A model-independent "General Search" for new physics with the ATLAS detector at LHC

The LHC particle collider accelerates bunches of colliding protons at an energy never reached before, and a completely new landscape of new physics has been opened. In this scenario the number of possible physics processes and signatures becomes virtually infinite, making the setup of dedicated analyses impossible. Moreover it is important being able to reveal new physics signals even in regions of the phase-space where it is less lucky to be found, or where suitable theoretical models are missing. In this Thesis a new model-independent “General Search” for the ATLAS experiment has been conceived. In fact, at the time this project started, no model-independent search was set for ATLAS. In the end the new analysis has been run over the first data at 7 TeV collected by the ATLAS experiment, and the results presented. The data have shown a very good agreement with the Standard Model expectation, and no evidence of new physics has been observed. But the strategy and methodology of the new model-independent General Search have been defined, ready to be used in a future version of this analysis, over a larger set of experimental data. During this work, a new innovative software framework has been also conceived and implemented. The package has been designed to ease the implementation of physics analysis code, using Computer-Aided Software Engineering (CASE) principles. The framework has been successfully used to analyze the very first LHC data, and then it has been transformed into a modular package, and presented to Physics and Computer Science international conferences, as a public open-source framework for data analysis.