Oversampling Menggunakan Pendekatan Latin Hypercube Sampling Dan K-Nearest Neighbors Untuk Meningkatkan Kinerja Klasifikasi
Kata Kunci:
Imbalanced Class, kinerja klasifikasi, k-Nearest Neighbor, Latih Hypercube Sampling, OversamplingAbstrak
Ketidakseimbangan kelas pada data (imbalanced class) merupakan tantangan signifikan dalam pengembangan model machine learning, yang sering kali menyebabkan penurunan kinerja model. Masalah ini sering ditemui dalam data nyata, di mana proporsi data antara kelas mayoritas dan minoritas sangat tidak seimbang. Salah satu pendekatan yang umum digunakan untuk mengatasi masalah ini adalah oversampling, yang berfungsi untuk menyeimbangkan distribusi kelas dengan menambahkan data sintetis ke kelas minoritas. Teknik oversampling yang paling populer adalah Synthetic Minority Oversampling Technique (SMOTE), meskipun metode ini memiliki kelemahan seperti menghasilkan data yang kurang beragam dan kemungkinan munculnya outlier. Sebagai solusi alternatif, penelitian ini mengusulkan penggunaan metode Latin-Hypercube Sampling (LHS) yang dikombinasikan dengan k-Nearest Neighbor (k-NN) untuk meningkatkan kinerja klasifikasi pada data yang tidak seimbang. Kombinasi LHS dan k-NN diharapkan dapat menghasilkan data sintetis yang lebih berkualitas, sehingga meningkatkan performa model klasifikasi yang diukur menggunakan confusion matrix. Data yang digunakan dalam penelitian ini berasal dari berbagai online repository seperti KEEL, Kaggle, UCI, serta satu dataset penjurusan siswa SMK di PekanbaruUnduhan
Referensi
H. Wang and H. Huang, “Feature Space Oversampling Technique for Imbalanced Classification,” 2019 6th Int. Conf. Information, Cybern. Comput. Soc. Syst. ICCSS 2019, pp. 93–99, 2019, doi: 10.1109/ICCSS48103.2019.9115430.
R. Sauber-Cole and T. M. Khoshgoftaar, “The use of generative adversarial networks to alleviate class imbalance in tabular data: a survey,” J. Big Data, vol. 9, no. 1, 2022, doi: 10.1186/s40537-022-00648-6.
A. Puri and M. K. Gupta, “Improved Hybrid Bag-Boost Ensemble with K-Means-SMOTE-ENN Technique for Handling Noisy Class Imbalanced Data,” Comput. J., vol. 65, no. 1, pp. 124–138, 2022, doi: 10.1093/comjnl/bxab039.
H. Mardiansyah, R. Widia Sembiring, and S. Efendi, “Handling Problems of Credit Data for Imbalanced Classes using SMOTEXGBoost,” J. Phys. Conf. Ser., vol. 1830, no. 1, 2021, doi: 10.1088/1742-6596/1830/1/012011.
J. M. Johnson and T. M. Khoshgoftaar, “Survey on deep learning with class imbalance,” J. Big Data, vol. 6, no. 1, 2019, doi: 10.1186/s40537-019-0192-5.
W. Ustyannie and S. Suprapto, “Oversampling Method To Handling Imbalanced Datasets Problem in Binary Logistic Regression Algorithm,” IJCCS (Indonesian J. Comput. Cybern. Syst., vol. 14, no. 1, p. 1, 2020, doi: 10.22146/ijccs.37415.
S. Mutmainah, “Penanganan Imbalance Data Pada Klasifikasi,” SNATi, vol. 1, pp. 10–16, 2021.
I. Kunakorntum, W. Hinthong, and P. Phunchongharn, “A Synthetic Minority Based on Probabilistic Distribution (SyMProD) Oversampling for Imbalanced Datasets,” IEEE Access, vol. 8, pp. 114692–114704, 2020, doi: 10.1109/ACCESS.2020.3003346.
