Prediction of earthquake magnitude using soft computing techniques: ANN and ANFIS

The present investigation aims to find out the performance of the Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) in predicting the earthquake magnitude. More specifically, a comparison is made between ANN and ANFIS, to check their efficiency in predicting earthquake magnitude. For this purpose, 47 data sets of the earthquake are collected from the period of 1906 to 2019. Time of occurrence, latitude, longitude, and focal depth of the earthquake is considered the input variables. Since there is no mathematical and empirical relationship between these variables to predict the earthquake magnitude, the problem is modelled using two different soft computing tool, i.e. ANN and ANFIS. Grid partitioning and subtractive clustering algorithms are employed to develop the model in ANFIS. The result showed that ANN is more efficient in predicting the earthquake magnitude when compared to ANFIS. Adopting this technique in predicting the earthquake magnitude is found to be very fast and economic.