Matteo Bodini

Bio

I am Matteo Bodini, Ph.D., Data Manager at the Department of Economics, Management and Quantitative Methods of the University of Milan, and member of the INdAM (National Institute for Advanced Mathematics "Francesco Severi") within the GNCS group (National Group for Scientific Computing) since 2019.

I obtained a Ph.D. in Computer Science (XXXIV cycle) on January 24th 2022 at the Department of Computer Science "Giovanni Degli Antoni" of the University of Milan with a thesis entitled "Design and Explainability of Machine Learning Algorithms for the Classification of Cardiac Abnormalities from Electrocardiogram Signals". Before starting my Ph.D. in 2018, I received a M.Sc., full marks with honors, in Computer Science from the University of Milan in July 2018, discussing a thesis entitled "Probabilistic Dynamical Models for the Analysis of Affective Facial Expressions", focused on emotion recognition with machine learning algorithms (thesis and publication are available for downloading).

During my Ph.D. I was part of the BiSP Lab (Biomedical Image and Signal Processing Lab) and my research interests were focused on the areas of machine learning and deep learning. In particular, I was interested in the problem of explainability of machine learning and deep learning algorithms. In the context of my Ph.D. I applied and explained machine learning and deep learning algorithms in the task of heart diseases recognition through electrocardiogram signals. I was even interested in time series analysis: I worked on the design and testing of new entropy measures, with applications to the context of bioengineering.

In addition to my work of Data Manager, I even enjoy teaching: so far I did 300+ funded hours as Professor and teaching assistant in computer science related courses at the Department of Computer Science, Department of Physics, Department of Economics, Management and Quantitative Methods, and Department of Political, Economical and Social Sciences, at University of Milan, both for bachelor and master degree programs. You can read more information about the courses for which I am (or I was) Professor or teaching assistant in the Teaching section of this website.

I can be reached at matteo.bodini@unimi.it. Currently I'm based at the Department of Economics, Management and Quantitative Methods, at University of Milan, Via Conservatorio 7, 20122 Milano, Italy, in the room 38 (intermediate floor between 1st and 2nd floor).

Journal papers

Here you can find my journal papers.

Opinion mining from machine translated Bangla reviews with stacked contractive auto-encoders
Matteo Bodini
Journal of Ambient Intelligence and Humanized Computing (2022).
[paper (open access)]

Opening the black box: interpretability of machine learning algorithms in electrocardiography
Matteo Bodini, Massimo Walter Rivolta, and Roberto Sassi
Philosophical Transactions of the Royal Society A. 2021; 379(2212): 20200253.
[paper, preprint]

A Two-Steps-Ahead Estimator for Bubble Entropy
George Manis, Matteo Bodini, Massimo Walter Rivolta, and Roberto Sassi
Entropy. 2021; 23(6):761.
[paper]

Aspect Extraction from Bangla Reviews Through Stacked Auto-Encoders
Matteo Bodini
Data. 2019; 4(3):121.
[paper]

Will the Machine Like Your Image? Automatic Assessment of Beauty in Images with Machine Learning Techniques
Matteo Bodini
Inventions. 2019; 4(3):34.
[paper]

A Review of Facial Landmark Extraction in 2D Images and Videos Using Deep Learning
Matteo Bodini
Big Data and Cognitive Computing. 2019; 3(1):14.
[paper]

Conference papers

Here you can find my conference papers.

Bubble Entropy of Fractional Gaussian Noise and Fractional Brownian Motion
George Manis, Matteo Bodini, Massimo Walter Rivolta and Roberto Sassi
Computing in Cardiology Conference (CinC), 2021
[paper available soon]

Classification of ECG Signals with Different Lead Systems Using AutoML
Matteo Bodini, Massimo Walter Rivolta, and Roberto Sassi
Computing in Cardiology Conference (CinC), 2021
[paper available soon]

Classification of 12-lead ECG with an Ensemble Machine Learning Approach
Matteo Bodini, Massimo Walter Rivolta, and Roberto Sassi
Computing in Cardiology Conference (CinC), 2020
[paper (open access)]

Interpretability Analysis of Machine Learning Algorithms in the Detection of ST-Elevation Myocardial Infarction
Matteo Bodini, Massimo Walter Rivolta, and Roberto Sassi
Computing in Cardiology Conference (CinC), 2020
[paper (open access), slides]

Analytical Formulation of Bubble Entropy for Autoregressive Processes
Matteo Bodini, Massimo Walter Rivolta, George Manis, and Roberto Sassi
11th Conference of the European Study Group on Cardiovascular Oscillations (ESGCO), 2020
[paper, preprint, slides]

Predictive sampling of facial expression dynamics driven by a latent action space
Giuseppe Boccignone, Matteo Bodini, Vittorio Cuculo, and Giuliano Grossi
IEEE 14th International Conference on Signal Image Technology & Internet Based Systems (SITIS), 2018
[paper, preprint, slides]

Single Sample Face Recognition by Sparse Recovery of Deep-learned LDA Features
Matteo Bodini, Alessandro D'Amelio, Giuliano Grossi, Raffaella Lanzarotti, and Jianyi Lin
Advanced Concepts for Intelligent Vision Systems (ACIVS), 2018
[paper, preprint, slides, code]

A New Proof of Dobiński’s Formula via an ad hoc Algebraic Structure
Matteo Bodini, Ottavio M. D’Antona, and Mauro Torelli
Combinatorics, 2018
[abstract, preprint, slides]