Tailoring machine learning for tackling the Large Hadron Collider problems

Machine learning has proven its vitality in devising complex systems. The wide range of tasking techniques allowed machine learning to be a qualified tool for analyzing the vast amount of data obtained from the Large Hadron Collider. These tasks can range from tracking pattern recognition, particle identification, to search for rare decays. As a new tool starts to spread, the need for modifying increases to adapt to the new users' tasks. One approach is the design of packages. Here I present a machine learning package in R, designed for analyzing the Large Hadron Collider data, focusing mainly on particle identification through the measured particle parameters and the detector system, and also spotting rare decays using the supervised learning based classification techniques of machine learning. Machine learning does not only focus on easing the data analysis process that is already taking place, but it offers an opportunity for implementing new search strategies for rare phenomena through novel techniques. This step aims to assist physicists and researchers, and it also takes a further step ahead of enabling users to personalize more the tool through having an open source of the package accessible to users.