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.

Theory after Dialogue: Christian-Muslim Engagement Today and Tomorrow

The phenomenon of Christian–Muslim dialogue has had a very chequered history. At varying times, three broad modes of engagement can be said to have operated: antipathy, affinity and inquiry, and these three modes can be found still in today's world. In some places, hostility and antipathy abound. In others, voices and actions express cordial friendship, détente and affinity. In this latter climate, the prospect of engagement in mutual inquiry and cooperative ventures is not only theoretically possible, but actively pursued, and in the first decade of the twenty-first century, a number of notable initiatives in the arena of mutual inquiry have taken place. This article addresses aspects of the context and development of Christian–Muslim dialogue as a modern phenomenon, and then turns to a review of three twenty-first century developments – the Building Bridges seminar series; the Stuttgart-based Christian–Muslim Theological Forum and the “Common Word” letter. It also reflects on the models and theology of dialogue, including not only theology for dialogue, but also theology in and – importantly – after dialogue.