L. Zhang et al., “A class imbalance loss for imbalanced object recognition,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., vol. 13, pp. 2778–2792, 2020, doi: 10.1109/JSTARS.2020.2995703.
Z. Wang and H. Wang, “Global Data Distribution Weighted Synthetic Oversampling Technique for Imbalanced Learning,” IEEE Access, vol. 9, pp. 44770–44783, 2021, doi: 10.1109/ACCESS.2021.3067060.
C. Liu et al., “Constrained Oversampling: An Oversampling Approach to Reduce Noise Generation in Imbalanced Datasets With Class Overlapping,” IEEE Access, vol. 10, no. July 2020, pp. 91452–91465, 2022, doi: 10.1109/ACCESS.2020.3018911.
J. Engelmann and S. Lessmann, “Conditional Wasserstein GAN-based oversampling of tabular data for imbalanced learning,” Expert Syst. Appl., vol. 174, no. December 2020, p. 114582, 2021, doi: 10.1016/j.eswa.2021.114582.
Y. Il Kang and S. Won, “Weight decision algorithm for oversampling technique on class-imbalanced learning,” ICCAS 2010 - Int. Conf. Control. Autom. Syst., pp. 182–186, 2010, doi: 10.1109/iccas.2010.5669889.
C. Liu, X. Wang, K. Wu, J. Tan, F. Li, and W. Liu, “Oversampling for imbalanced time series classification based on generative adversarial networks,” 2018 IEEE 4th Int. Conf. Comput. Commun. ICCC 2018, pp. 1104–1108, 2018, doi: 10.1109/CompComm.2018.8780808.
S. Korkmaz, M. A. ?ahman, A. C. Cinar, and E. Kaya, “Boosting the oversampling methods based on differential evolution strategies for imbalanced learning,” Appl. Soft Comput., vol. 112, p. 107787, 2021, doi: 10.1016/j.asoc.2021.107787.
S. K. Lee, S. J. Hong, and S. Il Yang, “Oversampling for Imbalanced Data Classification Using Adversarial Network,” 9th Int. Conf. Inf. Commun. Technol. Converg. ICT Converg. Powered by Smart Intell. ICTC 2018, pp. 1255–1257, 2018, doi: 10.1109/ICTC.2018.8539543.
V. A. Briones-Segovia, V. Jiménez-Villar, J. A. Carrasco-Ochoa, and J. F. Martínez-Trinidad, “A new oversampling method in the string space,” Expert Syst. Appl., vol. 183, no. November 2020, 2021, doi: 10.1016/j.eswa.2021.115428.
G. Douzas, R. Rauch, and F. Bacao, “G-SOMO: An oversampling approach based on self-organized maps and geometric SMOTE,” Expert Syst. Appl., vol. 183, no. May, p. 115230, 2021, doi: 10.1016/j.eswa.2021.115230.
S. Feng, J. Keung, X. Yu, Y. Xiao, and M. Zhang, “Investigation on the stability of SMOTE-based oversampling techniques in software defect prediction,” Inf. Softw. Technol., vol. 139, no. August 2020, p. 106662, 2021, doi: 10.1016/j.infsof.2021.106662.
H. Zhou, X. Dong, S. Xia, and G. Wang, “Weighted oversampling algorithms for imbalanced problems and application in prediction of streamflow[Formula presented],” Knowledge-Based Syst., vol. 229, p. 107306, 2021, doi: 10.1016/j.knosys.2021.107306.
J. Liu, “A minority oversampling approach for fault detection with heterogeneous imbalanced data,” Expert Syst. Appl., vol. 184, no. July, p. 115492, 2021, doi: 10.1016/j.eswa.2021.115492.
K. U. Syaliman, A. Labellapansa, and A. Yulianti, “Improving the Accuracy of Features Weighted k-Nearest Neighbor using Distance Weight,” no. ICoSET 2019, pp. 326–330, 2020, doi: 10.5220/0009390903260330.
