Data-oriented parsing (DOP) is one of the leading paradigms in statistical natural language processing. In this volume, a collection of computational linguists offer a state-of-the-art overview of DOP, suitable for students and researchers in natural language processing and speech recognition as well as for computational linguistics. This handbook begins with the theoretical background of DOP and introduces the algorithms used in DOP as well as in other probabilistic grammar models. After surveying extensions to the basic DOP model, the volume concludes with a close study of the applications that use DOP as a backbone: speech understanding, machine translation, and language learning.