1. Deep Learning: Predict Authorship of Tweets using Neural Networks

HEC Course: Deep Learning - I.Rudnytskyi

Today, American politicians frequently use Twitter to communicate their opinions and gain popularity. Based on differents datasets, we decided to test if an algorithm can recognise the different authors of the tweets. This would enable us to see if they have actually been written by the mentioned author (to avoid any hacking).

Click here to access the website

2.1 Data Visualization and Modeling: Explain Real Estate Price with Criminality and School Ranking in Washingthon DC

HEC Course: Data Science in Business Analytics - T.Vatter

The aim of my analysis was to determine if it is possible to predict real estate price in Washington DC based on school ranking and criminality data.

Firstly, I have decided to conduct an exploratory analysis to see if there is a link between property prices, schools, and crime. For this, I used interactive maps using the Leaflet Package. Secondly, I run four different generalized linear models, to have additional insight during the model comparison.

Click here to access the website

2.2 Shiny App: Explain Real Estate Price with Criminality and School Ranking in Washingthon DC

I developped this Shiny App as a add-on to my work on Real Estate in Washington DC. It works with the same datasets and can be used to estimate real estate price based on differents variables.

Click here to access the website

3. Forecasting: 28 days forecast of aggregated unit sales in Wisconsin store

HEC Course: Forecasting II - M.Wilhelm

This paper aims to predict sales using data provided by Walmart for several stores in the US. KAGGLE COMPETITION: The M Forecasting competitions are a series of events which test the forecasting ability of many different methods and models

Click here to access the website

Last update: 10/22