Single Exponential Smoothing-Multilayer Perceptron Untuk Peramalan Pengunjung Unik Jurnal Elektronik

Authors

  • Miftakhul Anggita Bima Ferdinand Universitas Negeri Malang
  • Aji Prasetya Wibawa Universitas Negeri Malang
  • Ilham Ari Elbaith Zaeni Universitas Negeri Malang
  • Harits Ar Rosyid Universitas Negeri Malang

DOI:

https://doi.org/10.12928/mf.v2i2.2034

Keywords:

Pengunjung Unik, Jurnal Elektronik, Forecasting, Multilayer Perceptron, Single Exponential Smoothing.

Abstract

Jumlah kunjungan rerata pengunjung unik per hari pada jurnal elektronik menunjukkan bahwa hasil terbitan karya ilmiah website tersebut menarik. Sehingga jumlah pengunjung unik dijadikan indikator penting dalam mengukur keberhasilan sebuah jurnal elektronik untuk memenuhi perluasan, penyebaran dan percepatan sistem akreditasi jurnal. Pengunjung Unik merupakan jumlah pengunjung per Internet Address (IP) yang mengakses sebuah jurnal elektronik dalam kurun waktu tertentu. Terdapat beberapa metode yang biasa digunakan untuk peramalan, diantaranya adalah Multilayer Perceptron (MLP). Kualitas data berpengaruh besar dalam membangun model MLP yang baik, karena sukses tidaknya permodelan pada MLP sangat dipengaruhi oleh data input. Salah satu cara untuk meningkatkan kualitas data adalah dengan melakukan smoothing pada data tersebut. Pada penelitian ini digunkan metode peramalan Multilayer Perceptron berdasarkan penelitian sebelumnya dengan kombinasi data training dan testing 80%-20% dengan asitektur 2-1-1 dan learning rate 0,4. Selanjutnya untuk meningkatkan kualitas data dilakukan smoothing dengan menerapkan metode Single Exponential Smoothing. Dari penelitian yang dilakukan diperoleh hasil terbaik menggunakan alpha 0.9 dengan hasil akurasi MSE 94.02% dan RMSE 75.54% dengan lama waktu eksekusi 580,27 detik.

The number of visits by the average unique visitor per day on electronic journals shows that the published scientific papers on the website are interesting. So that the number of unique visitors is used as an important indicator in measuring the success of an electronic journal to meet the expansion, dissemination and acceleration of the journal accreditation system. Unique Visitors is the number of visitors per Internet Address (IP) who access an electronic journal within a certain period of time. There are several methods commonly used for forecasting, including the Multilayer Perceptron (MLP). Data quality has a big influence in building a good MLP model, because the success or failure of modeling in MLP is greatly influenced by the input data. One way to improve data quality is by smoothing the data. In this study, the Multilayer Perceptron forecasting method was used based on previous research with a combination of training data and testing 80% -20% with a 2-1-1 architecture and a learning rate of 0.4. Furthermore, to improve data quality, smoothing is done by applying the Single Exponential Smoothing method. From the research conducted, the best results were obtained using alpha 0.9 with MSE accuracy of 94.02% and RMSE 75.54% with a long execution time of 580.27 seconds.

References

Jamaluddin, "Mengenal Elektronik Jurnal Dan Manfaatnya Bagi Pengembangan Karir Pustakawan," Pustak. madya di UPT Pustak. Unhas, vol. XIV, no. 2, pp. 38-44, 2015,

B. Siregar, E. B. Nababan, A. Yap, U. Andayani, and Fahmi, "Forecasting of raw material needed for plastic products based in income data using ARIMA method," Proceeding - 2017 5th Int. Conf. Electr. Electron. Inf. Eng. Smart Innov. Bridg. Futur. Technol. ICEEIE 2017, vol. 2018-Janua, no. April 2018, pp. 135-139, 2018, doi: 10.1109/ICEEIE.2017.8328777.

Muladi, S. A. Siregar, and A. P. Wibawa, "Double Exponential-Smoothing Neural Network for Foreign Exchange Rate Forecasting," Proc. - 2nd East Indones. Conf. Comput. Inf. Technol. Internet Things Ind. EIConCIT 2018, pp. 118-122, 2018, doi: 10.1109/EIConCIT.2018.8878591.

W. Lestari, A. P. Wibawa, and T. Widiyaningtyas, "PERAMALAN SESSIONS PADA SEBUAH WEBSITE E-JURNAL MENGGUNAKAN MULTILAYER PERCEPTRON," Skripsi, Univ. Negeri Malang, vol. 4, pp. 5-10, 2019.

X. Qin, C. Jiang, and J. Wang, "Online clustering for wind speed forecasting based on combination of RBF neural network and persistence method," in 2011 Chinese Control and Decision Conference (CCDC), 2011, pp. 2798-2802, doi: 10.1109/CCDC.2011.5968687.

