Missing meteorological data is one of the problems facing specialists and designers of water recources projects and it,s necessary to reconstruct them. There are different methods for infilling missing data. In this research, performance of tree model and neural network for infiiling missing evaporation data from 4 meteorological stations in khouzestan province, were evaluated. The data were divided into two periods: 4 and 12 years and in each period 5%, 10% and 20% of data were deliberately missed and had been filled by models. In tree model coefficient of determination for 4years period were: 85%, 75% and 85%, and for 12years period were: 90%, 83% and 84% respectively. In neural network model coefficient of determination for 4years period were: 85%, 75% and 85% and for 12years period were: 90%, 82% and 85% respectively. A higher coefficient value for 12 years period showed that models are more accurate to estimate missing data for longer term statistical data. By increasing missing data from 5% to 20%, accuracy of models were diminished. This research also indicated that both models have similar accuracy in the estimation of missing data.
Vahabi Mashhor, M., & Rahimikhoob, A. (2016). Comparison between neural network and M5 tree models
For reconstructing missing evaporation data of khouzestan. Journal of Water and Soil Conservation, 22(4), 187-202.
MLA
Marjan Vahabi Mashhor; Ali Rahimikhoob. "Comparison between neural network and M5 tree models
For reconstructing missing evaporation data of khouzestan". Journal of Water and Soil Conservation, 22, 4, 2016, 187-202.
HARVARD
Vahabi Mashhor, M., Rahimikhoob, A. (2016). 'Comparison between neural network and M5 tree models
For reconstructing missing evaporation data of khouzestan', Journal of Water and Soil Conservation, 22(4), pp. 187-202.
VANCOUVER
Vahabi Mashhor, M., Rahimikhoob, A. Comparison between neural network and M5 tree models
For reconstructing missing evaporation data of khouzestan. Journal of Water and Soil Conservation, 2016; 22(4): 187-202.
)
