Here, I share with you several projects where I try to learn more about data science techniques with hands-on experience and have fun at the same time!

  • Modelling the Influence of Weather Conditions on Traffic Accidents Using Machine Learning Algorithms

    The goal of this model is to help emergency services estimate and optimize resources based on weather conditions and the scope is the city of Calgary. It would help police services make a proactive plan for unit deployment to reduce emergency response times and help the city plan snow clearing services. Several supervised and unsupervised learning models are used such as linear regression, KNN, random forest, DBSCAN, K-means and mean shift.

    The final report, codes and data can be accessed on my GitHub repository.

  • Time Series Analysis - Global Warming

    The goal of this study is trying to confirm if global warming exists and to create critical thinking about the temperature tendency. I was motivated to work on this after seeing the popularity increase of weird theories like Earth being flat.

    The dataset is a compilation of several global temperature reports from 1900 to 2015 consolidated by a third-party organization extracted from Kaggle.

    Linear regression assumptions, autocorrelation and time series models are discussed here.

  • Mobile Price Prediction - SVM and Logistic Regression Comparison

    It was a quick project where I analyze the differences between SVM and logistic regression as classifier. The objective is to predict the class of mobile selling price based on its features. It is intended to help businesses when it is time to sell their products.