IMPLEMENTASI CLASS WEIGHT, DATA AUGMENTATION DAN RESNET101 UNTUK MENGATASI OVERFITTING DAN IMBALANCED DATA PADA KLASIFIKASI PISTACHIO
DOI:
https://doi.org/10.32699/biner.v5i2.11352Keywords:
Resnet101, Transfer learning, Class Weight, Data augmentation, Pistachio ClassificationAbstract
Klasifikasi kacang Pistachio varietas Kirmizi dan Siirt secara manual rentan terhadap kesalahan akibat kemiripan morfologi antar varietas, sementara penelitian sebelumnya menggunakan ResNet50 mengalami overfitting parah dengan gap accuracy 78% dan belum menangani ketidakseimbangan data secara eksplisit. Penelitian ini bertujuan mengimplementasikan ResNet101 dengan augmentasi data dan class weighting untuk meningkatkan akurasi klasifikasi sekaligus mengatasi overfitting dan imbalanced data. Metode yang digunakan adalah kerangka CRISP-DM dengan 2148 citra (rasio 80:10:10) dan empat skenario eksperimen kombinasi data augmentasi dan class weighting berbasis Transfer Learning ResNet101. Hasil menunjukkan ResNet101 dengan class weighting tanpa augmentasi mencapai accuracy, precision, recall, dan F1-score sebesar 98,31%, melampaui Inception-V3 (96%) sebesar 2,31 poin dan ResNet50 (86%) sebesar 12,31 poin, dengan class weighting lebih bermanfaat saat dikombinasikan augmentasi dibanding kondisi data bersih. Augmentasi justru menurunkan akurasi karena latar belakang citra yang homogen, menunjukkan efektivitasnya bergantung pada karakteristik dataset dan bukan berlaku universal. Disimpulkan bahwa ResNet101 dengan class weighting tanpa augmentasi merupakan konfigurasi optimal untuk dataset Pistachio berkarakteristik visual homogen, dan penelitian lanjutan disarankan mengeksplorasi arsitektur modern dengan dataset lebih bervariasi.
Downloads
References
[1] L. Lisda, K. Kusrini, and D. Ariatmanto, “Classification of Pistachio Nut Using Convolutional Neural Network,” Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi, vol. 8, no. 1, pp. 71–77, Jan. 2023, doi: 10.25139/inform.v8i1.5685.
[2] E. AVUÇLU, “Classification of Pistachio Images Using VGG16 and VGG19 Deep Learning Models,” International Scientific and Vocational Studies Journal, vol. 7, no. 2, pp. 79–86, Dec. 2023, doi: 10.47897/bilmes.1328313.
[3] D. Singh et al., “Classification and Analysis of Pistachio Species with Pre-Trained Deep Learning Models,” Electronics (Switzerland), vol. 11, no. 7, Apr. 2022, doi: 10.3390/electronics11070981.
[4] K. Azmi and S. Defit, “Implementasi Convolutional Neural Network (CNN) Untuk Klasifikasi Batik Tanah Liat Sumatera Barat,” Jurnal Unitek, vol. 16, no. 1, p. 2023, 2023, doi: 10.52072/unitek.v16i1.504.
[5] R. Faturrahman, Y. S. Hariyani, and S. Hadiyoso, “Klasifikasi Jajanan Tradisional Indonesia berbasis Deep Learning dan Metode Transfer Learning,” ELKOMIKA: Jurnal Teknik Energi Elektrik, Teknik Telekomunikasi, & Teknik Elektronika, vol. 11, no. 4, p. 945, Oct. 2023, doi: 10.26760/elkomika.v11i4.945.
[6] A. Vierisyah and R. Maulana Fajri, “KLASIFIKASI KANKER PARU PARU MENGGUNAKAN CNN DENGAN 5 ARSITEKTUR,” Jini Global : Journal of Intelligent Networks and IoT Global, 2023, doi: doi.org/10.36982/jinig.v1i2.3643.
[7] E. Goceri and A. A. Karakas, “COMPARATIVE EVALUATIONS OF CNN BASED NETWORKS FOR SKIN LESION CLASSIFICATION,” International Conferences Computer Graphics, Visualization, Computer Vision and Image Processing 2020, 2020.
