PhD in Computer Science

Learning with the Web. Structuring data for easing machine understanding

Learning with the Web. Structuring data for easing machine understanding

Academic year 2013/2014

Sommario del corso

• Course objectives
• Notes
• Teaching tools

Course objectives

Abstract: The Web offers a vast amount of unstructured content from which to discover or better understand entities, while also serving as the largest knowledge graph on which to ground content. Moreover, it offers a broad range of relationships

that already exist among entities. Most of the knowledge available on the Web is present as natural language text enclosed in Web documents aimed at human consumption. A common approach for obtaining programmatic access to such a knowledge uses information extraction techniques; it reduces texts written in natural languages to machine readable structures, from which it is possible to retrieve entities and relations, for instance obtaining answers to database-style queries. 

In the first part of my talk, I will explain the theoretical foundations of the Named Entity Recognition and Disambiguation (NERD) initiative [1], detailing the challenges of the named entity recognition and named entity linking tasks. I will show the results NERD achieved in shared tasks such as ETAPE 2012, #MSM'13 and the results in benchmark such as CoNLL-2003 and TAC KBP 2011. In the second part of the talk, I will present the research we conducted for topic generation from annotated streams of heterogeneous data coming from social platforms such as Dailymotion, Youtube, Twitter, Facebook, and G+.

Class schedule

Note

The seminar will be held by Giuseppe Rizzo.

Short Bio: Giuseppe Rizzo is an Associate Research Fellow at the Multimedia Communications Department of EURECOM since July 2013. Currently Giuseppe is involved in the LinkedTV project and in the recent past has collaborated in the OpenSem and EventMedia projects. His main research interests include Information Extraction, Linked Data and Web Science. He is a devoted Web evangelist, fascinated by smart web applications and knowledge extraction techniques; he likes to consider the Web as a mine of real world entities. He obtained his Master in Computer Science at the Politecnico di Torino, Turin, Italy after defending his thesis with the subject of Semantic Classification and Clustering of Emails. Mixing Web techniques and intelligent agents became a love and he decided to pursue these studies in a PhD. He benefited from a PhD fellowship from Politecnico di Torino at the Department of Control and Computer Engineering, supervised by the legendary prof. Raffaele Meo. During this period, he was visiting PhD student at EURECOM, Sophia Antipolis, France, supervised by prof. Raphael Troncy.

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