WDQ7004 Group Assignment
Group Members of Group 7
- Muhd Izani Zulkifli (22102674)
- Nur Fatihah Atikah Mohd Rashdan (17100287/2)
- Pavitra a/p Palanisamy (S2181697)
- Shameen Izwan Anthonysamy (s2180659)
Title
Obesity Level Predictive Modeling
Details Dataset
Dataset: Estimitaion of obesity Level
This dataset include data for the estimation of obesity levels in individuals from the countries of Mexico, Peru and Colombia, based on their eating habits and physical condition. It consists of 17 attributes and 2111 records, the records are labeled with the class variable NObesity (Obesity Level), that allows classification of the data using the values of Insufficient Weight, Normal Weight, Overweight Level I, Overweight Level II, Obesity Type I, Obesity Type II and Obesity Type III. 77% of the data was generated synthetically using the Weka tool and the SMOTE filter, 23% of the data was collected directly from users through a web platform.
Introduction
Obesity is a global health challenge with significant implications for individuals and society. As the prevalence of obesity continues to rise, understanding the factors contributing to obesity and developing effective predictive models are crucial for preventive healthcare interventions. Predictive modeling in the context of obesity aims to anticipate and identify individuals at risk, enabling timely interventions and personalized healthcare strategies.
Questions
Will various external factors frequency contribute to the prediction of body mass index (BMI)?
How can we predict the obesity level based on the various external factors contributing to body mass index (BMI)?
Objectives:
Predicting the body mass index (BMI) based on various features? (Linear Regression).
Classifying individuals into groups of Obesity (Classification).
Initialization
Import necessary libraries
library(dplyr)
library(plotly)
library(ggplot2)
library(tidyr)
library(ggcorrplot)
library(e1071)
library(caTools)
library(tidyverse)
library(caret)Data Ingestion
Load dataset of Obesity Level as dataframe
obesityDF <- read.csv("ObesityDataSet_raw_and_data_sinthetic.csv")Data Understanding
Check variables that attribute to the dataset
head(obesityDF)## Gender Age Height Weight family_history_with_overweight FAVC FCVC NCP
## 1 Female 21 1.62 64.0 yes no 2 3
## 2 Female 21 1.52 56.0 yes no 3 3
## 3 Male 23 1.80 77.0 yes no 2 3
## 4 Male 27 1.80 87.0 no no 3 3
## 5 Male 22 1.78 89.8 no no 2 1
## 6 Male 29 1.62 53.0 no yes 2 3
## CAEC SMOKE CH2O SCC FAF TUE CALC MTRANS
## 1 Sometimes no 2 no 0 1 no Public_Transportation
## 2 Sometimes yes 3 yes 3 0 Sometimes Public_Transportation
## 3 Sometimes no 2 no 2 1 Frequently Public_Transportation
## 4 Sometimes no 2 no 2 0 Frequently Walking
## 5 Sometimes no 2 no 0 0 Sometimes Public_Transportation
## 6 Sometimes no 2 no 0 0 Sometimes Automobile
## NObeyesdad
## 1 Normal_Weight
## 2 Normal_Weight
## 3 Normal_Weight
## 4 Overweight_Level_I
## 5 Overweight_Level_II
## 6 Normal_WeightData Preprocessing
1. Check for any NA values
colSums(is.na(obesityDF)) ## Gender Age
## 0 0
## Height Weight
## 0 0
## family_history_with_overweight FAVC
## 0 0
## FCVC NCP
## 0 0
## CAEC SMOKE
## 0 0
## CH2O SCC
## 0 0
## FAF TUE
## 0 0
## CALC MTRANS
## 0 0
## NObeyesdad
## 02. Do data cleaning process
- Round off the values of age variable from numeric to integer
- Load into new dataframe with age integer
obesityDF$Age <- round(obesityDF$Age)
age_obesity <- obesityDF- Check what the dataframe is about
#not necessary but if want to see what the changes of column name, can use this
str(age_obesity)## 'data.frame': 2111 obs. of 17 variables:
