Rainfall Prediction Using Data Mining Techniques - A Survey

Rainfall forecasting takes an important role in India not only for agriculture but also for minimising damages from floods, to reduce water scarcity and for efficient water resource management.There are various machine learning techniques used for Rainfall Forecasting. Among these techniques, Artificial Neural Networks (ANN) is the most widely used Machine Learning Technique for Rainfall Forecasting. Purpose of this paper is to identify and analyse the trends of Neural Network Techniques in Rainfall Forecasting.

In India, agriculture is the most important factor for survival of human being. For agriculture, the most important thing is water. i.e., rainfall. Nowadays rainfall prediction is a major problem. Predicting the amount of rainfall gives alertness to farmers by knowing early so that they can protect their crops and properties from rain. There are more techniques to predict the rainfall. The ML algorithms are best suited for prediction of rainfall. Here are some of the major ML algorithms used rapidly which are Auto regressive integrated moving average Model (ARIMA), Artificial neural network (ANN), Support Vector Machine, Logistic regression, and Self organizing map. And there are two models commonly used to predict periodic rainfall such as Linear and Non-linear models. ARIMA Model is the first used model. Although using ANN (Artificial neural network) the prediction of rain can easily completed by using Cascade NN, Layer recurrent network, or Back propagation NN. Artificial NN is similar as Biological neural networks.

Environmental Economics, 2016

The growing severe damage and sustained nature of the recent drought in some parts of the globe have resulted in the need to conduct studies relating to rainfall forecasting and effective integrated water resources management. This research examines and analyzes the use and ability of artificial neural networks (ANNs) in forecasting future trends of rainfall indices for Mkomazi Basin, South Africa. The approach used the theory of back propagation neural networks, after which a model was developed to predict the future rainfall occurrence using an environmental fed variable for closing up. Once this was accomplished, the ANNs’ accuracy was compared against a traditional forecasting method called multiple linear regression. The probability of an accurate forecast was calculated using conditional probabilities for the two models. Given the accuracy of the forecast, the benefits of the ANNs as a vital tool for decision makers in mitigating drought related concerns was enunciated. Keywor...

2021

Modeling of rainfall is the most challenging aspect of the hydrological processes because of the intricacy of atmospheric dynamics. In a country like India, where agriculture is the main source of income, long-term rainfall forecasting is critical. For modeling and analyzing rainfall data, the location of Surat (Gujarat) was chosen, and using an Artificial Neural Network (ANN) with various meteorological indices, distinct properties of the hidden neurons in the system were studied. For this analysis, 39 years of data (from 1981 to 2019) were used. The data was utilized for training in 75% of the cases, validation in 15% of the cases, and testing in 15% of the cases. The models were discovered to be capable of accurately predicting rainfall. In this study, it was observed that using Neural Network Tool (NNTOOL) to implement the algorithm achieved the best results when compared to Neural Fitting Tool (NFTOOL) i.e., a multilayer algorithm outperforms a single layer algorithm in terms o...

International Journal of Advanced Computer Science and Applications, 2019

Multilayer perceptron neural network (MLPNN) is considered as one of the most efficient forecasting techniques which can be implemented for the prediction of weather occurrence. As with any machine learning implementation, the challenge on the utilization of MLPNN in rainfall forecasting lies in the development and evaluation of MLPNN models which delivers optimal forecasting performance. This research conducted performance analysis of MLPNN models through data preparation, model designing, and model evaluation in order to determine which parameters are the best-fit configurations for MLPNN model implementation in rainfall forecasting. During rainfall data preparation, imputation process and spatial correlation evaluation of weather variables from various weather stations showed that the geographical location of the chosen weather stations did not have a direct correlation between stations with respect to rainfall behavior leading to the decision of utilizing the weather station having the most complete weather data to be fed in the MLPNN. By conducting performance analysis of MLPNN models with different combinations of training algorithms, activation functions, learning rate, and momentum, it was found out that MLPNN model having 100 hidden neurons with Scaled Conjugate Gradient training algorithm and Sigmoid activation function delivered the lowest RMSE of 0.031537 while another MLPNN model having the same number of hidden neurons, the same activation function but Resilient Propagation as training algorithm had the lowest MAE of 0.0209. The results of this research showed that performance analysis of MLPNN models is a crucial process in model implementation of MLPNN for week-ahead rainfall forecasting.

