Data Science Projects

Modeling of Area Burned by Forest Fires

view project
project

The objective of this study was to forecast the area burned by forest fires in Portugal's northern region. Machine learning was used to make this prediction using linear regression. Python was employed in this repository, together with libraries and packages such as Matplotlib, Numpy, Scikit-learn, and Seaborn.

project

The data set underwent pre-processing, data wrangling, and exploratory data analysis to identify variables that had high correlations with the target variable and to filter out major outliers. Our own modeling and testing cast doubt on the validity of linear regression as a reliable model to predict environmental factors, even though the authors of this data set built their own linear regression model. We delve further into the problems with the data collected and the difficulties it presents.

Shopping Behaviour Analysis

view project
project

This project is a shopping behaviour analysis based on a sample data of the american retail market. Through exploratory data analysis and data cleaning we could observe the means of each variable such as age, season and common modalitys of purchase.