Shallow Dialogue Processing Using Machine Learning Algorithms (or not)

This paper presents a shallow dialogue analysis model, aimed at human-human dialogues in the context of staff or business meetings. Four components of the model are defined, and several machine learning techniques are used to extract features from dialogue transcripts: maximum entropy classifiers for dialogue acts, latent semantic analysis for topic segmentation, or decision tree classifiers for discourse markers. A rule-based approach is proposed for solving cross-modal references to meeting documents. The methods are trained and evaluated thanks to a common data set and annotation format. The integration of the components into an automated shallow dialogue parser opens the way to multimodal meeting processing and retrieval applications.

The astrological works of Abraham Ibn Ezra : a literary and linguistic study with special reference to the Old French translation of Hagin

by Raphael Levy

The Athens Dialogues : An International Conference on Culture and Civilization

Fil: Fernández Deagustini, María del Pilar. Universidad Nacional de La Plata. Facultad de Humanidades y Ciencias de la Educación. Instituto de Investigaciones en Humanidades y Ciencias Sociales (UNLP-CONICET); Argentina.

The Athens Dialogues : An International Conference on Culture and Civilization

Fil: Fernández Deagustini, María del Pilar. Universidad Nacional de La Plata. Facultad de Humanidades y Ciencias de la Educación. Instituto de Investigaciones en Humanidades y Ciencias Sociales (UNLP-CONICET); Argentina.

The Athens Dialogues : An International Conference on Culture and Civilization

Fil: Fernández Deagustini, María del Pilar. Universidad Nacional de La Plata. Facultad de Humanidades y Ciencias de la Educación. Instituto de Investigaciones en Humanidades y Ciencias Sociales (UNLP-CONICET); Argentina.

Mean annual surface displacement of the Laurichard rock glacier, French Alps

The Laurichard active rock glacier is the permafrost-related landform with the longest record of monitoring in France, including an annual geodetic survey, repeated geoelectrical campaigns from 1979 onwards and continuous recording of ground temperature since 2003. These data were used to examine changes in creep rates and internal structure from 1986 to 2006. The control that climatic variables exert on rock glacier kinematics was investigated over three time scales. Between the 1980s and the early 2000s, the main observed changes were a general increase in surface velocity and a decrease in internal resistivity. At a multi-year scale, the high correlation between surface movement and snow thickness in the preceding December appears to confirm the importance of snow cover conditions in early winter through their influence on the ground thermal regime. A comparison of surface velocities, regional climatic datasets and ground sub-surface temperatures over six years suggests a strong relation between rock glacier deformation and ground temperature, as well as a role for liquid water due to melt of thick snow cover. Finally, unusual surface lowering that accompanied peak velocities in 2004 may be due to a general thaw of the top of the permafrost, probably caused both by two successive snowy winters and by high energy inputs during the warm summer of 2003.