Mean_Monthly_Met_Data_KMD

This research seeks to resolve the problems associated with inaccurate weather prediction in North Eastern Kenya. The economic cost of incorrect weather prediction is profound, especially for agriculture, energy, tourism, and transportation among other sectors. The research will improve the accuracy of long-range weather predictions by adopting the Radial Basis Function Network (R B F N) which was recommended by previous research. The RBFN is efficient in handling nonlinear data, computationally effective, and quick to learn. The study employed a mixed-methods research design. A descriptive research approach was used to get an in-depth understanding of the weather patterns and to facilitate the usage of historical meteorological data for quantitative analysis. The research used archived historical meteorological data from the Kenya Meteorological Department (2003-2013). Written requests to the Director, of Kenya Meteorological Department, were made. The weather data was used to fit the model of the designed system to create a set of patterns that can be used to make predictions. The raw features of the dataset shall be cleansed, pre-processed, and normalized whose features in the feature vectors were reduced to remove the noise and consequently improve the performance of the RBFN model. Experiments were conducted to determine the effect of different parameters such as the number of concealed neurons, amount of training, and input factors on the level of correctness of the strength of rainfall prediction. The study also explored the best possible combination of meteorological variables as predictor inputs, minimal training data requirements, and the determination of the number of hidden neurons which provided an efficient model. The expected outcomes of the research include a well-developed RBFN-based weather forecasting system for long-range predictions of the weather in North Eastern Kenya. The results are ultimately expected to provide valuable input in decision-making of economic sectors, disaster management, climate mitigations, etc., and thus contribute to labor from the above participant territories and economic development in the region