About the Dataset

The data contains 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.

The dataset can be found here : https://archive-beta.ics.uci.edu/ml/datasets/estimation+of+obesity+levels+based+on+eating+habits+and+physical+condition

• Columns Description :
  • Gender : Female/Male
  • Age : Numeric value
  • Height : Numeric value in meters
  • Weight : Numeric value in kilograms
  • Family History With Over Weight
  • FAVC : Frequent consumption of high caloric food,
  • FCVC : Frequency of consumption of vegetables
  • NCP : Number of main meals
  • CAEC : Consumption of food between meals
  • CH20 : Consumption of water daily
  • CALC : Consumption of alcohol
  • SCC : Calories consumption monitoring
  • FAF : Physical activity frequency
  • TUE : Time using technology devices
  • MTRANS : Transportation used

Data Input:

Read “obesity.csv” as obesity and rename the columns and round the number(value) into integer

obesity <- read.csv("obesity.csv", stringsAsFactors = T)
names(obesity) <- c("Gender", "Age", "Height", "Weight", "Family_History_with_Overweight",
       "Frequent_consumption_of_high_caloric_food", "Frequency_of_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_devices",
       "Consumption_of_alcohol", "Transportation_used", "Obesity")
obesity <- obesity %>% mutate_at(vars(Frequency_of_consumption_of_vegetables, Number_of_main_meals,Consumption_of_water_daily, Physical_activity_frequency, Time_using_technology_devices), funs(round(.,0)))

## Warning: `funs()` was deprecated in dplyr 0.8.0.
## Please use a list of either functions or lambdas: 
## 
##   # Simple named list: 
##   list(mean = mean, median = median)
## 
##   # Auto named with `tibble::lst()`: 
##   tibble::lst(mean, median)
## 
##   # Using lambdas
##   list(~ mean(., trim = .2), ~ median(., na.rm = TRUE))
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.

str(obesity)

## 'data.frame':    2111 obs. of  17 variables:
##  $ Gender                                   : Factor w/ 2 levels "Female","Male": 1 1 2 2 2 2 1 2 2 2 ...
##  $ 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           : Factor w/ 2 levels "no","yes": 2 2 2 1 1 1 2 1 2 2 ...
##  $ Frequent_consumption_of_high_caloric_food: Factor w/ 2 levels "no","yes": 1 1 1 1 1 2 2 1 2 2 ...
##  $ Frequency_of_consumption_of_vegetables   : num  2 3 2 3 2 2 3 2 3 2 ...
##  $ Number_of_main_meals                     : num  3 3 3 3 1 3 3 3 3 3 ...
##  $ Consumption_of_food_between_meals        : Factor w/ 4 levels "Always","Frequently",..: 4 4 4 4 4 4 4 4 4 4 ...
##  $ Smoke                                    : Factor w/ 2 levels "no","yes": 1 2 1 1 1 1 1 1 1 1 ...
##  $ Consumption_of_water_daily               : num  2 3 2 2 2 2 2 2 2 2 ...
##  $ Calories_consumption_monitoring          : Factor w/ 2 levels "no","yes": 1 2 1 1 1 1 1 1 1 1 ...
##  $ Physical_activity_frequency              : num  0 3 2 2 0 0 1 3 1 1 ...
##  $ Time_using_technology_devices            : num  1 0 1 0 0 0 0 0 1 1 ...
##  $ Consumption_of_alcohol                   : Factor w/ 4 levels "Always","Frequently",..: 3 4 2 2 4 4 4 4 2 3 ...
##  $ Transportation_used                      : Factor w/ 5 levels "Automobile","Bike",..: 4 4 4 5 4 1 3 4 4 4 ...
##  $ Obesity                                  : Factor w/ 7 levels "Insufficient_Weight",..: 2 2 2 6 7 2 2 2 2 2 ...

Save edited data to .csv format and load it as “obesity_new”

write_csv(obesity,"obesity_new.csv" )
obesity_new <- read.csv("obesity_new.csv", stringsAsFactors = T)
str(obesity_new)

