APLIKASI DATA MINING MENGGUNAKAN MULTIPLE LINEAR REGRESSION UNTUK PENGENALAN POLA CURAH HUJAN

Irwan Budiman, Artesya Nanda Akhlakulkarimah

Abstract


The development of information technology in today's era of globalization is growing rapidly. It also has created the development of a lot of data, including data about the weather. The method of data analysis that we used is multiple linear regression. F test, partial correlation test and coefficient of determination were used in this research. After we got a regression model with two independent variables, then we did testing for coefficient of determination. From the result, we knew that the relevance between the number of rainy days with the rainfall was very strong. The relevance between the duration of solar radiation with the rainfall was strong. Whereas, the relevance between the number of rainy days with duration of sun exposure was very strong. Coefficient of determination was 0.5778. It meant that multiple linear regression model had a reliability rate of 57,78%. The Conclusions of this research are the number of rainy days and duration of sun exposure are affecting significantly with rainfall. The regression model which used is 57,78%, it means that rainfall is influenced by 57,78% of independent variables which measured in this research.

Keywords: Multiple Linear Regression, data mining, rainfall.

Perkembangan teknologi informasi pada era globalisasi saat ini sangat berkembang pesat. Perkembangan ini juga telah melahirkan perkembangan banyak data, termasuk data-data tentang cuaca. Metode analisis data yang digunakan dengan multiple linear regression. Pada penelitian ini digunakan uji F, uji korelasi parsial dan koefisien determinasinya. Setelah didapatkan model regresi dengan dua variable bebas, kemudian dilakukan pengujian terhadap koefisien regresi. Dari hasil perhitungan, dapat diketahui keterkaitan antara jumlah hari hujan dengan curah hujan sangat kuat. Keterkaitan antara lama penyinaran dan curah hujan kuat. Sedangkan, keterkaitan antara jumlah hari hujan dan lama penyinaran sangat kuat. Koefisien determinasinya 0,5778. Artinya tingkat kecocokan model multiple linear regression memiliki tingkat kehandalan 57,78%. Kesimpulan dari penelitian ini adalah jumlah hari hujan dan lamanya penyinaran matahari berpengaruh signifikan terhadap curah hujan. Model regresi yang digunakan memberikan hasil 57,78% yang berarti curah hujan dipengaruhi oleh 57,78% variable bebas yang diukur pada penelitian ini.

Kata Kunci: Multiple Linear Regression, data mining, curah hujan.

Full Text:

PDF

References


Kurniadi, Eka, dkk. 2012. “Multiple linear regression Menggunakan Aplikasi Matlabâ€. Universitas Pendidikan Ganesha Singaraja. Bali.

Larose, Daniel T. 2006. “Data mining Methods and Modelsâ€. John Wiley & Sons Inc.Hoboken New Jersey.

Lesmana, Eman dan Riaman. 2013. “Penggunaan Model Regresi linear Berganda pada Program Penggemukan Sapi PO (Peranakan Ongole) serta Analisis BCR (Benefit Cost Ratio) Penggunaan Bahan Pakan Keringâ€. Prosiding Seminar Nasional Sains dan Teknologi Nuklir PTNBR-BATAN Bandung 4 Juli 2013.

Turban, E., Aronson Jay E. dan Liang T. 2005. “Decision Support Systems and Intelligent Systems Seventh Editionâ€. Andi. Yogyakarta.




DOI: http://dx.doi.org/10.20527/klik.v2i1.16

Copyright (c) 2016 KLIK - JURNAL ILMIAH ILMU KOMPUTER



Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.joomla
counter View My Stats