[8] U. N. Oktaviana, R. Hendrawan, A. D. K. Annas, and G. W. Wicaksono, “Classification of Rice Diseases based on Leaf Image Using Resnet101 Trained Model,” Jurnal RESTI, vol. 5, no. 6, pp. 1216–1222, Dec. 2021, doi: 10.29207/resti.v5i6.3607.
[9] E. Avuçlu, “Classification Of Pistachio Images With The ResNet Deep Learning Model,” Selcuk Journal of Agricultural and Food Sciences, Aug. 2023, doi: 10.15316/sjafs.2023.029.
[10] I. A. Özkan, M. Köklü, and R. Saraçoğlu, “Classification of pistachio species using improved k-NN classifier,” Progress in Nutrition, vol. 23, no. 2, Feb. 2021, doi: 10.23751/pn.v23i2.9686.
[11] H. Akbar and W. K. Sanjaya, “Kajian Performa Metode Class Weight Random Forest pada Klasifikasi Imbalance Data Kelas Curah Hujan,” Jurnal Sains, Nalar, dan Aplikasi Teknologi Informasi, vol. 3, no. 1, Dec. 2023, doi: 10.20885/snati.v3i1.30.
[12] D. Mualfah, R. Ardiansyah, and R. Gunawan, “Classification of DDoS attacks using the random forest method and class weight technique on the CICDDoS2019 dataset,” Jurnal CoSciTech (Computer Science and Information Technology), vol. 6, no. 3, pp. 530–535, Dec. 2025, doi: 10.37859/coscitech.v6i3.10731.
[13] B. Marsheila Islami and A. Nugroho, “Analisis Algoritma KNN dan Penerapan SMOTE dalam Deteksi Dini Kanker Paru-Paru,” Jurnal Qua Teknika, vol. 15, no. 2, p. 15, 2025, doi: 10.35457/quateknika.v15i02.4603.
[14] S. Sucipto, D. Dwi Prasetya, and T. Widiyaningtyas, “Educational Data Mining: Multiple Choice Question Classification in Vocational School,” MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer, vol. 23, no. 2, pp. 379–388, Mar. 2024, doi: 10.30812/matrik.v23i2.3499.
[15] M. Shelin Sahira, F. Dessela Putri, A. Ristyawan, and E. Daniati, “Penggunaan Algoritma K-Nearest Neighbor Untuk Klasifikasi Data Diabetes Pada Wanita,” Prosiding Seminar Nasional Inovasi Teknologi, vol. 8, pp. 2549–7952, 2024, doi: doi.org/10.29407/inotek.v8i1.4981.
[16] C. Chazar, M. A. Adli, J. Pardede, and M. Ichwan, “Pendekatan Augmentasi Citra Fundus pada Model EfficientNet untuk Klasifikasi Tingkat Keparahan Retinopati Diabetik dengan Dataset Tidak Seimbang,” Journal MIND Journal | ISSN, vol. 10, no. 2, pp. 180–194, 2025, doi: 10.26760/mindjournal.v10i2.180-194.
[17] J. Khatib Sulaiman, A. Nur Salsabila, M. Liebenlito, D. Urfina Zulkifli, and U. Syarif Hidayatullah Jakarta, “Perbandingan Deteksi Alzheimer: ViT, CNN dan ViT dengan Bobot pada Citra Medis,” Indonesian Journal of Computer Science Attribution, vol. 13, no. 1, pp. 2024–1401, 2024.
[18] T. S. Siadari and Y. S. Hariyani, “Klasifikasi Penyakit Periodontal Pada Citra Gigi Berwarna Menggunakan Arsitektur Berbasis Convolutional Neural Network,” JETT: Jurnal Elektro dan Telekomunikasi Terapan, vol. 11, pp. 13–19, 2024, doi: 10.25124/jett.v11i2.8090.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Biner : Jurnal Ilmiah Informatika dan Komputer
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
An author who publishes in this Journal agrees to the following terms:
- Author retains the copyright and grants the journal the right of first publication of the work simultaneously licensed under the Creative Commons Attribution-ShareAlike 4.0 License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal
- Author is able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book) with the acknowledgement of its initial publication in this journal.
- Author is permitted and encouraged to post his/her work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of the published work (See The Effect of Open Access).