## $ Gender : chr "Female" "Female" "Male" "Male" ...
## $ Age : num 21 21 23 27 22 29 23 22 24 22 ...
## $ Height : num 1.62 1.52 1.8 1.8 1.78 1.62 1.5 1.64 1.78 1.72 ...
## $ Weight : num 64 56 77 87 89.8 53 55 53 64 68 ...
## $ family_history_with_overweight: chr "yes" "yes" "yes" "no" ...
## $ FAVC : chr "no" "no" "no" "no" ...
## $ FCVC : num 2 3 2 3 2 2 3 2 3 2 ...
## $ NCP : num 3 3 3 3 1 3 3 3 3 3 ...
## $ CAEC : chr "Sometimes" "Sometimes" "Sometimes" "Sometimes" ...
## $ SMOKE : chr "no" "yes" "no" "no" ...
## $ CH2O : num 2 3 2 2 2 2 2 2 2 2 ...
## $ SCC : chr "no" "yes" "no" "no" ...
## $ FAF : num 0 3 2 2 0 0 1 3 1 1 ...
## $ TUE : num 1 0 1 0 0 0 0 0 1 1 ...
## $ CALC : chr "no" "Sometimes" "Frequently" "Frequently" ...
## $ MTRANS : chr "Public_Transportation" "Public_Transportation" "Public_Transportation" "Walking" ...
## $ NObeyesdad : chr "Normal_Weight" "Normal_Weight" "Normal_Weight" "Overweight_Level_I" ...- Add BMI column
- By calculate using “BMI = weight (kg) ÷ height2 (meters)”
age_obesity$BMI = age_obesity$Weight/(age_obesity$Height^2)
str(age_obesity)## 'data.frame': 2111 obs. of 18 variables:
## $ Gender : chr "Female" "Female" "Male" "Male" ...
## $ Age : num 21 21 23 27 22 29 23 22 24 22 ...
## $ Height : num 1.62 1.52 1.8 1.8 1.78 1.62 1.5 1.64 1.78 1.72 ...
## $ Weight : num 64 56 77 87 89.8 53 55 53 64 68 ...
## $ family_history_with_overweight: chr "yes" "yes" "yes" "no" ...
## $ FAVC : chr "no" "no" "no" "no" ...
## $ FCVC : num 2 3 2 3 2 2 3 2 3 2 ...
## $ NCP : num 3 3 3 3 1 3 3 3 3 3 ...
## $ CAEC : chr "Sometimes" "Sometimes" "Sometimes" "Sometimes" ...
## $ SMOKE : chr "no" "yes" "no" "no" ...
## $ CH2O : num 2 3 2 2 2 2 2 2 2 2 ...
## $ SCC : chr "no" "yes" "no" "no" ...
## $ FAF : num 0 3 2 2 0 0 1 3 1 1 ...
## $ TUE : num 1 0 1 0 0 0 0 0 1 1 ...
## $ CALC : chr "no" "Sometimes" "Frequently" "Frequently" ...
## $ MTRANS : chr "Public_Transportation" "Public_Transportation" "Public_Transportation" "Walking" ...
## $ NObeyesdad : chr "Normal_Weight" "Normal_Weight" "Normal_Weight" "Overweight_Level_I" ...
## $ BMI : num 24.4 24.2 23.8 26.9 28.3 ...- Reorder columns to put BMI immediately after Height and Weight
obesity_new <- age_obesity[, c("Gender", "Age", "Height", "Weight", "BMI", "family_history_with_overweight", "FAVC", "FCVC", "NCP", "CAEC", "SMOKE", "CH2O", "SCC", "FAF", "TUE", "CALC", "MTRANS", "NObeyesdad")]
str(obesity_new)## 'data.frame': 2111 obs. of 18 variables:
## $ Gender : chr "Female" "Female" "Male" "Male" ...
## $ Age : num 21 21 23 27 22 29 23 22 24 22 ...
## $ Height : num 1.62 1.52 1.8 1.8 1.78 1.62 1.5 1.64 1.78 1.72 ...
## $ Weight : num 64 56 77 87 89.8 53 55 53 64 68 ...
## $ BMI : num 24.4 24.2 23.8 26.9 28.3 ...
## $ family_history_with_overweight: chr "yes" "yes" "yes" "no" ...
## $ FAVC : chr "no" "no" "no" "no" ...
## $ FCVC : num 2 3 2 3 2 2 3 2 3 2 ...
## $ NCP : num 3 3 3 3 1 3 3 3 3 3 ...
## $ CAEC : chr "Sometimes" "Sometimes" "Sometimes" "Sometimes" ...
## $ SMOKE : chr "no" "yes" "no" "no" ...
## $ CH2O : num 2 3 2 2 2 2 2 2 2 2 ...
## $ SCC : chr "no" "yes" "no" "no" ...
## $ FAF : num 0 3 2 2 0 0 1 3 1 1 ...
## $ TUE : num 1 0 1 0 0 0 0 0 1 1 ...
## $ CALC : chr "no" "Sometimes" "Frequently" "Frequently" ...
## $ MTRANS : chr "Public_Transportation" "Public_Transportation" "Public_Transportation" "Walking" ...
## $ NObeyesdad : chr "Normal_Weight" "Normal_Weight" "Normal_Weight" "Overweight_Level_I" ...- Rename the column name for better reading
names(obesity_new) <- c("Gender",
"Age",
"Height",
"Weight",
"BMI",
"Family_History_with_Overweight",
