dmrks/Algerian_Forest_Fires_Project

# AlgerianForestFires_Project Algerian Forest Fires In this project, you’ll examine the forests dataset that includes variables on weather and fire risk for locations in Algerian forests from two different regions. By plotting and running multiple linear regressions, you’ll explore the relationships among variables including: temp – maximum temperature in degrees Celsius humid – relative humidity as a percentage region – location in Bejaia in the northeast of Algeria or Sidi Bel-abbes in the northwest of Algeria fire – whether a fire occurred (True) or didn’t (False) FFMC – Fine Fuel Moisture Code: measure of forest litter fuel moisture that incorporates temperature, humidity, wind, and rain ISI – Initial Spread Index: estimates spread potential of fire BUI – Buildup Index: estimates potential release of heat FWI – Fire Weather Index: measure of general fire intensity potential that incorporates ISI and BUI The dataset is a subset of data that was downloaded from the UCI Machine Learning Repository and then cleaned for analysis. Faroudja ABID et al., Predicting Forest Fire in Algeria using Data Mining Techniques: Case Study of the Decision Tree Algorithm, International Conference on Advanced Intelligent Systems for Sustainable Development (AI2SD 2019) , 08 - 11 July , 2019, Marrakech, Morocco. # Predicting Relative Humidity 1. The dataset has been loaded for you as forests. Since we will be running several multiple linear regressions in this project, we want to check the multicollinearity assumption before we get started. Create a table of correlations for the quantitative variables in forests and call it corr_grid. Then plot the correlations with a heat map to look for potential collinear variables. Make sure to show, then clear the plot. Are there any variables that should not go into a model together as predictors? 2. Let’s explore the relationship between relative humidity (humid) and maximum temperature (temp) by creating a scatter plot with humid on …