{"type":"ScholarlyArticle","title":"Reduction of Enumeration in Grammar Acquisition","authors":[{"name":"Stefano Crespi-Reghizzi","slug":"stefano-crespi-reghizzi","given":"Stefano","family":"Crespi-Reghizzi","affiliation":""}],"abstract":"The learner of a language is able to acquire a grammar on the basis of finite linguistic data, within stringent constraints of time and memory. The learner selects one out of the infinitely many grammars which are compatible with the data. The require ment of feasibility implies a reduction in the number of hypotheses which are compatible with the data, so that the learner can quickly select the best grammar (based on some evaluation measure) out of the hypotheses. Our grammar acquisition model applies to a non trivial subset of context-free grammars, for which the enumeration of potential hypotheses is greatly reduced to a point where a computer program for grammar acquisition becomes feasible. The availability of structure descriptions in the data not only contributes to a reduction of enumeration, but also results in a grammar which is \"strongly adequate\". As the sample is enlarged and tends to the source language, the grammar acquisition identifies the source grammar in the limit.","doi":null,"url":"https://ijcai.org/proceedings/1971/57","pdf":"https://ijcai.org/proceedings/1971/0057.pdf","markdown":"https://ijcai.org/proceedings/1971/57.md","bibtex":"https://ijcai.org/proceedings/1971/bibtex/57","volume":{"slug":"1971","title":"Proceedings of the Second International Joint Conference on Artificial Intelligence","short_title":"IJCAI 1971","url":"https://ijcai.org/proceedings/1971/","year":1971,"location":"London, UK","dates":"","editors":"D.C. Cooper","isbn":"","publisher":"International Joint Conferences on Artificial Intelligence","doi_prefix":""},"section":null,"track":"","number":57,"pages":{"first":546,"last":551},"keywords":[],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}