Loading...
Please wait, while we are loading the content...
Similar Documents
A Practical Approach for Knowledge-Driven Constructive Induction
| Content Provider | Semantic Scholar |
|---|---|
| Author | Lee, Huei Diana Iltc, Maria Carolina Monard Augusto, José Baranauskas |
| Copyright Year | 2004 |
| Abstract | Learning problems can be difficult for many reasons, one of them is inadequate representation space or description language. Features can be considered as a representational language; when this language contains more features than necessary, subset selection helps simplify the language. On the other hand, when this language is not sufficient to describe the problem, Feature Construction helps enrich the language. Feature Construction, also known as Constructive Induction, aims to discover missing information about the relationships between features and augments the space of features by inferring additional features. Thus, feature selection reduces the feature space while Feature Construction expands the feature space. In both cases, the main idea is to improve the representation space before searching for concept description, in order to improve the overall prediction accuracy of the generated concept description. This work is concerned with knowledge-driven Constructive Induction, which uses domain knowledge provided by the expert to search for a better representational space. The objective of this work is to propose an approach for practical Feature Construction when this is done with the aid of the user or the expert. We describe a series of experiments performed on four real world datasets using inducers C4.5rules and CN2. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://dfm.ffclrp.usp.br/~augusto/publications/2000-asai.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
