Picture Fuzzy Set – Based Approach for Classification
21st International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2026, Rome, İtalya, 15 - 19 Haziran 2026, cilt.3020 CCIS, ss.188-201, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Cilt numarası: 3020 CCIS
- Doi Numarası: 10.1007/978-3-032-28997-1_14
- Basıldığı Şehir: Rome
- Basıldığı Ülke: İtalya
- Sayfa Sayıları: ss.188-201
- Anahtar Kelimeler: Classification, k-Nearest Neighbor, Picture Fuzzy kNN, Picture Fuzzy Sets
- Kocaeli Üniversitesi Adresli: Evet
Özet
Picture Fuzzy Sets offer a flexible representation for handling uncertainty in data. In this study, we employ a relative frequency – based Picture Fuzzy Set modeling strategy within a k-Nearest Neighbor framework and propose a Picture Fuzzy kNN classifier using the Dα operator. The proposed framework consists of a formal mathematical definition of the relative frequency – based Picture Fuzzy Set representation followed by the algorithmic formulation of the Picture Fuzzy kNN classifier. The proposed method is evaluated on the User Knowledge Modeling dataset and compared with classical kNN using different distance measures and distance-weighting schemes. Experimental results demonstrate that our approach significantly improves classification performance.