Abstract
A method for conceptual clustering of a collection of texts represented with conceptual graphs is presented. It uses the incremental strategy to construct the cluster hierarchy and incorporates some characteristics attractive for text mining processes. For instance, it considers the structural information of the graphs, uses domain knowledge to detect the cluster with generalized descriptions, and uses a user-defined similarity measure between the graphs.
Original language | English |
---|---|
Pages (from-to) | 898-903 |
Number of pages | 6 |
Journal | Proceedings of the IEEE International Conference on Systems, Man and Cybernetics |
Volume | 2 |
DOIs | |
State | Published - 2001 |
Externally published | Yes |
Keywords
- Conceptual clustering
- Domain knowledge
- Similarity measure
- Summarization
- Users interests profile