We present the work of the Democritus University of Thrace (DUTH) team in TREC’s 2016 Contextual Suggestion Track. The goal of the Contextual Suggestion Track is to build a system capable of proposing venues which a user might be interested to visit, using any available contextual and personal information. First, we enrich the TREC-provided dataset by collecting more information on venues from ...

The difficulties of current classifications of depression can not be considered independent of the general complications of psy chiatric classification. A tedious period of heated debate between the proponent of unitary vs pluralist views, and between the advocates of the categorical vs dimensional views didn't prove fruitful. Current classifications share two main characteristics I.e, they are...

We apply phrase-based and neural models to a core task in interactive machine translation: suggesting how to complete a partial translation. For the phrase-based system, we demonstrate improvements in suggestion quality using novel objective functions, learning techniques, and inference algorithms tailored to this task. Our contributions include new tunable metrics, an improved beam search stra...

Which comes first—the lyrics or the music? Here we consider the lyrics-first approach to songwriting and seek to augment the process by developing a creativity-support tool which uses lyrics as a creative constraint for rhythm generation. We describe a novel algorithm for suggesting possible melodic rhythms to complement the accent structure of a given set of lyrics. The algorithm is composed o...

The field of the proposed project is opinion dynamics which involves the mathematical modeling of how groups of agents (people) influence one anothers’ opinions. This is very much a cross-disciplinary field, involving work by physicists, computer scientists, robotics engineers and, of course, social scientists. Mathematically, it is still very young, with a shortage of rigorous theorems even at...