/Master-Thesis

All the code used for my MSc Thesis: Search for Dark Matter using Machine Learning in dilepton and missing energy events with the ATLAS detector at the LHC, A tentative model independent approach

Primary LanguagePython

Master Thesis

Welcome to the repository for my Master's thesis titled "Search for Dark Matter using Machine Learning in dilepton and missing energy events with the ATLAS detector at the LHC, A tentative model independent approach" This repository contains the code and resources related to the thesis project. The repository is organized into several categories:

DataPrep

The DataPrep directory contains code and scripts for two main purposes:

  • Plotting the kinematic variables used in machine learning (ML) analysis.
  • Preparing the data files in a ML-friendly format.

Inside the DataPrep directory you will find more detailed information about the specific scripts and their usage.

EventSelector

The EventSelector directory includes the algorithm used for event selection. It works with both datapoints and Monte Carlo simulations from the Large Hadron Collider (LHC). The selected events are transformed into readable files suitable for plotting and further ML analysis. For more information on the algorithms and usage, please refer to the EventSelector directory.

ML

The ML directory is dedicated to the main algorithms developed for training a Neural Network and Boosted Decision Tree. These algorithms are specifically designed to distinguish signal (dark matter) from background (the Standard Model). For detailed information about the algorithms and their implementation, please refer to the ML directory.

Plots

The Plots directory contains all the plots generated from the various scripts used in the analysis. These plots provide visual representations of the data and results obtained during the research. Feel free to explore the Plots directory for further insights.

StatAnalysis

The StatAnalysis directory contains scripts used for performing a Bayesian analysis. This analysis aims to set limits on the proposed models by utilizing statistical methods. If you are interested in the details of the Bayesian analysis and its implementation, please refer to the StatAnalysis directory.

Please navigate into each respective category directory for more detailed information on scripts, usage, and additional resources.


If you have any questions or inquiries regarding this repository or my Master's thesis, please feel free to contact me! Thank you for your interest in this project!