NOVEL INCREMENTAL ALGORITHMS FOR ATTRIBUTE REDUCTION FROM DYNAMIC DECISION TABLES WHEN ADDING OBJECT S

• Tác giả: Thi Phuong Ho, Duc Phuong Truong
• Ngôn ngữ: eng
• Thông tin xuất bản: Tạp chí Khoa học - Trường Đại học Thủ đô Hà Nội: Khoa học Tự nhiên và Công nghệ, 2023
• Mô tả vật lý: tr.45665
• Bộ sưu tập: Metadata
• ID: 341097

Tóm tắt

In recent years, attribute reduction methods following fuzzy rough set approachhave attracted the attention of researchers because they improve the accuracy of theclassification model. However, most of the proposed methods are performed on the unchangeddecision table. In this paper, we build an incremental algorithm to find the approximate reductaccording to the combined filter-wrapper approach. Experimental results on a number ofsample datasets show that the proposed incremental algorithm is more efficient than someother incremental algorithms following the filter approach in terms of the number of reductiveset attributes and classification accuracy.