Baseline Experiments in the Extraction of Polish Valence Frames

Initial experiments in learning valence (subcategorisation) frames of Polish verbs from a morphosyntactically annotated corpus are reported here. The learning algorithm consists of a linguistic module, responsible for very simple shallow parsing of the input text (nominal and prepositional phrase recognition) and for the identification of valence frame cues (hypotheses), and a statistical module which implements three well-known inferential statistics (likelihood ratio, t test, binomial miscue probability test). The results of the three statistics are evaluated and compared with a baseline approach of selecting frames on the basis of the relative frequencies of frame/verb co-occurrences. The results, while clearly reflecting the many deficiencies of the linguistic analysis and the inadequacy of the statistical measures employed here for a free word order language rich in ellipsis and morphosyntactic syncretisms, are nevertheless promising.