ncmarie - PhDc

Welcome :)

My Phd thesis, started on september 2010, is focused on real-time and linked-data based exploratory search.

First popular search engines were released in the late nineties. They offered at that time a new and efficient paradigm to access the growing content of the web. Today, search is still the main way to access information on the web.

Actual search engines are very efficient for “lookup” tasks i.e. tasks when the user knows precisely what he wants and how to ask it. In this case keywords used are well-defined; the result space is limited and easy to navigate. However search engines are less efficient for exploratory search tasks such as learning or investigation. In this case the search objective is fuzzier. User manipulates iteratively an evolving set of keywords and has to synthesize an important amount of information coming from an open result space.

So-called exploratory search systems are built specifically to help user in these high cognitive complexity search activities. Linked data are particularly interesting for exploratory search systems. Traditional keyword-based search is based on link popularity. It does not take into account the meaning of the content. Thus it misses numerous aspects that can be interesting when providing results for exploratory search purpose.

In this context I am especially focused in real-time linked data processing and linked-data based navigation strategies for exploratory search purpose. An operational prototype of Discovery Hub, a DBpedia-based discovery engine, is currently developed with Damien Legrand and is accessible in beta mode. Don’t hesitate to request an access!

My PhD Thesis is Co-supervised by Myriam Ribière (Alcatel-Lucent Bell Labs) and Fabien Gandon (INRIA Sophia-Antipolis).

A bientôt ! Nicolas

ncmarie3@gmail.com

nicolas.marie@alcatel-lucent.com

nicolas.marie@inria.fr