Stereotype Extraction with Default Clustering
Abstract
The concept of stereotype seems to be really adapted when wishing to extract meaningful descriptions from data, especially when there is a high rate of missing values. This paper proposes a logical framework called default clustering based on default reasoning and local search techniques. The first experiment deals with the rediscovering of initial descriptions from artificial data sets, the second one extracts stereotypes of politicians in a real case generated from newspaper articles. It is shown that default clustering is more adapted in this context than the three classical clusterers considered.
Cite
Text
Velcin and Ganascia. "Stereotype Extraction with Default Clustering." International Joint Conference on Artificial Intelligence, 2005.Markdown
[Velcin and Ganascia. "Stereotype Extraction with Default Clustering." International Joint Conference on Artificial Intelligence, 2005.](https://mlanthology.org/ijcai/2005/velcin2005ijcai-stereotype/)BibTeX
@inproceedings{velcin2005ijcai-stereotype,
title = {{Stereotype Extraction with Default Clustering}},
author = {Velcin, Julien and Ganascia, Jean-Gabriel},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2005},
pages = {883-888},
url = {https://mlanthology.org/ijcai/2005/velcin2005ijcai-stereotype/}
}