"High_Caloric_Food_Consumption",
"Frequency_Consumption_of_Vegetables",
"Number_of_Main_Meals",
"Consumption_of_Food_Between_Meals",
"Smoke",
"Consumption_of_Water_Daily",
"Calories_Consumption_Monitoring",
"Physical_Activity_Frequency",
"Time_Using_Technology",
"Consumption_of_Alcohol",
"Transportation_Used",
"Obesity")- Remove ’_’ underscore in values of Obesity column
obesity_new$Obesity <- gsub("_", " ", obesity_new$Obesity)head(obesity_new)## Gender Age Height Weight BMI Family_History_with_Overweight
## 1 Female 21 1.62 64.0 24.38653 yes
## 2 Female 21 1.52 56.0 24.23823 yes
## 3 Male 23 1.80 77.0 23.76543 yes
## 4 Male 27 1.80 87.0 26.85185 no
## 5 Male 22 1.78 89.8 28.34238 no
## 6 Male 29 1.62 53.0 20.19509 no
## High_Caloric_Food_Consumption Frequency_Consumption_of_Vegetables
## 1 no 2
## 2 no 3
## 3 no 2
## 4 no 3
## 5 no 2
## 6 yes 2
## Number_of_Main_Meals Consumption_of_Food_Between_Meals Smoke
## 1 3 Sometimes no
## 2 3 Sometimes yes
## 3 3 Sometimes no
## 4 3 Sometimes no
## 5 1 Sometimes no
## 6 3 Sometimes no
## Consumption_of_Water_Daily Calories_Consumption_Monitoring
## 1 2 no
## 2 3 yes
## 3 2 no
## 4 2 no
## 5 2 no
## 6 2 no
## Physical_Activity_Frequency Time_Using_Technology Consumption_of_Alcohol
## 1 0 1 no
## 2 3 0 Sometimes
## 3 2 1 Frequently
## 4 2 0 Frequently
## 5 0 0 Sometimes
## 6 0 0 Sometimes
## Transportation_Used Obesity
## 1 Public_Transportation Normal Weight
## 2 Public_Transportation Normal Weight
## 3 Public_Transportation Normal Weight
## 4 Walking Overweight Level I
## 5 Public_Transportation Overweight Level II
## 6 Automobile Normal Weight- Save as new csv file
write.csv(obesity_new, "obesity_new1.csv", row.names = FALSE)Exploratory Data Analysis
1. Plot histogram BMI with gender differentiation
ggplot(obesity_new, aes(x = BMI, fill = Gender)) +
geom_histogram(position = "identity", alpha = 0.7, bins = 20) +
labs(title = "Histogram of BMI by Gender",
x = "BMI",
y = "Frequency") +
scale_fill_manual(values = c("Male" = "blue", "Female" = "pink")) +
theme_minimal()
2. Plot correlation between age and BMI
ggplot(obesity_new, aes(x = Age, y = BMI)) +
geom_bar(stat = "identity", fill = "lightcoral", width = 0.7) +
labs(title = "Relationship between Age and BMI",
x = "Age",
y = "BMI")
3. Alcohol Consumption, Family History with Overweight vs BMI
ggplot(obesity_new, aes(x = as.factor(obesity_new$Consumption_of_Alcohol), y = obesity_new$BMI, fill = obesity_new$Family_History_with_Overweight)) +
geom_bar(stat = "summary", fun = "mean", position = "dodge", width = 0.5) +
labs(title = "BMI by Alcohol Consumption and Family History", x = "Alcohol Consumption", y = "BMI") +
scale_fill_manual(values = c("#eae2b7", "#fcbf49") ) +
theme_minimal() +
labs(title = "Relationship between BMI, Alcohol Consumption and Family History with Overweight",
x = "Alcohol Consumption",
y = "BMI",
fill = 'Family History with Overweight')
4. Obesity vs Age
ggplot(obesity_new, aes(x=as.factor(obesity_new$Obesity), y=obesity_new$Age)) +
geom_boxplot(
color="#003049",
fill="#eae2b7",
alpha=0.2,
notch=TRUE,
notchwidth = 0.8,
outlier.colour="#d62828",
outlier.fill="#d62828",
outlier.size=3
) +
theme_minimal() +
labs(title = "Relationship between Obesity Level and Age", x = "Obesity Level", y = "Age")
5. Find the correlation between the numerical fields
numericFields <- dplyr::select_if(obesity_new, is.numeric)
r <- cor(numericFields, use="complete.obs")
ggcorrplot(r)
Modeling
1. BMI Prediction (Regression Model)
- Define a Rsquared function and remove rows with Null values
df_reg <- obesity_new
