PEMANFATAN DATA MINING UNTUK PRAKIRAAN CUACA
DOI:
https://doi.org/10.31172/jmg.v12i2.100Keywords:
data mining, association rule, classification tree, random forest, cuacaAbstract
Proses prakiraan cuaca memerlukan banyak komponen data cuaca, jumlah data yang besar serta kemampuan prakirawan. Hal ini menyebabkan ketepatan dan kecepatan prakiraan kurang terpenuhi. Untuk memecahkan masalah tersebut, telah dilakukan penelitian model prediksi menggunakan beberapa teknik data mining yakni Association Rule, C4.5, Classification dan Random Forest. Data masukan adalah data sinoptik 9 stasiun maritim tahun 2009. Data masukan tersebut terdiri dari kecepatan angin, tutupan awan, suhu udara dan suhu titik embun. Data untuk pengujian model adalah data sinoptik Stasiun Meteorologi Maritim Tanjung Priok sejak tahun 2002 hingga 2010. Dari serangkaian pembuatan, pemilihan dan pengujian model, hasil penelitian menunjukkan Association Rule mempunyai tingkat akurasi 60.9%, sedangkan C4.5 mempunyai tingkat akurasi 68.5%. Dengan demikian model prediksi yang dipilih adalah model prediksi C4.5. Komponen cuaca yang dominan memungkinkan terjadinya hujan adalah suhu udara, suhu titik embun, dan tutupan awan.
Weather forecasting process needs many weather components, big data size and forecaster experience. They cause the accuracy and rapid of forecasting were not well-fulfilled. In order to solve this problem, the research of prediction model was done by using Association Rule and Classification (C4.5, Classification Tree and Random Forest) methods. The input of model production were wind speed, cloud cover, dew point temperature and temperature from 9 marine stations in 2009. The input for testing the resulted model was synoptic data of Tanjung Priok Marine Station since 2002-2010. The result shows the accuracy of C4.5 is highest than others. Accuracy of C4.5 and Association Rule are about 68.5%, and 60.9%, respectively. Thus, the appropriate prediction model is the C4.5. Dominant weather component of C4.5 are cloud cover, dew point temperatur and temperature.
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are 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 an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their 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 published work.