An Extreme Learning Machine-Based Model for Cryptocurrencies Prediction

Cryptocurrency, in a short period of time has got wide popularity and considered as an investment asset. Prediction of cryptocurrency is a recent area of research interest and budding fast. The price trend of cryptocurrency behaves arbitrarily and fluctuates like other stock markets due to inherent volatility. Though few computational intelligence methods are available, sophisticated methodologies for accurate prediction of cryptocurrency are still lacking and need to be explored. Extreme learning machine (ELM) is a faster and better learning method for neural networks with solitary hidden layer and has enhanced generalization performance. This study proposes an ELM based approach for prediction of four emerging cryptocurrencies such as Litecoin, Ethereum, Ripple, and Bitcoin. The prediction ability of the proposed approach is compared with few similar methods such as RBFN, SVM, MLP, ARIMA, and LSE. From exhaustive simulation studies and comparative result analysis it is found that the ELM method performed better than others and hence can be suggested as an efficient tool for cryptocurrencies prediction.