Decision trees: a contribution to automatic interpretation of GUHA results
✍ Scribed by Zdeněk Renc; Lenka Setíkovská
- Book ID
- 104139948
- Publisher
- Elsevier Science
- Year
- 1985
- Weight
- 846 KB
- Volume
- 22
- Category
- Article
- ISSN
- 0020-7373
No coin nor oath required. For personal study only.
✦ Synopsis
The paper is concerned with a method of automatic interpretation of results obtained by the GUHA method with implicational quantifiers. These results are represented by a set of sentences (elementary implications) verified on a model (experimental data). On the basis of these sentences, the method described forms the so-called decision trees, with the aid of which each object of the model in question can be assigned one of the mutually exclusive properties from a set chosen in advance. In the introduction, the necessary information on the GUHA method is briefly outlined. In the following sections, the notion of a decision tree and its labelling are introduced and the means suitable for its formation are examined. In the concluding part, algorithms of generation of the (best) decision trees are described and their basic properties are inferred.