joedet/time_meal_dataset

# Dataset Feature Documentation

## User-Level Features
These describe individual characteristics that influence meal patterns and nutritional decisions.

---

### **Age (Numeric)**
- **Description:** Individual’s age in years. Simulated with different distributions across Meal_Time classes to improve model learning.
- **Range/Statistics:** Min=18, Max≈65, Mean≈40, Std≈9.5 (clipped 18–80).
- **Context:** Age influences metabolism, lifestyle, and meal preferences.

---

### **Gender (Categorical — Encoded: Female=0, Male=1)**
- **Description:** Binary gender with slight distribution differences across meal classes.
- **Categories:**
- **0 — Female**
- **1 — Male**
- **Context:** Gender may influence dietary energy requirements.

---

### **BMI (Numeric)**
- **Description:** Body Mass Index calculated from height/weight (simulated). Strongly correlated with Obesity_Stage.
- **Range:** 15–40 (Mean≈25, Std≈5).
- **Context:** Indicates body fatness; helps inform dietary patterns.

---

### **Obesity_Stage (Categorical — Encoded)**
- **Description:** Weight-status class derived from BMI with minor random noise.
- **Categories:**
- **0 — Normal (BMI < 25)**
- **4 — Overweight (25–29.9)**
- **1 — Obese I (30–34.9)**
- **2 — Obese II (35–39.9)**
- **3 — Obese III (≥40)**
- **Context:** Helps understand health risk and likely meal types.

---

### **Activity_Level (Categorical — Encoded)**
- **Description:** Physical activity intensity.
- **Categories:**
- **0 — Active**
- **1 — Moderate**
- **2 — Sedentary**
- **3 — Very Active**
- **Context:** Higher activity → more calories/protein.

---

### **Diet_Preference (Categorical — Encoded)**
- **Description:** Individual eating style.
- **Categories:**
- **0 — High-Protein**
- **1 — Low-Carb**
- **2 — Omnivore**
- **3 — Vegan**
- **4 — Vegetarian**
- **Context:** Influences meal composition and caloric intake.

---

## Food-Level Features

### **Portion_Size (Categorical — Encoded)**
- **Description:** Relative quantity of food consumed.
- **Categories:**
- **2 — Small**
- **1 — Medium**
- **0 — Large**
- **Context:** Affects total calorie intake and meal suitability.

---

### **Cuisine_Type (Categorical — Encoded)**
- **Description:** Type of cuisine for the meal.
- **Categories & Examples:**

#### **0 — Continental**
Western European–inspired meals.
Examples: calamari, chicken pie, baked broccoli, spaghetti carbonara.

#### **1 — Fast Food**
Quick-service meals, often high-calorie.
Examples: fries, burgers, pizza, tacos, nuggets.

#### **2 — Nigerian**
Mainstream Nigerian staples.
Examples: jollof rice, suya, efo-riro, moin-moin, pounded yam.

#### **3 — Traditional Local**
Indigenous or hyper-local Nigerian dishes.
Examples: okpa, ukwa, abacha, porridge yam, tuwo masara.

- **Context:** Cuisine affects nutritional profile and meal timing.

---

## Nutritional Indicators

### **Calories (Categorical — Encoded)**
- **Description:** Caloric density group instead of exact values.
- **Categories:**
- **1 — Low (<400–500 cal)**
- **2 — Medium (~500–700 cal)**
- **0 — High (>700–900 cal)**
- **Context:** Allows meal-time suitability estimation.

---

## Target Variable

### **Meal_Time (Categorical — Encoded)**
- **Description:** Final label to predict.
- **Categories:**
- **0 — Breakfast**
- **1 — Dinner**
- **2 — Lunch**
- **Context:** Classification objective based on user and meal attributes.