Conceptual clustering of structured objects: a goal-oriented approach
Artificial Intelligence
| Hi-index | 0.00 |
The ability of a system to classify objects (physical or abstract) accurately and approximately is essential for performing such Artificial Intelligence tasks as Natural Language Processing, Inductive Reasoning, and building Decision Support Systems and Expert Systems. This paper introduces an approach to data analysis which will allow us to determine whether a set of objects is precisely describable, approximately describable, or non-describable. This approach also provides a conceptual description of the set whenever it is precisely or approximately describable. Finally, this data analysis approach is used as a stepping stone to study foundations of knowledge representation.