TY - JOUR
T1 - Ovarian cancer classification using K-Nearest Neighbor and Support Vector Machine
AU - Wibowo, V. V.P.
AU - Rustam, Z.
AU - Hartini, S.
AU - Maulidina, F.
AU - Wirasati, I.
AU - Sadewo, W.
N1 - Funding Information:
This research is fully supported financially by The Indonesian Ministry of Research, Technology, with a KEMENRISTEK/BRIM PDUPT 2021 research grant scheme.
Publisher Copyright:
© Published under licence by IOP Publishing Ltd.
PY - 2021/3/29
Y1 - 2021/3/29
N2 - Ovarian cancer is one of the common malignancies in women and a known cause of death. This condition occurs when a tumor appears from the growth of abnormal cells in the ovary. It causes about 140.000 deaths out of 225.000 cases annually. Most women with ovarian cancer do not have distinctive signs and symptoms even at the late stage. Therefore, diagnosis at an early stage is necessary because it has a significant impact on the survival rate. Machine learning with various methods can be used in the medical field to classify diseases. Among the many methods, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) were used and analyzed in this study to classify ovarian cancer. The data used were from Al Islam Bandung Hospital consisting of 203 instances with 130 labeled ovarian cancer and 73 as non-ovarian. The results showed that the KNN produced higher results than SVM with 90.47% of accuracy and 94.11% of F1-score, while SVM produced accuracy and F1-score values of 90.47% and 92.30% respectively.
AB - Ovarian cancer is one of the common malignancies in women and a known cause of death. This condition occurs when a tumor appears from the growth of abnormal cells in the ovary. It causes about 140.000 deaths out of 225.000 cases annually. Most women with ovarian cancer do not have distinctive signs and symptoms even at the late stage. Therefore, diagnosis at an early stage is necessary because it has a significant impact on the survival rate. Machine learning with various methods can be used in the medical field to classify diseases. Among the many methods, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) were used and analyzed in this study to classify ovarian cancer. The data used were from Al Islam Bandung Hospital consisting of 203 instances with 130 labeled ovarian cancer and 73 as non-ovarian. The results showed that the KNN produced higher results than SVM with 90.47% of accuracy and 94.11% of F1-score, while SVM produced accuracy and F1-score values of 90.47% and 92.30% respectively.
UR - http://www.scopus.com/inward/record.url?scp=85103891589&partnerID=8YFLogxK
U2 - 10.1088/1742-6596/1821/1/012007
DO - 10.1088/1742-6596/1821/1/012007
M3 - Conference article
AN - SCOPUS:85103891589
SN - 1742-6588
VL - 1821
JO - Journal of Physics: Conference Series
JF - Journal of Physics: Conference Series
IS - 1
M1 - 012007
T2 - 6th International Conference on Mathematics: Pure, Applied and Computation, ICOMPAC 2020
Y2 - 24 October 2020
ER -
