Predicting gamefi’s daily market direction

This study presents predicting the direction of the GameFi market on a daily basis. GameFi is a Blockchain product that is a combination of gaming and trading. The customer can make profits from playing, buying, and selling items in-game. Although GameFi is easy to make money, the GameFi market has a high price deviation. As we know, machine learning can predict the future by using historical data or attribute data such as open, high, low, close, volume, and market cap, so many researchers try to create machine learning models to predict the market value of stock, gold, gas, and recently cryptocurrency. The main purpose of this paper is to predict the price direction of GameFi on a daily basis by comparing the performance of SVM and Random Forest on selected features. The experimental result shows that SVM prediction performance result is best at 64.1% with all original features, while the Random Forest performance is best at 85% with some selected features. This study concludes that a suitable model for predicting the direction of the GameFi market on a daily basis is Random Forest.