Publication View

Learning Two-Tiered Descriptions of Flexible Concepts: The Poseidon System (1992)

Abstract
Abstract. This paper describes a method for learning flexible concepts. by which are meant concepts that lack precise definition and are contextqlependent. To describe such concepts, the method employs a two-tiered represen-tation. in which the first tier captures explicitly basic concept properties, and the second tier characterizes allowable concept's modifications and context dependency. In e proposed method. the first tier, called Base Concept Representation (BCR), is created in two phases. In phase 1, the AQ-15 rule learning program is applied to induce a complete and consistent concept description from supplied examples. In phase 2, this description is optimized according to a domain-dependent quality criterion. The second tier, called the inferential concept interpretation dCI). consists of a procedure for flexible matching, and a set of inference rules. The proposed method has been implemented in the POSEIDON system. and experimentally tested on two real-world problems: [earning the concept of an acceptable umon contract, and learning voting patterns of Republicans and Democrats in the U.S. Congress. For comparison, a few other learning methods were also applied to the same problems. These methods included simple variants of exemplar-based learning, and an ID-3-tyl: decision tree learning, implemented m the ASSISTANT program. In the exl:riments, POSEIDON generated concept descriptions that were both, more accurate and also substantially simpler than those produced by the other methods.

Publication details
Download http://citeseerx.ist.psu.edu/viewdoc/summary?doi=?doi=10.1.1.18.9808
Source http://www.mli.gmu.edu/papers/90-95/92-01.pdf
Contributors CiteSeerX
Repository CiteSeerX - Scientific Literature Digital Library and Search Engine (United States)
Keywords learning imprecise concepts, inductive learning, learning flexible concepts, twotiered concept
Type text
Language English
Relation 10.1.1.18.4267, 10.1.1.127.1694, 10.1.1.33.4894, 10.1.1.64.909, 10.1.1.19.4185, 10.1.1.50.9283, 10.1.1.18.9502, 10.1.1.52.5864, 10.1.1.51.3784, 10.1.1.70.8341, 10.1.1.32.5377, 10.1.1.70.7172, 10.1.1.14.6525, 10.1.1.23.7904, 10.1.1.20.9288, 10.1.1.102.8729, 10.1.1.26.7446, 10.1.1.5.740, 10.1.1.19.6848, 10.1.1.72.2479, 10.1.1.14.4729, 10.1.1.33.6519, 10.1.1.100.1930, 10.1.1.14.3371, 10.1.1.18.9949, 10.1.1.19.1766, 10.1.1.19.255, 10.1.1.19.2830, 10.1.1.19.3907, 10.1.1.19.5014, 10.1.1.19.671, 10.1.1.19.6811, 10.1.1.26.1364, 10.1.1.30.8828, 10.1.1.37.2436, 10.1.1.38.4827, 10.1.1.47.6350, 10.1.1.51.6810, 10.1.1.53.1471, 10.1.1.62.9351, 10.1.1.70.980, 10.1.1.72.5964, 10.1.1.72.9991, 10.1.1.92.239