"The first million is hardest to get": building a large tagged corpus as automatically as possible
COLING '90 Proceedings of the 13th conference on Computational linguistics - Volume 3
| Hi-index | 0.00 |
PC Beta is a PC oriented tool for corpus work in this term's broadest possible sense. With PC Beta one can prepare text for corpus work, e.g. standardize texts in different ways (very important when texts from different sources together will constitute a corpus), one can process texts, and one can analyze texts. Making ordinary concordances and similar things with PC Beta is, of course, very simple, and, in fact, PC Beta gives "concordance making" a new dimension. One can perform morphological analyses, one can use PC Beta as a "tagger", i.e. provide the words with different kinds of tags. In all, PC Beta is a versatile program, and it is in many cases the only program needed (together with function belonging to the MS/PC-DOS operative system) for pursuing a complete corpus project.The program's main distinctive feature is simplicity: it is rule controlled, and the rules adhere to a format that any linguist can learn to understand very quickly. But beware, in spite of its innocent appearence the program is a little tiger.