✈️ Flight Delay Analysis & Prediction

This project analyzes airline flight delay data and builds models to predict both:

• Whether a flight will be delayed (classification).
• How many minutes a flight will be delayed (regression).

It uses EDA, data cleaning, and machine learning on features like carrier delay counts, weather delays, NAS delays, late aircraft delays, and more.

---

📊 Features

✅ Exploratory Data Analysis (EDA) with visualizations:

• Histograms, boxplots, heatmaps of delay causes.
• Correlation heatmaps.

✅ Classification model:

• Predicts if a flight will be delayed (>15 minutes) or on time.
• RandomForestClassifier with performance metrics and confusion matrix.

✅ Regression model:

• Predicts exact arrival delay in minutes.
• LinearRegression with error metrics and actual vs. predicted plots.

✅ Prediction on new data:

• Generate delay predictions for custom flight scenarios.

✅ Cleaned dataset:

• Missing values handled.
• Outliers filtered.

---

📁 Dataset

Uses cleaned flight delay data (CleanedData.csv) containing columns like:

• carrier_ct, weather_ct, nas_ct, late_aircraft_ct
• carrier_delay, weather_delay, nas_delay, late_aircraft_delay
• arr_cancelled, arr_diverted, month
• arr_delay (target for regression)

---

🚀 How to Run

1. Clone the repository:

   git clone https://github.com/sargamjain5/FlightDelay.git
   cd FlightDelay