Y. S. Park and S. Lek, Artificial Neural Networks: Multilayer Perceptron for Ecological Modeling, vol. 28. Elsevier, 2016.

N. J. Nilsson, "INTRODUCTION TO MACHINE LEARNING AN EARLY DRAFT OF A PROPOSED TEXTBOOK Department of Computer Science," Mach. Learn., vol. 56, no. 2, pp. 387-99, 2005, doi: 10.1016/j.neuroimage.2010.11.004.

T. Marwala, "Chapter 1. יתמיאמ," The Jerusalem Talmud,First order: Zeraim, Tractate Berakhot, no. 2018, pp. 1-10, 2018, doi: 10.1515/9783110800487.39.

U. Hamida, "Penggunaan Artificial Neural Network (Ann) Untuk Memodelkan Kebutuhan Energi Untuk Transportasi," J. Teknol. dan Manaj., vol. 12, no. 2, pp. 57-65, 2014.

D. Sutarya, "PEMROSESAN AWAL DATA RUNTUN WAKTU HASIL PENGUKURAN UNTUK IDENTIFIKASI SISTEM TUNGKU SINTER DEGUSSA," J. BATAN, vol. 09, no. 16, pp. 1-12, 2016.

N. S. Muhamad and A. M. Din, "Exponential Smoothing Techniques on Time Series," Proc. 6th Int. Conf. Comput. Informatics, ICOCI 2017, no. 217, pp. 62-68, 2017.

A. Fahlevi, F. A. Bachtiar, and B. D. Setiawan, "Perbandingan Holt ' s dan Winter ' s Exponential Smoothing untuk Peramalan Indeks Harga Konsumen Kelompok Transportasi , Komunikasi dan Jasa Keuangan," J. Pengemb. Teknol. Inf. dan Ilmu Komput. Univ. Brawijaya, vol. 2, no. 12, pp. 6136-6145, 2018.

E. Oktafiani, F. Andriyani, and Noeryati, "Aplikasi Pemulusan Eksponensial Dari Brown Dan Dari Holt Untuk Data Yang Memuat Trend," Pros. Semin. Nas. Apl. Sains Teknol. Periode III Yogyakarta, no. November, pp. 447-455, 2012.

S. G. K. Patro and K. K. sahu, "Normalization: A Preprocessing Stage," Iarjset, vol. 2, no. 3, pp. 20-22, 2015, doi: 10.17148/iarjset.2015.2305.

C. Saranya and G. Manikandan, "A study on normalization techniques for privacy preserving data mining," Int. J. Eng. Technol., vol. 5, no. 3, pp. 2701-2704, 2013.

A. Sudarsono, "Jaringan Syaraf Tiruan Untuk Memprediksi Laju Pertumbuhan Penduduk Menggunakan Metode," Media Infotama, vol. 12, no. 1, pp. 61-69, 2016.

H. Amakdouf, M. El Mallahi, A. Zouhri, A. Tahiri, and H. Qjidaa, "Classification and recognition of 3d image of charlier moments using a multilayer perceptron architecture," Procedia Comput. Sci., vol. 127, pp. 226-235, 2018, doi: 10.1016/j.procs.2018.01.118.

T. Minemoto, T. Isokawa, H. Nishimura, and N. Matsui, "Feed forward neural network with random quaternionic neurons," Signal Processing, vol. 136, no. November 2016, pp. 59-68, 2017, doi: 10.1016/j.sigpro.2016.11.008.

E. Eǧrioǧlu, Ç. H. Aladaǧ, and S. Günay, "A new model selection strategy in artificial neural networks," Appl. Math. Comput., vol. 195, no. 2, pp. 591-597, 2008, doi: 10.1016/j.amc.2007.05.005.

M. Cerjan, A. Petričić, and M. Delimar, "HIRA model for short-term electricity price forecasting," Energies, vol. 12, no. 3, 2019, doi: 10.3390/en12030568.

B. Warsito, Subanar, Abdurakhman, and Widodo, "Penentuan Bobot Model Neural Network Untuk Data Time Series," Konf. Nas. Mat. XVI, 2012.

M. H. Gholizadeh and M. Darand, "Forecasting precipitation with artificial neural networks (case study: Tehran)," J. Appl. Sci., vol. 9, no. 9, pp. 1786-1790, 2009, doi: 10.3923/jas.2009.1786.1790.

S. D. Anggraini, "PREDIKSI NILAI TUKAR MATA UANG ASING MENGGUNAKAN EXTREME LEARNING MACHINE," MATHunesa J. Ilm. Mat., vol. 3, no. 6, pp. 110-115, 2017.

Downloads

Published

2020-08-29

Issue

Section

Articles