INTERNATIONAL JOURNAL OF RESEARCH AND ANALYTICAL REVIEWS, 2019

An accurate rainfall forecasting is a challenging problem for agriculture dependent countries like India for analyzing the crop productivity, use of water resources and pre-planning of water resources. Statistical techniques for rainfall forecasting cannot perform well for long-term rainfall forecasting due to the dynamic nature of climate phenomena. The daily rainfall dataset for Navsari and Anand regions are obtained and converted to monthly and yearly datasets for experimentations. Auto Regressive Integrated Moving Average (ARIMA), Feed Forward Neural Network (FFNN), Radial Basis Function Neural Network (RBFNN), and Time Delay Neural Network (TDNN) techniques are used for rainfall prediction. These techniques use input parameters such as evaporation, wind speed, maximum temperature, minimum temperature, and relative humidity for rainfall prediction. The proposed approach uses Genetic Algorithm (GA) for optimizing biases and weights of neural network. The results of ARIMA model are compared with the results obtained using three neural network models. From results, it is found that the neural network techniques performed well compared to the ARIMA model. The GA-TDNN is trained with Levenberg Marquardt Back Propagation algorithm and gives better performance than other NN models. The GA-TDNN model achieved monthly rainfall prediction accuracy of 89.83% and 81.38% for Anand and Navsari regions, respectively. In yearly prediction, the model achieved accuracy of 87.87%, and 87.86% for Anand and Navsari respectively.

European Journal of Electrical Engineering and Computer Science

Rainfall prediction is an important meteorological problem that can greatly affect humanity in areas such as agriculture production, flooding, drought, and sustainable management of water resources. The dynamic and nonlinear nature of the climatic conditions have made it impossible for traditional techniques to yield satisfactory accuracy for rainfall prediction. As a result of the sophistication of climatic processes that produced rainfall, using quantitative techniques to predict rainfall is a very cumbersome task. The paper proposed four non-linear techniques such as Artificial Neural Networks (ANN) for rainfall prediction. ANN has the capacity to map different input and output patterns. The Feed Forward Neural Network (FFNN), Cascade Forward Neural Network (CFNN), Recurrent Neural Network (RNN), and Elman Neural Network (ENN) were used to predict rainfall. The dataset used for this work contains some meteorological variables such as temperature, wind speed, humidity, rainfall, v...

Jurnal Informatika

Changing climatic patterns are caused by changes in variables, such as rainfall and air temperature that occur continuously in the long term. Rainfall itself is influenced by several weather factors such as air humidity, wind speed, air pressure, and temperature. This study experimented to test a combination of 9 additional weather parameters such as dew point, wind gusts, cloud cover, humidity, rainfall, air pressure, air temperature, wind direction, and wind speed to predict daily rainfall for one year using the main parameters of the rainfall time series. Prediction is done using Artificial Neural Network (ANN). The ANN architecture used is to use 3 to 11 input parameters, 1 hidden layer totaling 60 neurons with the ReLu activation function, and 1 neuron in the output layer without an activation function. ANN without additional weather parameters obtained an MSE of 0.01654, while prediction using additional weather parameters obtained an MSE of 0.00884. So the combination of rainfall time series parameters with additional weather parameters is proven to provide a smaller MSE value KEYWORDS ANN Rainfall forecasting ReLU activation function This is an open-access article under the CC-BY-SA license

International Journal of Innovative Research in Computer Science & Technology

Prediction of rainfall for a river basin is of utmost importance for planning and design of irrigation and drainage systems as also for command area development. Since the distribution of rainfall varies over space and time, it is required to analyze the data covering long periods and recorded at various locations to arrive at reliable information for decision support. Further, such data need to be analyzed in different ways, depending on the issue under consideration. In the present study, Extreme Value Type-1 (EV1) distribution based on statistical approach and Multi Layer Perceptron (MLP) network based on Artificial Neural Network (ANN) is adopted for prediction of rainfall at Fatehabad and Hansi. The performance of the statistical and ANN approaches used in rainfall predication are evaluated by model performance indicators viz., correlation coefficient, model efficiency and mean absolute percentage error. The study shows the MLP is found to be better suited network for prediction of rainfall at Fatehabad whereas EV1 for Hansi.

Procedia Computer Science, 2019

Accurate rainfall forecasting plays a significant role for weather stations as it serves to warn people about incoming natural disasters. This paper presents an implementation of week-ahead rainfall forecast that utilizes Multilayer Perceptron Neural Network (MLPNN) in processing historical rainfall data. Proper data preparation, model implementation and performance evaluation were conducted to two MLPNN models which yields promising results in predicting week-ahead rainfall. The MLPNN architecture was a supervised feed-forward neural network having 11 input neurons consisting of different weather variables along with various hidden neurons and 7 output neurons representing the week-ahead forecast. The MLPNN models which were SCG-Tangent and SCG-Sigmoid, produced a MAE of 0.01297 and 0.1388 and RMSE of 0.01512 and 0.01557, respectively. This viable implementation of MLPNN in rainfall forecasting hopes to provide organizations and individuals with lead-time for the strategic and tactical planning of activities and courses of action related to rainfall.