Z. Pan, Y. Wang, and W. Ku, “A new general nearest neighbor classification based on the mutual neighborhood information,” Knowledge-Based Syst., vol. 121, pp. 142–152, 2017, doi: 10.1016/j.knosys.2017.01.021.
Ö. F. Ertu?rul and M. E. Ta?luk, “A novel version of k nearest neighbor: Dependent nearest neighbor,” Appl. Soft Comput. J., vol. 55, pp. 480–490, 2017, doi: 10.1016/j.asoc.2017.02.020.
A. A. Nababan, O. S. Sitompul, and Tulus, “Attribute Weighting Based K-Nearest Neighbor Using Gain Ratio Attribute Weighting Based K-Nearest Neighbor Using Gain Ratio,” 2018.
X. Sun et al., “Smart Sampling for Reduced and Representative Power System Scenario Selection,” IEEE Open Access J. Power Energy, vol. 8, no. May, pp. 293–302, 2021, doi: 10.1109/OAJPE.2021.3093278.
Q. Wang et al., “Modified Algorithms for Fast Construction of Optimal Latin-Hypercube Design,” IEEE Access, vol. 8, pp. 191644–191658, 2020, doi: 10.1109/ACCESS.2020.3032122.
L. Zhu et al., “Pendekatan Resampling Data Untuk Menangani Masalah Ketidakseimbangan Kelas,” IEEE Access, vol. 8, no. 1, pp. 31–38, 2021, doi: 10.35508/jicon.v10i1.6554.
X. Wang, J. Xu, T. Zeng, and L. Jing, “Local distribution-based adaptive minority oversampling for imbalanced data classification,” Neurocomputing, vol. 422, pp. 200–213, 2021, doi: 10.1016/j.neucom.2020.05.030.
T. Kurbiel, H. G. Gckler, and D. Alfsmann, “A novel approach to the design of oversampling low-delay complex-modulated filter bank Pairs,” EURASIP J. Adv. Signal Process., vol. 2009, 2009, doi: 10.1155/2009/692861.
J. Mendes, M. Freitas, H. Siqueira, A. Lazzaretti, S. Stevan, and S. Pichorim, “Comparative Analysis Among Feature Selection of sEMG Signal for Hand Gesture Classification by Armband,” IEEE Lat. Am. Trans., vol. 18, no. 6, pp. 1135–1143, 2020.
A. Islam, S. B. Belhaouari, A. U. Rehman, and H. Bensmail, “K Nearest Neighbor OveRsampling approach: An open source python package for data augmentation,” Softw. Impacts, vol. 12, no. February, p. 100272, 2022, doi: 10.1016/j.simpa.2022.100272.
M. Kumar, N. K. Rath, A. Swain, and S. K. Rath, “Feature Selection and Classification of Microarray Data using MapReduce based ANOVA and K-Nearest Neighbor,” Procedia Comput. Sci., vol. 54, pp. 301–310, 2015, doi: 10.1016/j.procs.2015.06.035.
P. Nair and I. Kashyap, “Classification of medical image data using k nearest neighbor and finding the optimal k value,” Int. J. Sci. Technol. Res., vol. 9, no. 4, pp. 221–226, 2020.
G. I. Okolo, S. Katsigiannis, and N. Ramzan, “IEViT: An enhanced vision transformer architecture for chest X-ray image classification,” Comput. Methods Programs Biomed., vol. 226, p. 107141, 2022, doi: 10.1016/j.cmpb.2022.107141.
J. A. Romero-del-Castillo, M. Mendoza-Hurtado, D. Ortiz-Boyer, and N. García-Pedrajas, “Local-based k values for multi-label k-nearest neighbors rule,” Eng. Appl. Artif. Intell., vol. 116, no. June, p. 105487, 2022, doi: 10.1016/j.engappai.2022.105487.
