Summary: | 碩士 === 雲林科技大學 === 資訊管理系碩士班 === 96 === Information plays an important role in enterprises, no matter in decision supporting and business strategies making all provide efficient supports to executive managers. However, information is getting more and more today, how to handle high dimensions data and high complexity data are the key issues of this research.
Multi-attribute data usually possesses high data dimension and high data complexity. In order to solve above problems, this research proposes a new information fusion method which is briefly described as follows: (1) Reduce data dimensions by PCA method. (2) Calculate integrated values by OWA operator. (3) Cluster data instance into specific group by FCM and train classification accuracy of training data. (4) Validate classification accuracy of testing data. In this research, there five datasets adopted to verify performances of proposed method, i.e. Iris, Lung cancer, WBC, SPA50A and SPA50B. The experiments results show that classification accuracies rates and of proposed method obviously surpass the listing methods and OWA operator can effectively offer corresponding weights to those important principal components. It is conducive to enhance the performance of proposed method.
|