Identification of Higgs boson events from the background by machine learning.

Almost a decade since its detection in the Large Hadron Collider, the Higgs boson remains a key component in the confirmation of the standard model (SM) of particle physics. Current experimental efforts to determine the Higgs boson decay channels, predicted by the SM, rely on the accurate identification of Higgs boson events from a noisy background. Machine learning methods have emerged as new ways to accurately classify rare events present in large datasets. This study compares three supervised classification models in their ability to accurately and efficiently identify signal events: (i) boosted decision trees, (ii) support vector machine, and (iii) neural networks. With an accuracy of 83.88%, F1-score of 81.72% and a training time of 8.4 seconds using the simulated dataset from the 2014 ATLAS Higgs boson Machine Learning Challenge, a histogram gradient boosted decision tree was determined to be the most effective classification model for identifying Higgs boson events. The worst performing algorithm was the support vector machine with the lowest accuracy at 80.35%, F1-score of 77.12%, and the second lowest training time of 50 minutes.