PhD Alumni

Eynard Davide

Present position: Post-doc at Politecnico di Milano and UniversitÓ della Svizzera Italiana

Thesis title:  A Virtuous Cycle of Semantics and Participation
Advisor:  Marco Colombetti
Research area:  Artificial Intelligence, Robotics and Computer Vision
Thesis abstract:  
Participative systems are a particular class of social systems, in which people can interact, share information, or both. What characterizes them is not just a passive presence of users, but the fact that they actually 'take part' into some activity, providing new value to the community thanks to their explicit or implicit contributes. These systems have gained a great consensus both inside corporations and on the World wide web, and have become particularly interesting not only as a possibly successful business, but also as an object of academic research. In fact, as they deal at the same time with technologies and people, they present a plethora of interesting research problems that can be studied from the point of views of many different sciences. The main objective of our research project is to study participative systems and the possibility to enhance them through semantics. We believe that the advantages of a contamination between these systems and Semantic Web technologies could be twofold: on the one hand, the huge quantity of information created by the participation of many users could be better managed and searched thanks to added semantics; on the other hand, Semantic Web community could exploit spontaneous participation to increase the amount of knowledge described through formal representations, making it available to many other applications. Following the double nature of this research (the social and the technological one) our research question is split in two: given a community, a task and a context, we want to understand (i) what is the
most suitable tool to foster user participation and produce useful information, and (ii) how we could employ semantics to make this better. Our effort in this project has been to find both the technical insights and the design paradigms to choose the best participative system for a particular community and its activity, or evaluate the health of an existing one, and address part of its limitations using Semantic Web technologies (after recognizing if it is possible in the first place). As a result, we provide a set of suggestions and patterns which can be used to bootstrap or fine tune a social system. To this knowledge we add the results of our experimentations on semantic participative systems. We describe how two of the main tools to manage semantics, that is ontologies and reasoners, could be exploited to provide enhanced user interfaces for information visualization and consumption; we show how intrinsic limits in classification which depend on the ambiguity of English words could be addressed by semantic disambiguation techniques; finally, we show the advantages of sharing information as linked data in email servers and browser history, and how to use already available open information to make the browsing experience better and act as an incentive for user participation in collective activities.