AI and data driven results have been applied to different fields and achieved outperforming and promising results. In this exploration work we apply k- Nearest Neighbors, eXtreme Gradient Boosting and Random Forest classifiers for detecting the trend problem of three cryptocurrency requests. We use these classifiers to design a strategy to trade in those requests. Our input data in the trials include price data with and without specialized pointers in separate tests to see the effect of using them. Our test results on unseen data are veritably promising and show a great eventuality for this approach in helping investors with an expert system to exploit the request and gain profit. Our loftiest profit factor for an unseen 66 day span is1.60. We also bandy limitations of these approaches and their implicit impact on Effective request thesis.
Download the below PDF file. Download Profitable Strategy Design for Trades