S. Suyanto, P. E. Yunanto, T. Wahyuningrum, and S. Khomsah, “A multi-voter multi-commission nearest neighbor classifier,” J. King Saud Univ. - Comput. Inf. Sci., vol. 34, no. 8, pp. 6292–6302, 2022, doi: 10.1016/j.jksuci.2022.01.018.
X. Zhang, H. Xiao, R. Gao, H. Zhang, and Y. Wang, “K-nearest neighbors rule combining prototype selection and local feature weighting for classification,” Knowledge-Based Syst., vol. 243, 2022, doi: 10.1016/j.knosys.2022.108451.
N. García-Pedrajas and D. Ortiz-Boyer, “Boosting k-nearest neighbor classifier by means of input space projection,” Expert Syst. Appl., vol. 36, no. 7, pp. 10570–10582, 2009, doi: 10.1016/j.eswa.2009.02.065.
S. Ougiaroglou and G. Evangelidis, “Fast and accurate k-nearest neighbor classification using prototype selection by clustering,” Proc. 2012 16th Panhellenic Conf. Informatics, PCI 2012, no. i, pp. 168–173, 2012, doi: 10.1109/PCi.2012.69.
Z. Pan, Y. Wang, and W. Ku, “A new k-harmonic nearest neighbor classifier based on the multi-local means,” Expert Syst. Appl., vol. 67, pp. 115–125, 2017, doi: 10.1016/j.eswa.2016.09.031.
J. Wang, P. Neskovic, and L. N. Cooper, “Improving nearest neighbor rule with a simple adaptive distance measure,” Pattern Recognit. Lett., vol. 28, no. 2, pp. 207–213, 2007, doi: 10.1016/j.patrec.2006.07.002.
K. U. Syaliman, E. B. Nababan, and O. S. Sitompul, “Improving the accuracy of k-nearest neighbor using local mean based and distance weight,” J. Phys. Conf. Ser., vol. 978, no. 1, pp. 1–6, 2018, doi: 10.1088/1742-6596/978/1/012047.
Y. Yuliska and K. U. Syaliman, “Peningkatan Akurasi K-Nearest Neighbor Pada Data Index Standar Pencemaran Udara Kota Pekanbaru,” IT J. Res. Dev., vol. 5, no. 1, pp. 11–18, 2020, doi: 10.25299/itjrd.2020.vol5(1).4680.
X. Wu et al., Top 10 algorithms in data mining, vol. 14, no. 1. 2008. doi: 10.1007/s10115-007-0114-2.
Unduhan
Diterbitkan
Terbitan
Bagian
Lisensi
Hak Cipta (c) 2024 Jurnal Komputer Terapan
Artikel ini berlisensiCreative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Penulis yang mempublikasikan naskah pada jurnal ini setuju tentang hal-hal berikut ini:
- Penulis memegang hak cipta untuk proses, prosedur, atau artikel yang dideskripsikan pada naskah dan memberikan hak kepada jurnal untuk pertama kali mempublikasikan naskah dibawah Creative Commons Attribution License yang memperbolehkan orang lain untuk menyebarkan hasil penelitian dengan pengakuan hak cipta penulis dan jurnal.
- Penulis memegang hak untuk mempublikasikan kembali semua atau sebagian hasil penelitian pada naskah lain tetapi tidak memberikan hak kepada pihak ketiga lain untuk mencetak dan mempublikasikan ulang hasil penelitian yang sama.
- Penulis diperbolehkan dan disarankan untuk menginformasikan naskah yang telah terbit secara online (seperti pada repository institusi atau website milik penulis) yang akan meningkatkan produktivitas rujukan dan sitasi dari naskah yang telah terbit.
Ciptaan disebarluaskan di bawah Lisensi Creative Commons Atribusi-NonKomersial-BerbagiSerupa 4.0 Internasional.
