Polyglot Data Science Example

mkdir output

library(hrbrthemes) # For a clean typography-focused theme

df %>%
  group_by(Sleep_Hours) %>%
  summarise(mean_coffee = mean(Coffee_Intake)) %>%
  ggplot(aes(x = Sleep_Hours, y = mean_coffee)) +
  # Use points and lines together for better readability
  geom_line(color = "#6F4E37", size = 1.2, alpha = 0.8) +
  geom_point(color = "#6F4E37", size = 3) +
  # Add a smoothing line to see the overall trend
  geom_smooth(method = "lm", color = "darkred", linetype = "dashed", se = FALSE) +
  # Polish the labels and theme
  labs(
    title = "Does Coffee Intake Affect Sleep Duration?",
    subtitle = "Average cups of coffee vs. hours of sleep recorded",
    x = "Sleep Duration (Hours)",
    y = "Average Coffee Intake (Cups)",
    caption = "Data source: Revised Coffee Health Dataset"
  ) +
  ## theme_ipsum() + # Or theme_minimal() if you don't have hrbrthemes
  theme_ipsum() +
  theme(
    plot.title = element_text(face = "bold", size = 16),
    axis.title = element_text(size = 12)
  )

ggplot(df, aes(y = BMI, x = "")) +
  geom_boxplot(fill = "#6F4E37", color = "#3C2A21", alpha = 0.7, notch = TRUE) +
  geom_jitter(color = "black", size = 0.5, alpha = 0.2, width = 0.1) +
  labs(
    title = "Distribution of BMI",
    subtitle = "Analysis of 10,000 coffee consumers",
    y = "Body Mass Index (BMI)",
    x = NULL
  ) +
  theme_minimal() +
  theme(plot.title = element_text(face = "bold"))

Q1 <- quantile(df$BMI, .25)
Q3 <- quantile(df$BMI, .75)
overwt <- 24.9
underwt <- 18.5

# Number of overweight outliers of 10k sample size
nrow(df[df$BMI > overwt,])/10000

# Number of underweight outliers of 10k sample size
nrow(df[df$BMI < underwt,])/10000

df %>% mutate(Body_Condition = factor(case_when(
                BMI < underwt ~"low",
                BMI >= underwt & BMI < overwt ~ "normal",
                BMI >= overwt ~ "high"
              ))) -> df

summary(df$Body_Condition)

ggplot(df, aes(x = Body_Condition)) +
  geom_bar(fill = c("lightblue", "lightgreen", "coral"), alpha = 0.8) +  # Custom colors for bars
  labs(title = "Distribution of Body Conditions",
       x = "Body Condition",
       y = "Count") +
  theme_minimal(base_size = 15) +  # Base font size for readability
  theme(plot.title = element_text(hjust = 0.5, size = 20, face = "bold"),  # Centered title
        plot.subtitle = element_text(hjust = 0.5, size = 14, face = "italic"),
        axis.title.x = element_text(size = 16),
        axis.title.y = element_text(size = 16),
        axis.text.x = element_text(size = 14),
        axis.text.y = element_text(size = 14),
        panel.grid.major.y = element_line(color = "grey", linetype = "dashed"),  # Customize grid lines
        panel.grid.minor = element_blank()) +
  geom_text(stat = 'count', aes(label = ..count..), position = position_stack(vjust = 0.5), size = 5, color = "black")  # Add count labels

# Espresso, Cool Taupe, Oat Milk
coffee_colors <- c("low" = "#D9B99B", "normal" = "#7D6B5D", "high" = "#241611")

ggplot(df, aes(x = Body_Condition, y = Coffee_Intake, fill = Body_Condition)) +
  # Adds a "notched" boxplot to show confidence intervals of the median
  geom_boxplot(notch = TRUE, alpha = 0.7, color = "#241611", outlier.shape = NA) +
  # Adds jittered points to show the actual distribution of 10k rows
  geom_jitter(width = 0.2, alpha = 0.1, size = 0.5, color = "#241611") +
  scale_fill_manual(values = coffee_colors) +
  labs(
    title = "Coffee Consumption by Body Condition",
    subtitle = "Comparing intake levels across self-reported health states",
    x = "Body Condition",
    y = "Daily Coffee Intake (Cups)"
  ) +
  theme_minimal(base_size = 14) +
  theme(
    legend.position = "none",
    plot.title = element_text(face = "bold", color = "#241611"),
    panel.grid.major.x = element_blank()
  )

prop.table(table(df$Smoking))
prop.table(table(df$Alcohol_Consumption))

library(tidyverse)
library(scales)

df %>%
  # Select and clean up data
  select(Smoking, Alcohol_Consumption) %>%
  pivot_longer(everything(), names_to = "Metric", values_to = "Status") %>%
  mutate(
    Metric = gsub("_", " ", Metric),
    Status = if_else(Status == TRUE, "Yes", "No")
  ) %>%
  # Calculate percentages for labels
  count(Metric, Status) %>%
  group_by(Metric) %>%
  mutate(pct = n / sum(n)) %>%

  # Plot
  ggplot(aes(x = Metric, y = pct, fill = Status)) +
  geom_col(position = "fill", width = 0.7, color = "white", size = 0.5) +
  # Add direct labels inside the bars
  geom_text(aes(label = percent(pct, accuracy = 0.1)),
            position = position_stack(vjust = 0.5),
            color = "white", fontface = "bold") +
  scale_y_continuous(labels = percent) +
  scale_fill_manual(values = c("No" = "#A67B5B", "Yes" = "#3C2A21")) +
  labs(
    title = "Lifestyle Distribution: Smoking vs. Alcohol",
    subtitle = "Relative proportions within the consumer population",
    x = NULL, y = "Percentage of Participants",
    fill = "Status:"
  ) +
  theme_minimal(base_size = 14) +
  theme(
    panel.grid.major.x = element_blank(),
    legend.position = "top",
    plot.title = element_text(face = "bold")
  )

library(ggplot2)

ggplot(df, aes(x = Caffeine_mg, y = Heart_Rate)) +
  # Add points with transparency to handle 10k rows
  geom_point(alpha = 0.2, color = "#6F4E37", size = 1) +
  # Add a trend line
  geom_smooth(method = "lm", color = "#E44D2E", fill = "#FFD700", alpha = 0.2) +
  labs(
    title = "Caffeine Intake vs. Heart Rate",
    subtitle = "Assessing the physiological impact of caffeine levels",
    x = "Caffeine (mg)",
    y = "Heart Rate (BPM)"
  ) +
  theme_minimal() +
  theme(plot.title = element_text(face = "bold"))

library(ggplot2)

ggplot(df, aes(x = Coffee_Intake, fill = Gender)) +
  geom_density(alpha = 0.5, color = "white") +
  scale_fill_manual(values = c("Male" = "#241611", "Female" = "#7D685D", "Other" = "#D9B99B")) +
  labs(
    title = "Coffee Intake Distribution by Gender",
    x = "Cups of Coffee per Day",
    y = "Density",
    fill = "Gender"
  ) +
  theme_minimal() +
  theme(legend.position = "top", plot.title = element_text(face = "bold"))