Veronica Distefano is postdoctoral researcher at ECLT. She graduated from the University of Salento and obtained the PhD in Statistics at the same university. She was awarded a fellowship for a postdoc position in that university for developing research on strategic management of health data sets and for developing computational models to investigate cohorts of long-term cancer patients. As a postdoc at ECLT she is contributing by developing sparse principal components analysis and random forest models for big data sets.
Her research interests regard the development of new analytical methodologies to analyze the information contained in different type of data (environmental, social and economic) for big data sets. Currently she is dealing with the socio-economic data and its disparate application for the analysis of multidimensional systems, the variables selection in social and economic applications and forecasting. Among the tools used Sparse Principal Components analysis, Random Forest model and Clustering Algorithms. The recent research is devoted to analyze and modeling big data and develop procedures to summarize the main characteristics of data and identify outliers. Techniques for this aim include cluster analysis and dimension reduction. In addition, the integration of PCA with optimization tools such as Genetic Algorithm (GA) has also found relevant application in recent works. Other research topics include Spatial Analysis (with GIS) and Geostatistical Techniques among these methods Ordinary Kriging (OK) and CoKriging are techniques for estimating or predicting the spatial phenomenon.
High Dimensional Models, Statistical Modelling, Clustering; Principal component analysis, sparse principal component analysis (sparse PCA).
Last update: 12/01/2021