Project 1 of the Pattern Classification and Machine Learning EPFL Course
Authors: Baptiste Raemy, Cem Orhan, Dario Martinez
This project presents a real world implementation of the techniques described in the Pattern Classification and Machine Learning course at EPFL. The data set used corresponds to actual CERN particle accelerator data to recreate the process of "discovering" the Higgs particle, by reducing the problem into classifying as signal and background. After a first exploratory data analysis, the most important features were identified. Feature processing was combined with each linear model studied in class to make a prediction and determine which method performs the best given the data.
Leaderboard Position: 16/117 https://inclass.kaggle.com/c/epfml-project-1/leaderboard