TY - GEN
T1 - Weather prediction using fuzzy rough clustering
AU - Iqbal, Mohammad
AU - Mukhlash, Imam
AU - Wibowo, Inu Laksito
N1 - Publisher Copyright:
© 2016 Author(s).
PY - 2016/6/17
Y1 - 2016/6/17
N2 - Weather prediction is an important factor that can brought broad impact for other fields such as agriculture, business, and environmental. There were several researches conducted to develop tools or theories for weather prediction. In past few years, some researcher analyzed some seasonal attributes such as daily average temperature and daily average humidity to get better result in weather prediction. In this research, integration of classification and clustering technique is employed to analyze and predict the weather pattern. For the simulation, we used climate data in Perak, Surabaya, East Java, Indonesia which consist of temperature and humidity numerical data processed into categorical data using fuzzy rough clustering. In this research, we do four simulations: a month, a year, five years and the whole datasets. We transform humidity, temperature attributes into categorical data and speed of wind, still numerical data. Simulation results show that this method can predict the weather very well with the average accuracy above 80%.
AB - Weather prediction is an important factor that can brought broad impact for other fields such as agriculture, business, and environmental. There were several researches conducted to develop tools or theories for weather prediction. In past few years, some researcher analyzed some seasonal attributes such as daily average temperature and daily average humidity to get better result in weather prediction. In this research, integration of classification and clustering technique is employed to analyze and predict the weather pattern. For the simulation, we used climate data in Perak, Surabaya, East Java, Indonesia which consist of temperature and humidity numerical data processed into categorical data using fuzzy rough clustering. In this research, we do four simulations: a month, a year, five years and the whole datasets. We transform humidity, temperature attributes into categorical data and speed of wind, still numerical data. Simulation results show that this method can predict the weather very well with the average accuracy above 80%.
UR - http://www.scopus.com/inward/record.url?scp=84984577966&partnerID=8YFLogxK
U2 - 10.1063/1.4953995
DO - 10.1063/1.4953995
M3 - Conference contribution
AN - SCOPUS:84984577966
T3 - AIP Conference Proceedings
BT - 2016 Conference on Fundamental and Applied Science for Advanced Technology, ConFAST 2016
A2 - Winanda, Rara Sandhy
A2 - Hidayah, Qonitatul
A2 - Yanto, Iwan Tri Riyadi
A2 - Irsalinda, Nursyiva
A2 - Aji, Oktira Roka
A2 - Kusuma, Damar Yoga
A2 - Inayati, Syarifah
PB - American Institute of Physics Inc.
T2 - 2016 Conference on Fundamental and Applied Science for Advanced Technology, ConFAST 2016
Y2 - 25 January 2016 through 26 January 2016
ER -