Research background
My research background lie in the development and application of machine learning models and Data Analysis technique fro different context (e.g industrial, environmental data, ecc) , with a strong focus on data quality, model reliability, and the integration of data-driven approaches into sustainable decision-making processes.
Academic background
I studied at the University of Modena and Reggio Emilia (Emilia-Romagna, Italy). My academic journey began with a Bachelor’s degree in Management Engineering. My thesis examined the application of Life Cycle Assessment (LCA) to foundry processes. The study focused on the environmental impact of the grey cast iron life cycle, from raw materials to the final product, using an analytical model to evaluate the contribution of the casting process within the supply chain. The assessment was carried out using openLCA software and the ReCiPe 2016 Endpoint (H) method, which evaluates impacts on human health, terrestrial ecosystems, and aquatic ecosystems. The results showed that induction furnace melting, as well as casting shakeout and cleaning operations, are the most impactful stages due to emissions of fumes and dust during production. The research aimed to support the development of LCA models integrated into process simulation tools to improve sustainability in manufacturing.
I then continued my academic journey with a Master’s degree in Digital Automation Engineering at the same university. My thesis focused on the analysis and modeling of soil chemical properties in vineyards using data-driven approaches. The work began with a data preprocessing phase, including outlier detection, handling of missing values, and the analysis of correlations between chemical variables. An Exploratory Data Analysis (EDA) was then performed to better understand the data structure and identify the most relevant variables for predicting soil chemical parameters. Subsequently, several machine learning algorithms were applied, including Gradient Boosting (GB), Random Forest (RF), and Partial Least Squares Regression (PLSR), in order to build predictive models capable of estimating soil chemical properties from sensor-derived measurements. The models were evaluated and compared based on their predictive accuracy and generalization performance to determine the most effective approach for applications in precision agriculture and efficient, sustainable soil management.
In 2025, I attended the Summer School on Agriculture, Forest and Environmental Geodata: Statistical Analysis, Modelling and Machine Learning. The program aimed to simulate the integration of different data sources, data handling strategies, and modelling procedures for geodata, and to support modelling activities in various fields, including agriculture, forest management, soil management, soil biodiversity, and their relationships with economic, environmental, and social indicators. SAFEST provided both theoretical and practical training on statistical methods and tools for geodata handling and analysis, with special emphasis on soil traits, crop yield, and plant biomass across agricultural, forest, and other land uses under variable management and environmental conditions. The methods were tailored to a wide range of scales, from plot to landscape to regional levels. The school was supported by the SHARInG-MeD project (PRIMA 2022) and the SUS-SOIL project (Horizon Europe).
Current researches
At IPCMS, I am focusing my research on the development and implementation of machine learning tools for data interoperability and correlation analysis in computational materials science for energy applications. More specifically, my work has two main objectives. The first is to assess the performance of Machine Learning Interatomic Potentials used to study both amorphous and crystalline phases of GST materials. The second objective is to investigate potential correlations between structural and thermal properties of Phase-Change Materials, with a particular focus on GST.