
Dettaglio Insegnamento
Anno Accademico 
2021/2022 
Corso di Studi 
Dott.  MI (1387) Data Analytics and Decision Sciences 
Anno di Corso 
1 
Codice Identificativo 
057332 
Denominazione Insegnamento 
TIMESERIES EXPLORATION WITH MACHINE AND DEEP LEARNING: FROM THEORY TO PRACTICE 
Tipo Insegnamento 
MONODISCIPLINARE 
Crediti Formativi Universitari (CFU) 
5.0 
Programma sintetico 
The course presents, in an integrated and comprehensive way, theoretical and practical aspects for the analysis and prediction of timestructured data. The course aspires at providing students an endtoend pipeline for timeseries analysis/prediction from data wrangling to model evaluation by also introducing novel "asaservice" approaches in the field of timeseries forecasting. Theoretical lectures will be complemented with practical lectures describing Python libraries, frameworks, and toolboxes for timeseries analysis and prediction. Finally, a specific focus is given on hands on application cases by considering three realworld application scenarios where theory and practice will be applied.
In more detail, the course is organized into the following theoretical (T) and practical (P) lectures (2h):
1) Introduction to timestructured data (T)
2) Data wrangling for timeseries (T)
3) Exploratory data analysis for timeseries (T)
4) Tools and mechanisms for timeseries wrangling and exploration (P)
5) Statistical models for time series prediction (T)
6) Handson statistical models for time series (P)
7) Machine learning for time series prediction (T)
8) Handson machine learning for time series (P)
9) Deep learning for time series prediction (T)
10) Handson deep learning for time series (P)
11) Measuring and evaluating the error (T)
12) Forecastingasaservice (T)
13) Handson applications in healthcare, finance and environmental monitoring (P) 
Settori Scientifico Disciplinari (SSD) 


Scaglione

Docente

Programma dettagliato

Da (compreso)

A (escluso)

A

ZZZZ

Roveri Manuel


