Section: New Results

Natural Language Processing

In [6] (presented by David Chatel at the ECML–PKDD and CAp'2014 conferences) we propose a new algorithm for semi-supervised spectral clustering and apply it to the task of noun phrase coreference resolution. The main insight is in the inclusion of pairwise constraints into spectral clustering: our algorithm learns a new representation space for the data together with a distance in this new space. The representation space is obtained through a constraint-driven linear transformation of a spectral embedding of the data, and constraints are expressed with a Gaussian function that locally reweights the similarities in the projected space. A global, non-convex optimization objective is then derived and the model is learned via gradient descent techniques. Our algorithm is evaluated on the CoNLL-2012 coreference resolution shared task dataset, and shows some encouraging results.

In [2] and [1] , we develop a new approach for the automatic identification of so-called implicit discourse relations. Specifically, our system combines hand labeled examples and automatically annotated examples based on explicit relations using several simple methods inspired by work in domain adaptation. Our system is evaluated empirically on the Annodis corpus, a French corpus annotated with discourse structures. Our system yields significant performance gains compared to only using hand-labeled data or using only automatically annotated data.