As we know, today there have been several variations in the price of Bitcoin as it is not stable, that is to say, the price of Bitcoin rises and falls exaggeratedly as there are several factors that focus on the opening and closing price values of Bitcoin, also the highest and lowest Bitcoin price values reached in a day. Bitcoin price is determined by supply and demand. If user demand for this cryptocurrency is high, the price will increase and if it is very low, will go to. The price of Bitcoin has fluctuated considerably in recent years. This means that it is constantly changing and is in a stable range without major changes. In this paper, a machine learning method, also called Machine Learning, is proposed, which will allow us to automatically search and interpret relevant information from a large amount of data. Machine learning is the branch of artificial intelligence science that creates automated learning systems. In the case study, the Recurrent Neural Network (LSTM) algorithm will be used for the predictive analysis of Bitcoin value. Recurrent Neural Networks are network layers for analyzing time series and time-series data. An LSTM recurrent neural network aims to learn long-term dependencies; that is to say, learn the dependencies of future values of a sequence on previous values. The results of this research work showed that, by applying the Machine Learning technique and the LSTM algorithm, it was possible to predict the Bitcoin price increase. These results could benefit different companies or financial institutions in the investment of money with the help of Bitcoin. Companies such as Microsoft, Destinia, WordPress, among many others, already allow purchases with bitcoins, or other cryptocurrencies, on their websites.