df_reg <- df_reg %>%
mutate_if(is.character, as.factor) %>%
na.omit()
set.seed(123)- Train, Test, Split
splitIndex <- createDataPartition(df_reg$BMI, p = 0.8, list = FALSE)
training_data <- df_reg[splitIndex, ]
testing_data <- df_reg[-splitIndex, ]- Create Linear Regression model and train based on Training Data
model <- lm(BMI ~ ., data = training_data)- Make predictions with created model using Testing Data
pred_reg <- round(as.numeric(predict(model, newdata = testing_data)),digits = 2)
pred_reg## [1] 23.93 23.80 22.96 27.60 22.13 27.84 27.13 22.62 21.45 19.78 27.90 21.55
## [13] 22.13 20.70 36.08 28.95 17.69 22.39 21.66 28.41 22.24 26.42 22.80 22.78
## [25] 21.40 25.77 23.91 23.48 26.48 16.80 26.36 26.41 21.93 26.08 21.16 21.66
## [37] 26.50 22.87 23.26 25.49 38.13 30.44 31.75 35.53 28.12 21.92 38.84 21.29
## [49] 20.82 23.40 25.56 20.06 21.43 22.30 21.85 23.52 23.24 19.68 19.68 20.98
## [61] 38.06 17.21 22.44 24.15 14.10 17.04 29.79 22.65 20.05 21.08 25.27 22.90
## [73] 28.99 28.73 18.43 22.53 28.19 22.01 32.06 17.71 16.89 24.80 37.19 25.09
## [85] 25.98 21.24 20.92 22.99 25.32 21.01 33.90 20.55 21.42 31.77 26.35 19.18
## [97] 17.96 41.33 49.67 39.60 17.24 17.98 15.85 19.20 19.02 15.96 17.53 17.99
## [109] 18.08 16.98 16.70 19.47 17.94 16.48 17.94 15.00 15.39 18.90 16.52 18.16
## [121] 17.77 17.97 17.60 16.91 17.54 17.84 19.25 15.83 16.56 18.26 18.97 18.64
## [133] 16.89 16.33 16.68 16.29 17.30 15.26 17.70 17.45 15.53 15.16 19.12 16.65
## [145] 17.67 18.01 17.86 17.77 19.05 26.44 26.81 26.29 26.04 25.80 26.48 24.83
## [157] 25.71 25.36 26.52 25.43 25.13 26.22 25.72 26.55 26.09 27.75 25.36 26.50
## [169] 26.50 25.95 26.49 24.88 24.75 26.86 26.19 27.00 26.17 25.31 24.75 26.07
## [181] 25.82 26.18 25.58 25.88 26.12 26.30 25.91 24.99 25.62 26.23 25.49 25.26
## [193] 26.09 25.41 28.18 28.71 27.99 28.19 27.86 29.05 26.77 28.28 28.79 29.63
## [205] 29.28 28.16 27.98 28.16 28.25 27.55 27.80 28.47 28.56 27.25 28.73 27.73
## [217] 27.71 27.98 28.73 27.10 28.84 28.61 27.88 28.24 28.78 26.76 27.50 28.61
## [229] 28.03 28.07 27.66 27.35 28.30 27.85 27.94 29.39 28.22 28.18 28.74 27.65
## [241] 29.05 28.88 29.11 28.31 27.81 33.52 31.26 32.69 31.58 31.56 31.72 31.32
## [253] 30.80 31.03 32.96 32.11 31.57 30.88 32.55 31.66 32.58 31.21 32.37 32.08
## [265] 31.92 31.53 33.06 33.09 33.93 31.94 34.13 34.25 32.18 31.23 31.74 31.18
## [277] 31.88 33.77 31.69 31.79 32.30 32.39 31.82 32.61 30.03 32.51 32.91 31.50
## [289] 33.48 32.67 30.94 33.64 32.28 33.61 33.49 30.92 33.52 32.58 33.71 31.48
## [301] 33.00 32.55 33.32 36.52 38.00 37.58 37.83 36.62 36.02 36.29 36.39 36.61
## [313] 37.89 36.04 37.80 35.39 36.26 36.97 36.21 35.62 36.66 35.85 37.33 38.17
## [325] 36.02 36.12 37.04 36.27 38.04 37.07 36.12 38.16 37.19 36.86 36.28 38.30
## [337] 37.42 36.19 35.65 37.29 36.56 35.96 36.15 35.94 35.11 37.13 36.29 36.74
## [349] 36.47 35.79 36.08 36.15 38.42 37.61 35.28 36.05 37.13 36.43 36.50 35.63
## [361] 42.29 44.00 41.36 38.63 44.55 41.36 43.61 43.60 39.58 41.05 38.22 44.44
## [373] 40.94 46.79 41.10 40.60 44.10 38.19 38.26 44.11 47.25 43.96 39.54 40.81
## [385] 39.88 40.26 42.95 41.11 41.42 44.05 47.68 40.33 41.06 40.58 40.13 39.17
## [397] 44.27 47.01 40.72 40.91 44.40 44.14 44.84 44.71 44.77 43.96 43.63 41.10
## [409] 39.36 40.06 43.97 41.17 39.61 38.33 44.13 44.21 42.57 41.04 44.00 44.14- Check for prediction accuracy using Mean Square Error
mse_reg <- mean((testing_data$BMI - pred_reg)^2)
summary(model)##
## Call:
## lm(formula = BMI ~ ., data = training_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.40885 -0.31094 0.01986 0.33013 2.47279
