Memory-Based Text Chunking

The abundance of data often hinders computational efficiency in information retrie val, information extraction and parsing. Several data reduction techniques have been proposed t o v rcome this problem. A recently blooming approach to data reduction in computational linguist ics chunking. In (Daelemans, Buchholz, and Veenstra, 1999), we describe experiments on NP and V P (noun and verb phrase) chunking using Memory-Based Learning ( MBL ) carried out in the context of subject/object identification. In this paper we will show that NP, VP and PP ( prepositional phrase) chunking is possible with a precision and recall around 94-95%.