➗Basic algorithm

As mentioned earlier, this report introduces a novel approach to determine the movement of the cryptocurrency market. The model is optimized through different parameter settings as mentioned in Table 2. The model discussed in this report is designed to predict the closing price of Solana on the Spot Market (SOL/USDT). The algorithms used employ CNN architecture to predict the market direction with categorical target variables, while the model in this article has a continuous target variable. The models used in the reports by Hoseinzade E, Haratizadeh S (2019) and Kara Y, AcarBoyacioglu M, Baykan ÖK (2011) were employed to evaluate the model. To address the issue of prediction variables, our model predictions are also transformed into categorical variables using the method discussed in the report by Hoseinzade E, Haratizadeh S (2019):

C(t): closing price of “th candle"