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 4.850e+01 7.715e-01 62.864
## GenderMale 5.410e-01 4.471e-02 12.099
## Age 1.156e-03 3.271e-03 0.354
## Height -2.636e+01 3.566e-01 -73.931
## Weight 2.608e-01 2.756e-03 94.599
## Family_History_with_Overweightyes 4.228e-01 4.560e-02 9.272
## High_Caloric_Food_Consumptionyes 1.162e-01 5.012e-02 2.319
## Frequency_Consumption_of_Vegetables 1.212e-01 3.178e-02 3.815
## Number_of_Main_Meals 6.277e-02 2.006e-02 3.129
## Consumption_of_Food_Between_MealsFrequently -2.891e-01 1.002e-01 -2.885
## Consumption_of_Food_Between_Mealsno 9.128e-02 1.334e-01 0.684
## Consumption_of_Food_Between_MealsSometimes -8.814e-02 9.337e-02 -0.944
## Smokeyes -2.703e-01 1.038e-01 -2.604
## Consumption_of_Water_Daily -7.265e-02 2.508e-02 -2.896
## Calories_Consumption_Monitoringyes -1.343e-01 7.221e-02 -1.860
## Physical_Activity_Frequency -2.396e-02 1.893e-02 -1.266
## Time_Using_Technology 1.081e-02 2.524e-02 0.428
## Consumption_of_AlcoholFrequently 6.844e-02 5.895e-01 0.116
## Consumption_of_Alcoholno 1.846e-01 5.853e-01 0.315
## Consumption_of_AlcoholSometimes -5.956e-02 5.855e-01 -0.102
## Transportation_UsedBike 4.064e-04 2.420e-01 0.002
## Transportation_UsedMotorbike -3.408e-01 2.405e-01 -1.417
## Transportation_UsedPublic_Transportation -2.703e-01 4.619e-02 -5.851
## Transportation_UsedWalking 2.001e-03 9.998e-02 0.020
## ObesityNormal Weight 8.860e-01 7.096e-02 12.485
## ObesityObesity Type I 3.298e+00 1.403e-01 23.516
## ObesityObesity Type II 3.737e+00 1.822e-01 20.516
## ObesityObesity Type III 6.200e+00 2.157e-01 28.749
## ObesityOverweight Level I 1.949e+00 9.612e-02 20.278
## ObesityOverweight Level II 2.301e+00 1.147e-01 20.055
## Pr(>|t|)
## (Intercept) < 2e-16 ***
## GenderMale < 2e-16 ***
## Age 0.723719
## Height < 2e-16 ***
## Weight < 2e-16 ***
## Family_History_with_Overweightyes < 2e-16 ***
## High_Caloric_Food_Consumptionyes 0.020515 *
## Frequency_Consumption_of_Vegetables 0.000141 ***
## Number_of_Main_Meals 0.001782 **
## Consumption_of_Food_Between_MealsFrequently 0.003962 **
## Consumption_of_Food_Between_Mealsno 0.493915
## Consumption_of_Food_Between_MealsSometimes 0.345283
## Smokeyes 0.009294 **
## Consumption_of_Water_Daily 0.003827 **
## Calories_Consumption_Monitoringyes 0.063075 .
## Physical_Activity_Frequency 0.205859
## Time_Using_Technology 0.668479
## Consumption_of_AlcoholFrequently 0.907599
## Consumption_of_Alcoholno 0.752558
## Consumption_of_AlcoholSometimes 0.918990
## Transportation_UsedBike 0.998660
## Transportation_UsedMotorbike 0.156594
## Transportation_UsedPublic_Transportation 5.89e-09 ***
## Transportation_UsedWalking 0.984030
## ObesityNormal Weight < 2e-16 ***
## ObesityObesity Type I < 2e-16 ***
## ObesityObesity Type II < 2e-16 ***
## ObesityObesity Type III < 2e-16 ***
## ObesityOverweight Level I < 2e-16 ***
## ObesityOverweight Level II < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5736 on 1661 degrees of freedom
## Multiple R-squared: 0.995, Adjusted R-squared: 0.9949
## F-statistic: 1.129e+04 on 29 and 1661 DF, p-value: < 2.2e-162. Obesity Level Prediction (Classification Model)
- Generate random elements without replacement
df_svm <- obesity_new
df_svm$Gender = sample(c("Male", "Female"), 2111, replace = TRUE)
df_svm$Obesity <- sample(c('Normal Weight','Insufficient Weight', 'Overweight Level I','Overweight Level II',
'Obesity Type I','Obesity Type II','Obesity Type III'), 2111, replace = TRUE)- Convert the variables into factors
df_svm$Gender <- as.factor(df_svm$Gender)
df_svm$Obesity <- as.factor(df_svm$Obesity)
set.seed(456)- Train, Test, Split
train_index <- sample(seq_len(nrow(df_svm)), 0.7 * nrow(df_svm))
train_data <- df_svm[train_index, ]
test_data <- df_svm[-train_index, ]- Create Support Vector Machine model and train based on Training Data
model <- svm(formula = Obesity ~ .,
data = train_data,
kernel = 'linear')- Make predictions with created model using Testing Data
pred_svm <- predict(model, newdata = test_data)
pred_svm## 2 6 7 14
## Overweight Level II Obesity Type II Normal Weight Obesity Type II
## 16 19 21 25
## Overweight Level II Obesity Type III Insufficient Weight Normal Weight
## 28 30 33 38
## Obesity Type II Insufficient Weight Insufficient Weight Obesity Type II
## 39 45 49 50
## Normal Weight Overweight Level II Insufficient Weight Obesity Type II
## 51 52 53 54
## Insufficient Weight Obesity Type II Insufficient Weight Overweight Level II
## 55 58 59 63
## Overweight Level II Obesity Type III Insufficient Weight Obesity Type I
## 72 75 79 81
## Overweight Level II Obesity Type II Normal Weight Overweight Level II
## 83 84 87 89
## Normal Weight Overweight Level II Normal Weight Obesity Type III
## 93 94 96 98
## Overweight Level II Insufficient Weight Insufficient Weight Obesity Type III
## 102 103 104 105
## Overweight Level II Obesity Type III Normal Weight Obesity Type III
## 109 114 120 124
## Overweight Level II Obesity Type I Overweight Level II Normal Weight
## 129 134 139 140
## Obesity Type III Normal Weight Overweight Level II Obesity Type II
## 145 146 157 158
## Overweight Level II Obesity Type I Obesity Type III Obesity Type II
## 163 166 167 174
## Normal Weight Overweight Level II Overweight Level II Overweight