Implementasi Metode Adaptive Neuro-Fuzzy Inference System (ANFIS) terhadap Prediksi Tingkat Kemiskinan di Kabupaten/Kota Jawa Timur

Tsalsya Ni'matul Aulia, Evawati Alisah, Erna Herawati

Abstract


This study aims to evaluate the effectiveness of the Adaptive Neuro-Fuzzy Inference System (ANFIS) in predicting poverty levels in the Regencies/Cities of East Java Province. ANFIS is a hybrid method that combines fuzzy logic and artificial neural networks, enabling it to handle complex socio-economic data with inherent uncertainty. The model uses the open unemployment rate, population growth rate, and school participation rate for ages 16–18, while the target variable is the poverty level. The data were obtained from the central statistics agency (BPS) and cover the years 2022, 2023, and 2024. Modeling was carried out using a time-based split approach, in which 2022 data were used as training data to predict the 2023 poverty level, and 2023 data were used as testing data to predict the 2024 poverty level. The training process employed a hybrid algorithm combining Least Squares Estimation (LSE) and backpropagation. Evaluation was conducted using Mean Squared Error (MSE) to assess the model’s accuracy. The results indicate that ANFIS is capable of producing accurate predictions of poverty levels, with an MSE value of 3.98 × 10⁻²⁴ in the testing phase. Therefore, ANFIS is considered effective in analyzing and predicting socio-economic phenomena at the regional level.

Keywords


ANFIS; Time-Based Split; MSE.

Full Text:

PDF

References


[1] Y. Huang, “Advances in artificial neural networks – methodological development and application,” Algorithms, vol. 2, no. 3, pp. 973–1007, 2009. doi: 10.3390/algor2030973.

[2] Badan Pusat Statistik Provinsi Jawa Timur, Poverty profile of jawa timur province 2023, https://jatim.bps.go.id/en/publication/2024/03/26/0a47f129594de9a197806916/profil-kemiskinan- provinsi- jawa- timur- 2023.html, Diakses 23 November 2024.

[3] A. Yahya and R. E. Agustina, “Pembangunan desa dan tingkat kemiskinan kabupaten di indonesia,” Jurnal Litbang Sukowati, vol. 6, no. 1, pp. 98–108, 2022. doi: 10.32630/sukowati.v6i1.329.

[4] S. S. Aldin and H. Sözer, “Comparative analysis of ann and anfis models in thermal data prediction,” Journal of Advanced Computational Techniques, vol. 11, no. 1, pp. 45–56, 2022. doi: 10.31462/jcemi.2022.02119139.

[5] A. Matsniya et al., “Penerapan adaptive neuro fuzzy inference system (anfis) dalam prediksi produksi tembakau di jember,” InComTech: Jurnal, vol. 13, no. 1, pp. 51–68, 2023. doi: 10.22441/incomtech.v13i1.15655.

[6] A. Wisudawaty and T. Djatna, “Penerapan anfis untuk prediksi penjualan keripik kacampring,” Indonesian Journal of Business and Technology, vol. 12, no. 1, pp. 45–52, 2024. doi: 10.24961/j.tek.ind.pert.2024.34.1.1.

[7] T. Rohana et al., “Kajian model backpropagation dan hybrid anfis dalam memprediksi pertumbuhan penduduk di kabupaten karawang,” Journal of Information System Research (JOSH), vol. 4, no. 2, pp. 374–381, 2023. doi: 10.47065/josh.v4i2.2547.

[8] J. S. R. Jang, “Anfis: Adaptive-network-based fuzzy inference system,” IEEE Transactions on Systems, Man, and Cybernetics, vol. 23, no. 3, pp. 665–685, 1993. doi: 10.1109/21.256541.

[9] R. Medar et al., “Impact of training and testing data splits on accuracy of time series forecasting in machine learning,” in 2017 International Conference on Computing, Communication, Control and Automation (ICCUBEA), 2017, pp. 1–6. doi: 10.1109/ICCUBEA.2017.8463779.

[10] V. Cerqueira et al., “Evaluating time series forecasting models: An empirical study on performance estimation methods,” Machine Learning, vol. 109, no. 11, pp. 1997–2028, 2020. doi: 10.1007/s10994-020-05910-7.

[11] I. Lestari et al., “Pengaruh jumlah penduduk, tingkat pendidikan, dan upah terhadap pengangguran di jawa timur,” INDEPENDENT: Journal of Economics, vol. 4, no. 2, pp. 141–148, 2024.

[12] G. M. Putriaji et al., “Optimal adaptive neuro-fuzzy inference system architecture for time series forecasting with calendar effect,” Sains Malaysiana, vol. 51, no. 9, pp. 2765–2777, 2022. doi: 10.17576/jsm-2022-5109-23.

[13] M. Bilgili et al., “One-day ahead forecasting of energy production from run-of-river hydroelectric power plants with a deep learning approach,” Scientia Iranica, vol. 29, no. 4, pp. 1853–1867, 2022. doi: 10.24200/SCI.2022.58636.5825.

[14] U. Sholihah, “Penerapan logika fuzzy tsukamoto untuk memprediksi curah hujan di kabupaten kotawaringin timur,” EJECTS: E-Journal Computer, Technology and Informations System, vol. 1, no. 2, p. 78, 2022.

[15] J. Heaton, “Review of: Goodfellow, bengio, and courville, deep learning,” Genetic Programming and Evolvable Machines, vol. 19, no. 3, pp. 305–307, 2017. doi: 10.1007/s10710017-9314-z.

[16] A. R. Putra et al., “Level optimum hyperparameter tuning epoch dalam klasifikasi citra bencana kebakaran,” Jurnal Teknologi Rekayasa (JTERA), vol. 7, no. 2, pp. 209–216, 2022.doi: 10.31544/jtera.v7.i2.2022.209-216.




DOI: https://doi.org/10.18860/jrmm.v5i4.37645

Refbacks

  • There are currently no refbacks.