TY - CHAP
T1 - Kernel-based fuzzy clustering for sinusitis dataset
AU - Rustam, Zuherman
AU - Putri, Nadisa Karina
AU - Pandelaki, Jacub
AU - Nugroho, Widyo Ari
AU - Utami, Dea Aulia
AU - Hartini, Sri
N1 - Funding Information:
Acknowledgement. We wish to express our gratitude to the University of Indonesia and PIT.9 2019 research grant scheme (ID number NKB-0039/UN2.R3.1/HKP.05.00/2019) for facilitating this study. This work was also supported by the Department Radiology of Dr. Cipto Man-gunkusumo’s Hospital, and we are so grateful to them. We also thank all the reviewers for the improvements made.
Publisher Copyright:
© Springer Nature Switzerland AG 2020.
PY - 2020/1/1
Y1 - 2020/1/1
N2 - Sinusitis is a condition resulting from inflammation of sinus walls. In handling the disease, machine learning method is often used to find more precise and accurate treatment plan for patients. For instance, fuzzy clustering is widely used for pattern recognition and data mining. Due to uncertainty and ambiguity, this method is used to overcome the non-linearity of medical dataset. In this study, fuzzy clustering was provided with kernel methods. We used some Kernel methods such as, Kernelized Fuzzy c-Means (KFCM), Kernelized Possibilistic c-Means (KPCM), Kernelized Fuzzy Possibilistic c-Means (KFPCM), and Kernelized Possibilistic Fuzzy c-Means (KPFCM) for clustering sinusitis dataset. The dataset was retrieved from Cipto Mangunkusumo Hospital Jakarta, Indonesia, which contains 4 features and 200 instances of this condition. These level of accuracy and model performance are used to compare these approaches. The result showed that KFCM has the highest accuracy for categorizing sinusitis dataset with accuracy of 96.97% and running time of 0.01 s.
AB - Sinusitis is a condition resulting from inflammation of sinus walls. In handling the disease, machine learning method is often used to find more precise and accurate treatment plan for patients. For instance, fuzzy clustering is widely used for pattern recognition and data mining. Due to uncertainty and ambiguity, this method is used to overcome the non-linearity of medical dataset. In this study, fuzzy clustering was provided with kernel methods. We used some Kernel methods such as, Kernelized Fuzzy c-Means (KFCM), Kernelized Possibilistic c-Means (KPCM), Kernelized Fuzzy Possibilistic c-Means (KFPCM), and Kernelized Possibilistic Fuzzy c-Means (KPFCM) for clustering sinusitis dataset. The dataset was retrieved from Cipto Mangunkusumo Hospital Jakarta, Indonesia, which contains 4 features and 200 instances of this condition. These level of accuracy and model performance are used to compare these approaches. The result showed that KFCM has the highest accuracy for categorizing sinusitis dataset with accuracy of 96.97% and running time of 0.01 s.
KW - Kernelized Fuzzy c-Means (KFCM)
KW - Kernelized Fuzzy Possibilistic c-Means (KFPCM)
KW - Kernelized Possibilistic c-Means (KPCM)
KW - Kernelized Possibilistic Fuzzy c-Means (KPFCM)
KW - Sinusitis
UR - http://www.scopus.com/inward/record.url?scp=85081580725&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-43002-3_5
DO - 10.1007/978-3-030-43002-3_5
M3 - Chapter
AN - SCOPUS:85081580725
T3 - Lecture Notes in Networks and Systems
SP - 44
EP - 54
BT - Lecture Notes in Networks and Systems
PB - Springer
ER -