Level II
## 175 178 183 190
## Obesity Type I Insufficient Weight Obesity Type II Normal Weight
## 192 193 198 199
## Overweight Level II Overweight Level II Obesity Type II Obesity Type III
## 207 216 224 226
## Insufficient Weight Normal Weight Normal Weight Overweight Level II
## 227 232 233 238
## Obesity Type II Normal Weight Overweight Level II Normal Weight
## 241 244 246 248
## Insufficient Weight Normal Weight Overweight Level I Obesity Type II
## 251 260 264 270
## Normal Weight Obesity Type III Insufficient Weight Insufficient Weight
## 271 273 275 278
## Insufficient Weight Obesity Type II Overweight Level II Obesity Type II
## 282 285 287 289
## Obesity Type II Obesity Type I Insufficient Weight Normal Weight
## 294 296 300 303
## Overweight Level II Insufficient Weight Insufficient Weight Insufficient Weight
## 311 313 315 318
## Insufficient Weight Insufficient Weight Normal Weight Normal Weight
## 321 326 332 336
## Normal Weight Obesity Type II Insufficient Weight Obesity Type II
## 342 347 348 354
## Overweight Level II Overweight Level II Insufficient Weight Obesity Type I
## 358 365 369 379
## Obesity Type I Normal Weight Obesity Type I Obesity Type II
## 382 383 391 393
## Obesity Type II Obesity Type III Normal Weight Overweight Level II
## 397 398 400 407
## Insufficient Weight Obesity Type II Obesity Type III Overweight Level II
## 408 409 412 417
## Obesity Type III Overweight Level II Overweight Level II Overweight Level II
## 418 420 421 422
## Obesity Type II Obesity Type I Insufficient Weight Insufficient Weight
## 423 426 427 430
## Obesity Type II Normal Weight Overweight Level II Overweight Level II
## 432 435 436 437
## Obesity Type II Insufficient Weight Insufficient Weight Obesity Type II
## 439 444 447 452
## Obesity Type II Normal Weight Obesity Type II Overweight Level II
## 467 468 469 471
## Obesity Type III Obesity Type III Obesity Type II Obesity Type II
## 473 479 480 481
## Obesity Type III Insufficient Weight Obesity Type II Insufficient Weight
## 483 487 488 490
## Obesity Type III Insufficient Weight Overweight Level I Obesity Type III
## 491 497 500 505
## Overweight Level II Insufficient Weight Obesity Type II Normal Weight
## 506 508 509 513
## Normal Weight Normal Weight Insufficient Weight Obesity Type III
## 518 520 523 525
## Obesity Type III Normal Weight Obesity Type III Obesity Type III
## 527 532 534 541
## Obesity Type III Normal Weight Normal Weight Overweight Level II
## 544 547 555 560
## Normal Weight Obesity Type II Obesity Type II Insufficient Weight
## 563 564 571 574
## Obesity Type II Insufficient Weight Obesity Type II Insufficient Weight
## 579 587 588 591
## Obesity Type III Obesity Type II Obesity Type II Obesity Type II
## 593 597 601 605
## Normal Weight Normal Weight Obesity Type III Obesity Type III
## 607 611 612 616
## Insufficient Weight Normal Weight Overweight Level II Obesity Type III
## 617 622 628 635
## Overweight Level II Normal Weight Insufficient Weight Normal Weight
## 637 639 643 646
## Obesity Type II Normal Weight Normal Weight Normal Weight
## 648 649 651 653
## Normal Weight Obesity Type III Overweight Level II Obesity Type III
## 654 656 658 662
## Obesity Type III Obesity Type II Insufficient Weight Obesity Type III
## 669 672 675 677
## Insufficient Weight Normal Weight Obesity Type II Obesity Type II
## 678 680 683 690
## Overweight Level II Normal Weight Insufficient Weight Obesity Type II
## 694 700 701 704
## Normal Weight Insufficient Weight Insufficient Weight Normal Weight
## 713 714 715 722
## Overweight Level II Obesity Type III Obesity Type III Insufficient Weight
## 729 731 732 733
## Normal Weight Insufficient Weight Normal Weight Obesity Type I
## 741 742 747 752
## Obesity Type III Obesity Type III Obesity Type I Obesity Type II
## 757 763 764 766
## Obesity Type III Obesity Type I Obesity Type I Obesity Type I
## 768 773 775 776
## Normal Weight Normal Weight Obesity Type I Obesity Type I
## 777 781 785 786
## Obesity Type I Normal Weight Obesity Type III Obesity Type III
## 788 790 803 805
## Obesity Type III Obesity Type I Obesity Type III Insufficient Weight
## 817 818 820 823
## Normal Weight Normal Weight Normal Weight Insufficient Weight
## 824 829 834 838
## Normal Weight Obesity Type I Obesity Type I Normal Weight
## 842 847 849 854
## Normal Weight Insufficient Weight Insufficient Weight Obesity Type I
## 856 860 863 865
## Obesity Type I Normal Weight Obesity Type III Normal Weight
## 866 867 868 871
## Obesity Type III Obesity Type III Obesity Type III Obesity Type III
## 884 885 889 893
## Obesity Type I Insufficient Weight Obesity Type II Obesity Type III
## 897 902 909 919
## Obesity Type I Normal Weight Normal Weight Obesity Type I
## 922 926 927 929
## Obesity Type I Obesity Type I Normal Weight Normal Weight
## 937 956 957 959
## Obesity Type I Obesity Type III Insufficient Weight Obesity Type I
## 963 965 968 971
## Obesity Type III Obesity Type III Obesity Type III Normal Weight
## 973 975 976 981
## Insufficient Weight Insufficient Weight Insufficient Weight Insufficient Weight
## 982 986 987 989
## Insufficient Weight Normal Weight Overweight Level II Insufficient Weight
## 994 997 1000 1001
## Obesity Type III Insufficient