Freelancer: nmonnerat
Paylaşın:
Girdiyi Rapor Et

Crypto currency prediction

We may use some API to access the historical data of the cryptocurrency, such as Bitcoin Etherium, etc. As examples of API we have GeckCoin, Cryptocompare, CoinMarketCap. From the API we have access of historical prices, trading volume and market capitalization of the cryptocurrency. With the data we may use some Machine Learn algorithms such as linear regression, deep learnig (nerual network), and implement it with tools such as SciKit-Learn or TensorFlow. We may use part of the data to train the models and part to validate the models and measure their accuracy. We may use some metrics such as mean square error, root mean square error (RMSE), or mean absolute error to measure the accuracy of the models. We may store the predictions in a database for future use. The limit proposed of 60% seems to be feasible, but it is important to emphasize that the accuracy may be higher than 60% without outliers, but it will not be a certainty of precision.

Genel Açıklama Panosu

Henüz mesaj yok.