PhD in Computer Science

Data Fusion for Smarter Cities

Academic year 2012/2013

Sommario del corso

• Course objectives
• Support activities
• Program
• Teaching tools

Course objectives

In this seminar I will present the work that I am doing at IBM Research Ireland lab 1 ,

and give an overview of the major research topics in the lab.

In the first part of the seminar, I will present queriocity, a platform for

querying integrated representations of heterogeneous city data. Queriocity supports

cataloging, exploring, integrating, processing, and transforming urban information.

Such information comes from the Web, government authorities, social networks, data

streams and Linked Data sources. The ultimate goal of queriocity is to simplify access

to the huge data spaces originating from urban environments, and to make it easier to

find and extract (aggregated) datasets to feed complex analytics and machine learning

tasks. We are experimenting queriocity with public and restricted data provided by

Dublin (Ireland).

In the second part of the seminar, I will present our automated traffic diagnoser.

Based on data extracted from queriocity, our system produces quasi real-time

automated explanations of traffic congestions. We combine reasoning over semantic

representations of historical data (such as traffic congestions data) with an AI diagno-

sis approach. We are testing an initial implementation of the system with real data,

and we are planning a trial with Dublin city.

Support activities

The seminar will be held by

Marco Luca Sbodio

IBM Ireland Research Lab

SHORT BIO:

I am a Research Software Engineer 3 working at IBM Ireland Research Lab. I am

part of the City Fabric team in the Smarter Cities Technology Centre.

I work on automatic techniques for integrating massive semi-structured data and

for recovering their semantics. I am interested in data fusion and management, and

in making sense of the large amount of data available on the Web. A major goal

in my work is to unlock unstructured and semi-structured data to enable querying,

extraction, and complex analytics tasks.

My interests include Software Engineering, Artificial Intelligence (automated plan-

ning), Intelligent Agents, Linked Data, Semantic Web and Privacy.

Prior to joining IBM Research I worked for Hewlett-Packard for 11 years. I have

a PhD in Computer Science from Politecnico di Torino (Italy) and Universite de

Technologie de Compiegne (France).

1 www.ibm.com/ie/research/

2 http://www.dublinked.ie/datastore/datastore.php

3 http://researcher.ibm.com/person/ie-marco.sbodio

Program

The seminar will be held at:

Dipartimento di Automatica e Informatica

Sala riunioni 3 (4 piano)

13 Novembre 2012, ore 11.45

Politecnico di Torino

Corso Duca degli Abruzzi, 24

Map:

http://www.polito.it/php/mappa/?sigla=STRUT,D1010&lang=it

Class schedule

• Teaching materials
• Exams and finals
• Go to Moodle
• Visit the forums