Guiding Example Acquisition by Generating Scenarios
Abstract
Example acquisition often represents an important bottleneck when using learning techniques. This paper presents a method for guiding the acquisition of new learning examples, in the context of empirical learning. The output knowledge of the learning algorithm and the examples that have been used are analyzed by a module, called an example postprocessor, which generates specifications (called scenarios) for new examples considered strategic for improving the prediction accuracy of the output knowledge. These specifications are proposed to the expert, who provides, if possible, corresponding new examples to be added to the learning set.
Cite
Text
Niquil. "Guiding Example Acquisition by Generating Scenarios." International Conference on Machine Learning, 1992. doi:10.1016/B978-1-55860-247-2.50049-8Markdown
[Niquil. "Guiding Example Acquisition by Generating Scenarios." International Conference on Machine Learning, 1992.](https://mlanthology.org/icml/1992/niquil1992icml-guiding/) doi:10.1016/B978-1-55860-247-2.50049-8BibTeX
@inproceedings{niquil1992icml-guiding,
title = {{Guiding Example Acquisition by Generating Scenarios}},
author = {Niquil, Yves},
booktitle = {International Conference on Machine Learning},
year = {1992},
pages = {348-354},
doi = {10.1016/B978-1-55860-247-2.50049-8},
url = {https://mlanthology.org/icml/1992/niquil1992icml-guiding/}
}