## 'data.frame':    2111 obs. of  17 variables:
##  $ Gender                                   : Factor w/ 2 levels "Female","Male": 1 1 2 2 2 2 1 2 2 2 ...
##  $ 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           : Factor w/ 2 levels "no","yes": 2 2 2 1 1 1 2 1 2 2 ...
##  $ Frequent_consumption_of_high_caloric_food: Factor w/ 2 levels "no","yes": 1 1 1 1 1 2 2 1 2 2 ...
##  $ Frequency_of_consumption_of_vegetables   : int  2 3 2 3 2 2 3 2 3 2 ...
##  $ Number_of_main_meals                     : int  3 3 3 3 1 3 3 3 3 3 ...
##  $ Consumption_of_food_between_meals        : Factor w/ 4 levels "Always","Frequently",..: 4 4 4 4 4 4 4 4 4 4 ...
##  $ Smoke                                    : Factor w/ 2 levels "no","yes": 1 2 1 1 1 1 1 1 1 1 ...
##  $ Consumption_of_water_daily               : int  2 3 2 2 2 2 2 2 2 2 ...
##  $ Calories_consumption_monitoring          : Factor w/ 2 levels "no","yes": 1 2 1 1 1 1 1 1 1 1 ...
##  $ Physical_activity_frequency              : int  0 3 2 2 0 0 1 3 1 1 ...
##  $ Time_using_technology_devices            : int  1 0 1 0 0 0 0 0 1 1 ...
##  $ Consumption_of_alcohol                   : Factor w/ 4 levels "Always","Frequently",..: 3 4 2 2 4 4 4 4 2 3 ...
##  $ Transportation_used                      : Factor w/ 5 levels "Automobile","Bike",..: 4 4 4 5 4 1 3 4 4 4 ...
##  $ Obesity                                  : Factor w/ 7 levels "Insufficient_Weight",..: 2 2 2 6 7 2 2 2 2 2 ...

Data Inspection :

Change selected columuns to factor format

obesity_new <- obesity_new %>% mutate(Frequency_of_consumption_of_vegetables = as.factor(Frequency_of_consumption_of_vegetables),
                              Number_of_main_meals = as.factor(Number_of_main_meals),
                              Consumption_of_water_daily = as.factor(Consumption_of_water_daily),
                              Physical_activity_frequency = as.factor(Physical_activity_frequency),
                              Time_using_technology_devices = as.factor(Time_using_technology_devices))

List old Factor Name and Create the new Factor Name

old_1 <- c("1", "2", "3")
old_2 <- c("1", "2", "3", "4")
old_3 <- c("1", "2", "3")
old_4 <- c("0", "1", "2", "3")
old_5 <- c("0", "1", "2")

new_1 <- c("Never", "Sometimes", "Always")
new_2 <- c("1", "2", "3", "3+")
new_3 <- c("Less than a liter", "Between 1 and 2 L", "More than 2 L")
new_4 <- c("I do not have", "1 - 2 times", "3 - 4 times", "More than 4 times")
new_5 <- c("0–2 hours", "3–5 hours", "More than 5 hours")

Assign the New Factor Name into dataframe

obesity_new$Frequency_of_consumption_of_vegetables <- do.call(
  fct_recode,
  c(list(obesity_new$Frequency_of_consumption_of_vegetables), setNames(old_1, new_1)))

obesity_new$Number_of_main_meals <- do.call(
  fct_recode,
  c(list(obesity_new$Number_of_main_meals), setNames(old_2, new_2)))

obesity_new$Consumption_of_water_daily <- do.call(
  fct_recode,
  c(list(obesity_new$Consumption_of_water_daily), setNames(old_3, new_3)))

obesity_new$Physical_activity_frequency <- do.call(
  fct_recode,
  c(list(obesity_new$Physical_activity_frequency), setNames(old_4, new_4)))

obesity_new$Time_using_technology_devices <- do.call(
  fct_recode,
  c(list(obesity_new$Time_using_technology_devices), setNames(old_5, new_5)))

Check the Missing Value of the data

colSums(is.na(obesity_new))

##                                    Gender 
##                                         0 
##                                       Age 
##                                         0 
##                                    Height 
##                                         0 
##                                    Weight 
##                                         0 
##            Family_History_with_Overweight 
##                                         0 
## Frequent_consumption_of_high_caloric_food 
##                                         0 
##    Frequency_of_consumption_of_vegetables 
##                                         0 
##                      Number_of_main_meals 
##                                         0 
##         Consumption_of_food_between_meals 
##                                         0 
##                                     Smoke 
##                                         0 
##                Consumption_of_water_daily 
##                                         0 
##           Calories_consumption_monitoring 
##                                         0 
##               Physical_activity_frequency 
##                                         0 
##             Time_using_technology_devices 
##                                         0 
##                    Consumption_of_alcohol 
##                                         0 
##                       Transportation_used 
##                                         0 
##                                   Obesity 
##                                         0

Change Height matric from Meter to Centimeter by multiply the value with 100

obesity_new <- obesity_new %>% mutate_at(vars(Age, Weight), funs(round(.,0)))
obesity_new$Height <- obesity_new$Height *100

