Design and development of a clustering algorithm to compute and simulate people behaviours
keywords DATA ANALYSIS, DATA ANALYTICS, DATA MINING, MODEL-BASED APPROACH, MODELING
Reference persons ANDREA ACQUAVIVA, LORENZO BOTTACCIOLI, EDOARDO PATTI
Research Groups ELECTRONIC DESIGN AUTOMATION - EDA, Energy Center Lab, GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ICT4SS - ICT FOR SMART SOCIETIES
Thesis type EXPERIMENTAL
Description This Thesis aims at analyzing information about the "Time of Use", an ISTAT (National Institute of Statistics) survey, for the years 2008 and 2013. The "Time of use" survey consists on daily individual diaries where respondents reported their daily activities with 10 minutes time steps.
The main objective of this thesis consists on developing and testing an algorithm to cluster users according to their different behaviors and their categories (example of clustering features are: age, degree, qualification, job and gender)
The second objective of this thesis consists on improving and enhance a multi-agent simulator with the clustering algorithm.
The multi-agent simulator has been developed in our previous work. It emulates the activities of people in a city and translate them into energy consumption. This simulator helps on testing novel services (e.g. Demand Response and Demand Side Management) and new business strategies in a Smart Grid scenario.
Deadline 13/03/2021 PROPONI LA TUA CANDIDATURA