Weight Normal Weight Normal Weight
## 1005 1009 1011 1013
## Overweight Level II Normal Weight Insufficient Weight Insufficient Weight
## 1015 1017 1022 1027
## Normal Weight Normal Weight Obesity Type I Normal Weight
## 1029 1037 1039 1045
## Obesity Type III Normal Weight Normal Weight Overweight Level II
## 1047 1048 1056 1058
## Normal Weight Overweight Level II Normal Weight Normal Weight
## 1061 1064 1070 1073
## Normal Weight Normal Weight Obesity Type I Obesity Type II
## 1075 1077 1079 1082
## Normal Weight Insufficient Weight Normal Weight Overweight Level II
## 1083 1087 1089 1091
## Overweight Level II Normal Weight Normal Weight Insufficient Weight
## 1093 1095 1100 1102
## Insufficient Weight Obesity Type I Overweight Level II Normal Weight
## 1103 1104 1106 1110
## Overweight Level II Overweight Level II Insufficient Weight Obesity Type I
## 1112 1116 1119 1127
## Insufficient Weight Insufficient Weight Obesity Type III Obesity Type I
## 1128 1129 1136 1137
## Normal Weight Normal Weight Obesity Type III Insufficient Weight
## 1139 1140 1149 1152
## Overweight Level II Normal Weight Insufficient Weight Normal Weight
## 1153 1160 1163 1167
## Normal Weight Normal Weight Normal Weight Obesity Type III
## 1174 1177 1179 1182
## Normal Weight Overweight Level II Overweight Level II Normal Weight
## 1184 1185 1189 1198
## Normal Weight Normal Weight Overweight Level II Obesity Type I
## 1200 1205 1207 1211
## Normal Weight Normal Weight Normal Weight Normal Weight
## 1216 1217 1218 1222
## Obesity Type III Normal Weight Obesity Type I Normal Weight
## 1223 1234 1236 1238
## Normal Weight Normal Weight Normal Weight Obesity Type II
## 1240 1243 1244 1247
## Normal Weight Obesity Type III Obesity Type II Insufficient Weight
## 1248 1250 1253 1257
## Obesity Type I Insufficient Weight Insufficient Weight Insufficient Weight
## 1260 1263 1265 1266
## Normal Weight Insufficient Weight Obesity Type III Insufficient Weight
## 1274 1280 1281 1282
## Obesity Type I Normal Weight Insufficient Weight Insufficient Weight
## 1285 1287 1289 1290
## Obesity Type I Obesity Type I Obesity Type I Normal Weight
## 1294 1295 1296 1298
## Normal Weight Normal Weight Obesity Type III Normal Weight
## 1303 1312 1313 1316
## Insufficient Weight Obesity Type I Obesity Type III Obesity Type III
## 1321 1322 1327 1334
## Insufficient Weight Normal Weight Obesity Type III Obesity Type I
## 1341 1349 1355 1358
## Insufficient Weight Insufficient Weight Insufficient Weight Normal Weight
## 1359 1362 1365 1367
## Insufficient Weight Insufficient Weight Obesity Type I Obesity Type I
## 1369 1371 1372 1374
## Normal Weight Normal Weight Insufficient Weight Insufficient Weight
## 1377 1378 1382 1384
## Insufficient Weight Normal Weight Obesity Type III Obesity Type I
## 1387 1389 1393 1394
## Normal Weight Normal Weight Obesity Type III Insufficient Weight
## 1395 1399 1401 1404
## Insufficient Weight Normal Weight Normal Weight Insufficient Weight
## 1418 1423 1424 1427
## Obesity Type III Normal Weight Obesity Type II Normal Weight
## 1435 1437 1439 1445
## Obesity Type I Normal Weight Obesity Type III Obesity Type I
## 1446 1453 1472 1475
## Normal Weight Insufficient Weight Normal Weight Normal Weight
## 1479 1480 1482 1483
## Insufficient Weight Insufficient Weight Obesity Type I Obesity Type I
## 1484 1486 1492 1494
## Obesity Type III Insufficient Weight Insufficient Weight Insufficient Weight
## 1498 1499 1500 1502
## Insufficient Weight Obesity Type I Obesity Type I Obesity Type I
## 1512 1513 1516 1522
## Insufficient Weight Insufficient Weight Normal Weight Insufficient Weight
## 1524 1528 1530 1532
## Normal Weight Normal Weight Normal Weight Normal Weight
## 1533 1536 1544 1552
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 1555 1557 1559 1560
## Normal Weight Obesity Type III Obesity Type II Normal Weight
## 1564 1565 1566 1568
## Normal Weight Normal Weight Normal Weight Normal Weight
## 1571 1572 1573 1576
## Insufficient Weight Normal Weight Insufficient Weight Normal Weight
## 1577 1580 1583 1584
## Insufficient Weight Insufficient Weight Insufficient Weight Normal Weight
## 1585 1586 1588 1593
## Normal Weight Insufficient Weight Normal Weight Normal Weight
## 1597 1598 1599 1603
## Insufficient Weight Obesity Type III Normal Weight Obesity Type II
## 1607 1613 1614 1616
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 1617 1624 1625 1627
## Normal Weight Insufficient Weight Normal Weight Insufficient Weight
## 1629 1633 1635 1640
## Normal Weight Normal Weight Insufficient Weight Normal Weight
## 1641 1643 1646 1650
## Insufficient Weight Insufficient Weight Insufficient Weight Normal Weight
## 1651 1652 1658 1659
## Normal Weight Normal Weight Obesity Type III Normal Weight
## 1663 1666 1667 1668
## Normal Weight Obesity Type III Normal Weight Insufficient Weight
## 1673 1675 1676 1685
## Normal Weight Normal Weight Normal Weight Insufficient Weight
## 1686 1687 1690 1693
## Insufficient Weight Obesity Type II Normal Weight Insufficient Weight
## 1694 1695 1700 1701
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 1702 1705 1706 1709
