Automating a Classification Task Based on an Augmented Thesaurus

Most classification tasks that have been tackled for automation are ones involving complex chains of causal reasoning. However, our daily lives are filled with simple classification tasks. We propose a lexicon-based solution to the problem of automating simple classifications, and hence provide an intelligent user interface for personal information management. Although a classical organization of a lexical database constitutes the backbone for our system as it provides a basis for semantic inherit ante, we found that its information content was far from being sufficient. In particular, we felt it necessary to augment the existing lexical database with various kinds of contextual information such as the user’s preference for classification criteria, personal information regarding the user, knowledge about the actions and objects associated with certain lexical items, and certain kinds of commonsense knowledge. We believe that our system provides a simple and easy-to-use interface to ...

• Type: Conference
• Year: 1997
• Where: IUI
• Authors: Eunok Paek, Hye-Jeong Jeon
• Keywords: Classification Tasks, IUI 1997, Personal Information Management, Simple Classifications, Software Engineering