 36.8
## 67 5 121 72 23 112 26.2
## 68 1 126 60 0 0 30.1
## 69 1 93 70 31 0 30.4
## DiabetesPedigreeFunction Age Result Outcome
## 1 0.904 26 0.69787807 Diabetes
## 2 0.483 26 0.34050577 NoDiabetes
## 3 0.565 49 0.33783836 NoDiabetes
## 4 0.905 52 0.76961152 Diabetes
## 5 0.304 41 0.39827848 NoDiabetes
## 6 0.118 27 0.12342951 NoDiabetes
## 7 0.177 28 0.17639188 NoDiabetes
## 8 0.261 30 0.03123857 NoDiabetes
## 9 0.176 22 0.26297406 NoDiabetes
## 10 0.148 45 0.73285378 Diabetes
## 11 0.674 23 0.23289782 NoDiabetes
## 12 0.295 24 0.50129268 Diabetes
## 13 0.439 40 0.31984267 NoDiabetes
## 14 0.441 38 0.75777856 Diabetes
## 15 0.352 21 0.16796184 NoDiabetes
## 16 0.121 32 0.09428466 NoDiabetes
## 17 0.826 34 0.92282539 Diabetes
## 18 0.970 31 0.83563179 Diabetes
## 19 0.595 56 0.17679794 NoDiabetes
## 20 0.415 24 0.19820536 NoDiabetes
## 21 0.378 52 0.22363330 NoDiabetes
## 22 0.317 34 0.07402098 NoDiabetes
## 23 0.289 21 0.24707179 NoDiabetes
## 24 0.349 42 0.33473911 NoDiabetes
## 25 0.251 42 0.39955233 NoDiabetes
## 26 0.265 45 0.14877788 NoDiabetes
## 27 0.236 38 0.33027291 NoDiabetes
## 28 0.496 25 0.27537859 NoDiabetes
## 29 0.433 22 0.33263712 NoDiabetes
## 30 0.326 22 0.42472516 NoDiabetes
## 31 0.141 22 0.06557517 NoDiabetes
## 32 0.323 34 0.23807540 NoDiabetes
## 33 0.259 22 0.29303589 NoDiabetes
## 34 0.646 24 0.89150589 Diabetes
## 35 0.426 22 0.12774058 NoDiabetes
## 36 0.560 53 0.08970935 NoDiabetes
## 37 0.284 28 0.16904239 NoDiabetes
## 38 0.515 21 0.17299519 NoDiabetes
## 39 0.600 42 0.12443027 NoDiabetes
## 40 0.453 21 0.19664471 NoDiabetes
## 41 0.293 42 0.20083845 NoDiabetes
## 42 0.785 48 0.82143392 Diabetes
## 43 0.400 26 0.14433641 NoDiabetes
## 44 0.219 22 0.09321648 NoDiabetes
## 45 0.734 45 0.72110866 Diabetes
## 46 1.174 39 0.96300322 Diabetes
## 47 0.488 46 0.36686147 NoDiabetes
## 48 0.358 27 0.75711972 Diabetes
## 49 1.096 32 0.39877391 NoDiabetes
## 50 0.408 36 0.83409753 Diabetes
## 51 0.178 50 0.44053443 NoDiabetes
## 52 1.182 22 0.63457214 Diabetes
## 53 0.261 28 0.30353413 NoDiabetes
## 54 0.223 25 0.09468293 NoDiabetes
## 55 0.222 26 0.80223746 Diabetes
## 56 0.443 45 0.70603391 Diabetes
## 57 1.057 37 0.54946023 Diabetes
## 58 0.391 39 0.51486844 Diabetes
## 59 0.258 52 0.23845382 NoDiabetes
## 60 0.197 26 0.17385465 NoDiabetes
## 61 0.278 66 0.86809089 Diabetes
## 62 0.766 22 0.10495172 NoDiabetes
## 63 0.403 43 0.93040106 Diabetes
## 64 0.142 33 0.07667843 NoDiabetes
## 65 0.171 63 0.28296664 NoDiabetes
## 66 0.340 27 0.31837533 NoDiabetes
## 67 0.245 30 0.18350215 NoDiabetes
## 68 0.349 47 0.20020349 NoDiabetes
## 69 0.315 23 0.07513332 NoDiabetesObservation:
1. When the predicted output was matched with the given output we found that 81 % of data was predicted correctly.