Cristina Monni is currently a post-doctoral researcher at the STAR group at USI Università della Svizzera Italiana.
Cristina got a PhD in theoretical physics in 2012, with a thesis on high-energy physics focused on gravitational effects in the vicinity of black holes.
Cristina was post-doc in software engineering at the AGILE group, University of Cagliari, from 2012 to 2015. In 2014 she spent six months as a visiting postdoctoral fellow at the Technion, Israel Institute of Technology, Israel.
In november 2014 Cristina joined officially the Computer Science department at the Technion, Israel Institute of Technology as a post-doctoral researcher.
In december 2015, Cristina joined the STAR group at USI to focus on models for cloud reliability.
Cristina's research topic at the STAR group is the development of new models and techniques for the reliability of complex IT infrastructures such as the cloud, focusing on anomaly detection, failure prediction and fault localisation.
In particular, Cristina develops techniques for time series analysis based on machine learning, deep learning, complex network theory and statistical hypothesis testing.
In the past years, Cristina worked on hypothesis testing and the statistical characterisation of software defectiveness, using graph theoretical methods for predicting the bug density in Object Oriented software (mainly Java).
Cristina recently developed an algorithm to improve the accuracy of fitting of fat-tailed distributions, such as power law, Yule, log-normal and Weibull.
Such distributions are usually chosen to fig big datasets coming from man-made phenomena, characterised by high levels of noise. The developed algorithm is an effective fitting procedure which decouples noise and gives better accuracy than other similar methods.