Mohamed-Assaf-UM/Algerian_Forest_Fires

# Algerian Forest Fires This project focuses on analyzing and classifying forest fire occurrences in two regions of Algeria: Bejaia and Sidi Bel-Abbes. The dataset, which consists of 244 instances, includes various meteorological and fire weather index components collected from June to September 2012. The data is divided into two classes: fire and not fire. The goal is to clean, preprocess, and analyze the data before applying a classification model to predict the occurrence of fires. ### Dataset Description The dataset includes 11 input features and 1 output feature: - **Date**: Represented by day, month, and year (2012). - **Temperature**: Maximum temperature at noon, ranging from 22°C to 42°C. - **RH**: Relative Humidity, ranging from 21% to 90%. - **Ws**: Wind speed in km/h, ranging from 6 to 29 km/h. - **Rain**: Total daily precipitation, ranging from 0 to 16.8 mm. - **Fire Weather Index (FWI) Components**: - **FFMC**: Fine Fuel Moisture Code (28.6 to 92.5). - **DMC**: Duff Moisture Code (1.1 to 65.9). - **DC**: Drought Code (7 to 220.4). - **ISI**: Initial Spread Index (0 to 18.5). - **BUI**: Buildup Index (1.1 to 68). - **FWI**: Fire Weather Index (0 to 31.1). - **Class**: Binary classification into two categories — fire or not fire. ### Data Preprocessing The dataset was preprocessed to handle missing values and correct inconsistent data entries: - Rows with null values were identified and removed. - A new column was added to distinguish between the two regions, with Bejaia labeled as `0` and Sidi Bel-Abbes as `1`. - Spaces in column names were stripped to ensure clean data representation. - Data types of key columns, such as day, month, year, temperature, and humidity, were converted from string to integer for more accurate analysis and modeling. ### Feature Transformation and Cleaning - The dataset contained some rows with inconsistencies, such as placeholder text in numerical fields. These were carefully handled and cleaned. - An additional region colu …