Improving the Accuracy of Rainfall Prediction using Optimized LSTM Model
S. Geetha 

S. Geetha, Assistant Professor (Sr. Grade), Department of Computer Applications, Mepco Schlenk Engineering College Autonomous, Sivakasi (Tamil Nadu), India.
Manuscript received on 28 March 2019 | Revised Manuscript received on 09 April 2019 | Manuscript Published on 18 April 2019 | PP: 947-950 | Volume-7 Issue-6S March 2019 | Retrieval Number: F03930376S19/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Prediction of rainfall is too complex and also it depends on many meteorological factors. India is flourishing country in agriculture. In earlier days, the rainfall is predicted even by common man in the village and they gone for farming. Now a day, due to drastic changes in the climate and weather, accurate prediction of rainfall becomes multifaceted. Information Technology is offering very good techniques for prediction. Deep Learning is one of the latest technology applied in the field of prediction. As the rainfall data is time-series data, Optimized LSTM (OptLSTM) is proposed in this paper. The data is collected from the various districts in Tamil Nadu and used for developing prediction model using LSTM and optimized the hyper parameters with Particle Swarm Optimization (PSO). Series of experiments are conducted to authenticate the proposed model is predicting accurately. The accuracy of the model is evaluated with evaluation measures MSE, RMSE, MAE. The performance of OptLSTM model is compared with other conventional models used for rainfall prediction. Out of those, OptLSTM presents better accuracy.
Keywords: LSTM, Rainfall, PSO, Evaluation Metrics, Prediction.
Scope of the Article: Regression and Prediction