{"type":"ScholarlyArticle","title":"The Utility of Feature Construcuon for Back-Propagation","authors":[{"name":"Harish Ragavan","slug":"harish-ragavan","given":"Harish","family":"Ragavan","affiliation":""},{"name":"Selwyn Piramuthu","slug":"selwyn-piramuthu","given":"Selwyn","family":"Piramuthu","affiliation":""}],"abstract":"The ease oflearning concepts from examples in empirical machine learning depends on the attributes used for describing the training data. We show that decision-tree based feature construction can be used to improve the performance of back-propagation (BP), an artificial neural network algorithm, both in terms of the convergence speed and the number of epochs taken by the BP algorithm to converge. We use disjunctive concepts to illustrate feature construction, and describe a measure of feature quality and concept difficulty. We show that a reduction in the difficulty ofthe concepts to be learned by constructing better representations increases the performance of BP considerably.","doi":null,"url":"https://ijcai.org/proceedings/1991/124","pdf":"https://ijcai.org/proceedings/1991/0124.pdf","markdown":"https://ijcai.org/proceedings/1991/124.md","bibtex":"https://ijcai.org/proceedings/1991/bibtex/124","volume":{"slug":"1991","title":"Proceedings of the Twelfth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 1991","url":"https://ijcai.org/proceedings/1991/","year":1991,"location":"Sydney, New South Wales, Australia","dates":"24-30 August 1991","editors":"John Mylopoulos and Ray Reiter","isbn":"","publisher":"International Joint Conferences on Artificial Intelligence","doi_prefix":""},"section":"Learning Be Knowledge Acquisition","track":"Learning Be Knowledge Acquisition","number":124,"pages":{"first":844,"last":849},"keywords":[],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}