Abstract
Based on entropy optimal theory, a new model for early-warning of crisis is established. Firstly, minimum J-divergence entropy is applied to feature extraction. Then the calculating result is classified to judge state of enterprise with a new clustering algorithm, maximum entropy clustering algorithm, which is a development and extension of hard C-means. Finally, an example in early-warning of enterprise crisis is given to validate the model. The results show the feasibility and validity of the model. The research work supplies a new way for early-warning of enterprise crisis.
| Original language | English |
|---|---|
| Pages (from-to) | 113-117+121 |
| Journal | Kongzhi yu Juece/Control and Decision |
| Volume | 24 |
| Issue number | 1 |
| Publication status | Published - Jan 2009 |
| Externally published | Yes |
Keywords
- Early-warning of enterprise crisis
- Entropy clustering algorithm
- Feature extraction
- Minimum J-divergence entropy