Portfolio - Francisca Maron

Portfolio

About

I am an Environmental Scientist with a Masters degree in Marine Science and Management from the University of Sydney, and a Masters degree in Applied Data Science from the University of Adelaide.

Early in my career, I learned how to code in R and Python in order to analyse environmental data, which made me realise my passion for programming and data science. I am very interested in contributing to the conservation of biodiversity, conducting relevant analyses and creating engaging visualisations that empower stakeholders to make data-driven decisions.

I am highly motivated to learn and implement new technologies, leveraging the skills and experience gained throughout my career.

In my spare time, I like going bouldering and rock climbing with friends, and participating in various outdoor activities such us hiking and camping. I also enjoy playing board games with friends and reading fiction books.

Download CV

Contact Me

Full name

Email address

Phone number

Message

University Projects

Python Project

The aim of this study was to find the best-suited model to predict bike sharing usage. The data used to train the model is from the bike-sharing system “Capital Bike Sharing”, located in Washington, D.C, USA. The conducted models were Random Forest, Linear Regression, Polynomial Regression and Support Vector Machine Regression. In order to compare the models, the Mean Square Error (MSE) and Root Mean Square Error (RMSE) were calculated. The final model was the one that presented the lowest MSE and RMSE, which in this case was the Support Vector Machine Regression with hyperparameter tuning. The project is divided into 5 Jupyter notebooks:

Exploratory Data Analysis

Random Forest

Linear Regression

Polynomial Regression

SVM Regression

R Project

The aim of this project was to predict the number of sales in North America for a PlayStation 4 game “The Fatal Empire” (Note: this is not a real game) for a gaming company. The game has sold 2.58 million copies in Japan, 0.53 million copies in Europe, and 0.1 million copies in other world regions.
In this project, the prediction of sales in North America was performed using two different predictive models (i.e. Lasso regression and Random Forest). The final model was selected by evaluating the highest performing model.

Full Report and Code

Julia Project

The objective of this project was to create a simulation model of a lawn mower factory. The model was focused on a particular machine that will breakdown at a specified time (150 minutes). The arrivals at the machine followed an exponential distribution, same as the breakdown time and the repair time. The machine took 25 minutes building a lawn mower. The code of the simulation and the simulation harness can be found below:

Factory Simulation