Domain knowledge to support the discovery process: user preferences

Bing Liu, Wynne HsuOxford 2002 · OpenAlex

Abstract

The goal of data mining is to discover useful knowledge (or rules) for the user. Past research has produced many efficient techniques for rule discovery from databases. However, these techniques often generate too many rules, which makes it very difficult for the user to analyze them in order to find those truly interesting/useful rules. In this article, we first discuss some issues involved in assisting the user to analyze the discovered rules. We then review a number of existing techniques that employ the user's preferences and knowledge about the domain to identify those potentially interesting/useful rules for the user.
BibTeX
@inproceedings{liu2002domain,
  author = {Bing Liu and Wynne Hsu},
  title = {{Domain knowledge to support the discovery process: user preferences}},
  booktitle = {Oxford},
  year = {2002},
}

Loading related content...

Figure