Unsupervised Lexical Learning with Categorical Grammars Using the LLL Corpus

Stephen Watkinson, Suresh Manandhar

Research output: Contribution to conferencePaperpeer-review


In this paper we report on an unsupervised approach to learning Categorial Grammar (CG) lexicons. The learner is provided with a set of possible lexical CG categories, the forward and backward application rules of CG and unmarked positive only corpora. Using the categories and rules, the sentences from the corpus are probabilistically parsed. The parses of this example and the set of parses of earlier examples in the corpus are used to build a lexicon and annotate the corpus. We report the results from experiments on two generated corpora and also on the more complicated LLL corpus, that contains examples from subsets of English syntax. These show that the system is able to generate reasonable lexicons and provide accurately parsed corpora in the process. We also discuss ways in which the approach can be scaled up to deal with larger and more diverse corpora.
Original languageUndefined/Unknown
Publication statusPublished - 1999

Cite this