Correlation Between Height and Weight in Type Of Obesity

obesity_cor <- obesity_new %>%
  select(c(Obesity, Height, Weight))

plotob_cor <- ggplot(data = obesity_cor, mapping = aes(x = Height, y = Weight, col = Obesity))+
  geom_point(aes(col = Obesity))+
  geom_smooth(method=lm , color="black", se=FALSE, formula = y~x) +
  scale_fill_viridis(discrete = T, option = "C") +
  labs(title = list(text = paste0('Correlation of Height and Weight')),
    x = "Height (cm)",
    y = "Weight (Kg)"
  ) +
  theme(legend.title = element_blank(),
        plot.title = element_text(face = "bold"),
        panel.background = element_rect(fill = "#ffffff"),
        axis.line.y = element_line(colour = "grey"),
        axis.line.x = element_line())
ggplotly(plotob_cor, tooltip = "text")

ob_corr <- cor(obesity_cor$Height, obesity_cor$Weight)

According to the data, the correlation between height and weight is weakly positive (0.46)

Height and Weight Distribution based on Gender

height_weight <- obesity_new %>%
  select(c(Gender, Height, Weight))
height_weight <- pivot_longer(data = height_weight,
                           cols = c("Height","Weight"),
                           names_to = "variabel")

plothw <- ggplot(data = height_weight, mapping = aes(x = Gender, y = value))+
  geom_boxplot(aes(fill=Gender), position = "dodge")+
  facet_wrap(vars(variabel)) + #memisahkan plot berdasarkan variable parameter
  labs(title = list(text = paste0('Height and Weight Distribution Based on Gender')),
    x = "Gender",
    y = "Height (cm) / Weight (Kg)"
  ) +
  theme(legend.title = element_blank(),
        plot.title = element_text(face = "bold"),
        panel.background = element_rect(fill = "#ffffff"),
        axis.line.y = element_line(colour = "grey"),
        axis.line.x = element_line())
  
ggplotly(plothw, tooltip = "text")

The box plots show the distribution of Height and Weight based on Gender wise. The plots highlight that the median height of females in the sample is significantly lower than that of males, with a few of males surpassing 1.98 meters (outliers).   In terms of their weights, though, the difference is not as significant. While, one individual with a weight of more than 165 kg is considered an outlier.

Obesity Type Distribution on Gender

obs_gender <- obesity_new %>%
  select(c(Gender, Obesity)) %>%
  group_by(Gender, Obesity)  %>% 
  summarise(total = n()) %>%
  mutate(label = glue("Total : {total}"))

plotog <- ggplot(data = obs_gender, aes(x = Obesity, y = total, fill = Gender, text = label))+
  geom_col(position = "dodge")+
  facet_wrap(vars(Gender)) + #memisahkan plot berdasarkan variable parameter
  scale_fill_viridis(discrete = T, option = "C") +
  labs(title = list(text = paste0('Obesity Type based on Gender')),
    x = "Gender",
    y = "Total"
  ) +
  theme(legend.title = element_blank(),
        axis.text.x = element_text(hjust = 1, angle = 20),
        plot.title = element_text(face = "bold"),
        panel.background = element_rect(fill = "#ffffff"),
        axis.line.y = element_line(colour = "grey"),
        axis.line.x = element_line()) +
  coord_flip()
  
ggplotly(plotog, tooltip = "text")

Obesity distribution based on Two Related Variabel

#DATA COSTUM A
calories_activity <- obesity_new %>%
  select(c(Frequent_consumption_of_high_caloric_food, Physical_activity_frequency, Obesity)) %>%
  group_by(Obesity, Frequent_consumption_of_high_caloric_food, Physical_activity_frequency ) %>%
  summarise(count = n()) %>%
  ungroup()
calories_activity <- calories_activity %>% mutate(label = glue("Total : {count}
                                                                Obesity Type : {Obesity}
                                                                Frequent Consumpyion Of High Caloric Food : {Frequent_consumption_of_high_caloric_food}
                                                                Frequency of Physical Activity : {Physical_activity_frequency}"))

plota <- ggplot(calories_activity, aes(x = Physical_activity_frequency, y = count, fill = Obesity, text = label)) +
        geom_col(position = "dodge") +
        facet_wrap(vars(Frequent_consumption_of_high_caloric_food)) +
        scale_fill_viridis(discrete = T, option = "C") +
        scale_x_discrete(expand = expansion(mult = c(0.0001, 0.0001)),
                         labels = function(x) str_wrap(x, width = 10))+
        labs(title = "Frequently Consuming High Caloric Food :",
             x = 'Frequency of Physical Activity',
             y = NULL,
             fill = NULL) +
        theme(legend.title = element_text(),
              axis.text.x = element_text(hjust = 1, angle = 20),
              plot.title = element_text(face = "bold"),
              panel.background = element_rect(fill = "#ffffff"),
              axis.line.y = element_line(colour = "grey"),
              axis.line.x = element_blank(),
              panel.grid = element_blank())
ggplotly(plota, tooltip = "text")

