Automatic Fiscal Stabiliser: Make it happen! EPC Discussion Paper, 2 May 2016

The eurozone crisis triggered a whole new series of innovations in EU economic governance in order to make the Union more resilient for the next economic downswing. But one of the more persistent issues are the socio-economic divergences between member states, identified by the Five Presidents’ Report as a major problem in the functioning of the Economic and Monetary Union (EMU). Debates took place in recent years about automatic stabilisers, and more specifically about the possibility of introducing an unemployment insurance within the EMU. While the need for some form of fiscal risk-sharing has become a dominant view in expert circles, there has been much less progress among the main political parties and stakeholders. In this study, Regula Hess and László Andor analyse the political feasibility of the adoption of an automatic fiscal stabiliser (AFS) for the eurozone by evaluating actors’ positions towards three distinctive proposals: 1) cyclical shock insurance, 2) reinsurance, 3) a European basic unemployment insurance; they included an empirical case study of France and Germany as the most relevant players within the intergovernmental bargaining constellation. Although the authors realise the current political context makes the adoption of an AFS improbable, Hess and Andor encourage stakeholders to further pursue the discussion, as windows of opportunities can open at any time, and even give some suggestions on what the parameters of the most feasible proposal might be.

JAABA: interactive machine learning for automatic annotation of animal behavior

We present a machine learning-based system for automatically computing interpretable, quantitative measures of animal behavior. Through our interactive system, users encode their intuition about behavior by annotating a small set of video frames. These manual labels are converted into classifiers that can automatically annotate behaviors in screen-scale data sets. Our general-purpose system can create a variety of accurate individual and social behavior classifiers for different organisms, including mice and adult and larval Drosophila.

Web mining for translators: automatic construction of comparable, genre-driven corpora

The aim of this paper is to evaluate the efficacy of the application WebBootCaT to create specialised corpora automatically, investigating the translation of articles of association from Italian into English. The first section reflects on the relevant literature and proposes the utility of corpora for translators. The second section discusses the methodology employed, and the third section analyses the results obtained and comments on how language professionals could possibly exploit the application to its full. The fourth section provides a few concrete usage examples of the thus built corpora, to then conclude that WebBootCaT is a genuinely powerful tool that could be implemented by professional translators in order to save time and improve their translations in the long term.