## Normal Weight Insufficient Weight Normal Weight Obesity Type III
## 1713 1716 1719 1720
## Normal Weight Insufficient Weight Normal Weight Obesity Type I
## 1722 1727 1729 1730
## Normal Weight Normal Weight Normal Weight Insufficient Weight
## 1731 1734 1736 1737
## Normal Weight Insufficient Weight Insufficient Weight Insufficient Weight
## 1738 1740 1747 1752
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 1753 1756 1758 1761
## Insufficient Weight Normal Weight Insufficient Weight Insufficient Weight
## 1769 1771 1777 1779
## Normal Weight Normal Weight Normal Weight Normal Weight
## 1782 1790 1791 1799
## Normal Weight Insufficient Weight Normal Weight Normal Weight
## 1803 1805 1808 1811
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 1813 1818 1830 1834
## Normal Weight Normal Weight Normal Weight Normal Weight
## 1842 1850 1854 1858
## Normal Weight Normal Weight Insufficient Weight Normal Weight
## 1859 1865 1868 1869
## Normal Weight Insufficient Weight Normal Weight Normal Weight
## 1870 1872 1877 1883
## Normal Weight Obesity Type II Insufficient Weight Normal Weight
## 1886 1889 1894 1898
## Normal Weight Normal Weight Normal Weight Normal Weight
## 1906 1919 1922 1931
## Normal Weight Insufficient Weight Normal Weight Normal Weight
## 1936 1937 1941 1945
## Normal Weight Normal Weight Normal Weight Insufficient Weight
## 1947 1948 1949 1950
## Normal Weight Normal Weight Normal Weight Normal Weight
## 1952 1955 1958 1969
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 1973 1974 1975 1982
## Insufficient Weight Insufficient Weight Insufficient Weight Normal Weight
## 1985 1986 1997 1998
## Normal Weight Insufficient Weight Insufficient Weight Normal Weight
## 2000 2002 2008 2010
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 2011 2012 2019 2021
## Normal Weight Insufficient Weight Normal Weight Normal Weight
## 2022 2023 2024 2025
## Normal Weight Normal Weight Normal Weight Normal Weight
## 2028 2039 2043 2045
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 2055 2056 2058 2062
## Insufficient Weight Normal Weight Normal Weight Normal Weight
## 2063 2066 2072 2073
## Normal Weight Normal Weight Insufficient Weight Obesity Type II
## 2074 2076 2077 2081
## Normal Weight Normal Weight Normal Weight Normal Weight
## 2085 2086 2087 2090
## Normal Weight Normal Weight Normal Weight Normal Weight
## 2091 2094 2098 2106
## Normal Weight Normal Weight Normal Weight Insufficient Weight
## 2108 2109
## Insufficient Weight Insufficient Weight
## 7 Levels: Insufficient Weight Normal Weight Obesity Type I ... Overweight Level II- Check for prediction accuracy using Confusion Matrix
mse_svm <- confusionMatrix(pred_svm, test_data$Obesity)
mse_svm## Confusion Matrix and Statistics
##
## Reference
## Prediction Insufficient Weight Normal Weight Obesity Type I
## Insufficient Weight 23 11 15
## Normal Weight 36 43 38
## Obesity Type I 4 9 7
## Obesity Type II 7 4 7
## Obesity Type III 21 8 6
## Overweight Level I 1 0 0
## Overweight Level II 14 8 7
## Reference
## Prediction Obesity Type II Obesity Type III Overweight Level I
## Insufficient Weight 29 26 18
## Normal Weight 29 31 42
## Obesity Type I 9 12 6
## Obesity Type II 6 13 8
## Obesity Type III 12 7 9
## Overweight Level I 1 0 0
## Overweight Level II 4 5 6
## Reference
## Prediction Overweight Level II
## Insufficient Weight 28
## Normal Weight 30
## Obesity Type I 7
## Obesity Type II 9
## Obesity Type III 9
## Overweight Level I 0
## Overweight Level II 9
##
## Overall Statistics
##
## Accuracy : 0.1498
## 95% CI : (0.123, 0.18)
## No Information Rate : 0.1672
## P-Value [Acc > NIR] : 0.8907
##
## Kappa : 0.0077
##
## Mcnemar's Test P-Value : <2e-16
##
## Statistics by Class:
##
## Class: Insufficient Weight Class: Normal Weight
## Sensitivity 0.21698 0.51807
## Specificity 0.75947 0.62613
## Pos Pred Value 0.15333 0.17269
## Neg Pred Value 0.82851 0.89610
## Prevalence 0.16719 0.13091
## Detection Rate 0.03628 0.06782
## Detection Prevalence 0.23659 0.39274
## Balanced Accuracy 0.48823 0.57210
## Class: Obesity Type I Class: Obesity Type II
## Sensitivity 0.08750 0.066667
## Specificity 0.91516 0.911765
## Pos Pred Value 0.12963 0.111111
## Neg Pred Value 0.87414 0.855172
## Prevalence 0.12618 0.141956
## Detection Rate 0.01104 0.009464
## Detection Prevalence 0.08517 0.085174
## Balanced Accuracy 0.50133 0.489216
## Class: Obesity Type III Class: Overweight Level I
## Sensitivity 0.07447 0.000000
## Specificity 0.87963 0.996330
## Pos Pred Value 0.09722 0.000000
## Neg Pred Value 0.84520 0.859177
## Prevalence 0.14826 0.140379
## Detection Rate 0.01104 0.000000
## Detection Prevalence 0.11356 0.003155
## Balanced Accuracy 0.47705 0.498165
## Class: Overweight Level II
## Sensitivity 0.09783
## Specificity 0.91882
## Pos Pred Value 0.16981
## Neg Pred Value 0.85714
## Prevalence 0.14511
## Detection Rate 0.01420
## Detection Prevalence 0.08360
## Balanced Accuracy 0.50832summary(pred_svm)## Insufficient Weight Normal Weight Obesity Type I Obesity Type II
## 150 249 54 54
## Obesity Type III Overweight Level I Overweight Level II
## 72 2 53Conclusion
BMI Prediction (Regression Model)
Model able to predict BMI, holds RSE value of 0.995.
Obesity Level Prediction (Classification Model)
Model able to predict Obesity Level, though, requires more tuning as the accuracy is only 15%.