A transparent classification model using a hybrid soft computing method
Due to the inherent complexity of many real-world problems, classification models have become an important tool for solving pattern recognition tasks in many disciplines such as medicine, finance and management. Accuracy and transparency are two important criteria that should be satisfied by any cla...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
2009
|
| Subjects: |
| Summary: | Due to the inherent complexity of many real-world problems, classification models have become an important tool for solving pattern recognition tasks in many disciplines such as medicine, finance and management. Accuracy and transparency are two important criteria that should be satisfied by any classification model. In this paper, a transparent and relatively accurate classifier is developed using a hybrid soft computing technique. The initial fuzzy model is first generated using a clustering method and the transparency and accuracy of the model are then simultaneously optimized using a multi-objective evolutionary technique. The proposed model is tested on two real problems; the first one is related to credit scoring problem while the other is on medical diagnosis. All the data sets used in this study are publicly available at UCI repository of machine learning database. |
|---|