## DATA CUSTOM B
calories_meals <- obesity_new %>%
  select(c(Calories_consumption_monitoring, Number_of_main_meals, Obesity)) %>%
  group_by(Calories_consumption_monitoring, Number_of_main_meals, Obesity) %>%
  summarise(count = n()) %>%
  ungroup()
calories_meals <- calories_meals %>% mutate(label = glue("Total : {count}
                                                          Obesity Type : {Obesity}
                                                          Calories Consumpyion Monitorting : {Calories_consumption_monitoring}
                                                          Number of Main Meals : {Number_of_main_meals}"))

plotb <- ggplot(calories_meals, aes(x = Number_of_main_meals, y = count, fill = Obesity, text = label)) +
        geom_col(position = "dodge") +
        facet_wrap(vars(Calories_consumption_monitoring)) +
        scale_fill_viridis(discrete = T, option = "C") +
        scale_x_discrete(expand = expansion(mult = c(0.0001, 0.0001)))+
        labs(title = "Calories Consumption Monitoring",
             x = "Number of Main Meals",
             y = NULL,
             fill = "Obesity Type") +
        theme(legend.title = element_blank(),
              axis.text.x = element_text(hjust = 1),
              plot.title = element_text(face = "bold"),
              panel.background = element_rect(fill = "#ffffff"),
              axis.line.y = element_line(colour = "grey"),
              axis.line.x = element_blank(),
              panel.grid = element_blank())
ggplotly(plotb, tooltip = "text")

## DATA CUSTOM 3
gene_snack <- obesity_new %>%
  select(c(Family_History_with_Overweight, Consumption_of_food_between_meals, Obesity)) %>%
  group_by(Family_History_with_Overweight, Consumption_of_food_between_meals, Obesity) %>%
  summarise(count = n()) %>%
  ungroup()
gene_snack <- gene_snack %>% mutate(label = glue("Total : {count}
                                                          Obesity Type : {Obesity}
                                                          Family History with Overweight : {Family_History_with_Overweight}
                                                          Consumption of food between of Meals : {Consumption_of_food_between_meals}"))

#"Distribution of Obesity Type on Food Consumption Between Meals and Family History of Overweight"
plotc <- ggplot(gene_snack, mapping = aes(x = Consumption_of_food_between_meals, y = count, fill = Obesity, text = label)) +
        geom_col(position = "dodge") +
        facet_wrap(vars(Family_History_with_Overweight)) +
        scale_fill_viridis(discrete = T, option = "C") +
        scale_x_discrete(expand = expansion(mult = c(0.0001, 0.0001)))+
        labs(title = "Family History with Overweight",
             x = "Consumption Of Food Between Meals",
             y = NULL,
             fill = "Obesity Type") +
        theme(legend.title = element_blank(),
              axis.text.x = element_text(),
              plot.title = element_text(face = "bold"),
              panel.background = element_rect(fill = "#ffffff"),
              axis.line.y = element_line(colour = "grey"),
              axis.line.x = element_blank(),
              panel.grid = element_blank())
ggplotly(plotc, tooltip = "text")

## DATA COSTUM 4
alcohol_water <- obesity_new %>%
  select(c(Consumption_of_alcohol, Consumption_of_water_daily, Obesity)) %>%
  group_by(Consumption_of_alcohol, Consumption_of_water_daily, Obesity) %>%
  summarise(count = n()) %>%
  ungroup()
alcohol_water <- alcohol_water %>% mutate(label = glue("Total : {count}
                                                        Obesity Type : {Obesity}
                                                        Alcohol Consumption : {Consumption_of_alcohol}
                                                        Water Daily Consumption : {Consumption_of_water_daily}"))

plotd <- ggplot(alcohol_water, mapping = aes(x = Consumption_of_alcohol, y = count, fill = Obesity, text = label)) +
        geom_col(position = "dodge") +
        facet_wrap(vars(Consumption_of_water_daily)) +
        scale_fill_viridis(discrete = T, option = "C") +
        scale_x_discrete(expand = expansion(mult = c(0.0001, 0.0001)))+
        labs(title = "Daily Water Consumption",
             x = "Alcohol Consumption",
             y = NULL,
             fill = "Obesity Type") +
        theme(legend.title = element_blank(),
              axis.text.x = element_text(hjust =1, angle =20),
              plot.title = element_text(face = "bold"),
              panel.background = element_rect(fill = "#ffffff"),
              axis.line.y = element_line(colour = "grey"),
              axis.line.x = element_blank(),
              panel.grid = element_blank()) + coord_flip()
ggplotly(plotd, tooltip = "text")

