DOI: 10.28979/jarnas.1242840 ISSN:

An Educational Approach Higgs Boson Hunting using Machine Learning Classification Algorithms on ATLAS Open Data

Ayşe BAT
  • General Medicine
Show PDF Cite
In this study, the performance of different classification algorithms used to distinguish the signal which is H→ττ from background events are investigated. The open source data by the ATLAS experiment for the machine learning competition was used as the data set. The study generally focuses on making the signal and background separation of machine learning algorithms in terms of high-energy physics. This article compares the results of Linear Support Vector Machines (SVM), Radical SVM, Logistic Regression, k-Nearest Neighbors (KNN), XGBoost classifier, and AdaBoost classifier algorithms. The XGBoost classifier algorithm gave the best results with an AUC of 0.84 ±0.259 x 10^(-2). Radical SVM and AdaBoost algorithms were the second-best algorithms with the same 0.81 AUC. However, the AdaBoost algorithm shows more stable results statistically.

More from our Archive