## DATA CUSTOM 5
smoke_veggie <- obesity_new %>%
  select(c(Frequency_of_consumption_of_vegetables, Smoke, Obesity)) %>%
  group_by(Frequency_of_consumption_of_vegetables, Smoke, Obesity) %>%
  summarise(count = n()) %>%
  ungroup()
smoke_veggie <- smoke_veggie %>% mutate(label = glue("Total : {count}
                                                      Obesity Type : {Obesity}
                                                      Frequency of Vegetable Consumption : {Frequency_of_consumption_of_vegetables}
                                                      Smoke Type : {Smoke}"))

plote <- ggplot(smoke_veggie, mapping = aes(x = Frequency_of_consumption_of_vegetables, y = count, fill = Obesity, text = label)) +
        geom_col(position = "dodge") +
        facet_wrap(vars(Smoke)) +
        scale_fill_viridis(discrete = T, option = "C") +
        scale_x_discrete(expand = expansion(mult = c(0.0001, 0.0001)))+
        labs(title = "Smoke Category",
             x = "Frequency of Vegetable Consumption",
             y = NULL,
             fill = "Obesity Type") +
        theme(legend.title = element_blank(),
              axis.text.x = element_text(hjust = 1),
              plot.title = element_text(face = "bold"),
              panel.background = element_rect(fill = "#ffffff"),
              axis.line.y = element_line(colour = "grey"),
              axis.line.x = element_blank(),
              panel.grid = element_blank())
ggplotly(plote, tooltip = "text")

BMI Distribution on Age by Gender:

obesity_new$bmi <- obesity_new$Weight/(obesity_new$Height/100)**2
obesity_new$bmi <- round(obesity_new$bmi, 1)

obesity_bmi <- obesity_new %>%
  select(c(bmi, Age, Gender)) %>% mutate(label = glue("BMI : {bmi}
                                                        Age : {Age}
                                                        Gender : {Gender}"))

plotob_bmi <- ggplot(data = obesity_bmi, aes(x = Age, y = bmi, fill = Gender, text = label))+
  geom_point(aes(col = Gender), alpha = 0.5, col ="black") +
  labs(title = list(text = paste0('BMI on Age Distribution')),
    x = "Age",
    y = "BMI"
  ) +
  theme(legend.title = element_blank(),
        plot.title = element_text(face = "bold"),
        panel.background = element_rect(fill = "#ffffff"),
        axis.line.y = element_line(colour = "grey"),
        axis.line.x = element_line())
ggplotly(plotob_bmi, tooltip = "text")

Normal BMI on Adults (>= 18 y.o)

obesity_bmi_a <- obesity_bmi[obesity_bmi$Age >18 & obesity_bmi$bmi > 18.5 & obesity_bmi$bmi < 24.9,]
#obesity_bmi_a %>% group_by(Gender) %>% summarise(count = n())

plotob_bmi_a <- ggplot(data = obesity_bmi[obesity_bmi$Age >18 & obesity_bmi$bmi > 18.5 & obesity_bmi$bmi < 24.9,], aes(x = bmi, fill = Gender))+
  geom_histogram(position = "dodge", bins = 30) + 
  scale_y_continuous(breaks = seq(5,15,5), limits=c(0,15)) +
  labs(title = "Normal BMI on Adults",
             x = "BMI",
             y = NULL,
             fill = "Gender") +
        theme(legend.title = element_blank(),
              axis.text.x = element_text(hjust = 1),
              plot.title = element_text(face = "bold"),
              panel.background = element_rect(fill = "#ffffff"),
              axis.line.y = element_line(colour = "grey"),
              axis.line.x = element_blank(),
              panel.grid = element_line(colour = "black"))
ggplotly(plotob_bmi_a, tooltip = "text")

According to the graph, there are more adult females (56) than adult males (45) in the Normal BMI range who are above the age of 18. Furthermore, while this figure represents the Normal BMI range of 18.5 to 24.9, the majority of the sample is in the second half of the range, implying slightly heavier weight than the range’s median.