One of the most popular investment assets nowadays is Bitcoin. The financial market volatility of bitcoin's price has drawn the attention of researchers and investors alike to the ways in which its price fluctuates. The paper proposes an evaluation of Short Term Financial Marketing Bitcoin Prediction: A Comparative Analysis of Large-Kernel Attention Graph Convolutional Networks across Various Prediction Horizons (BP-ST-LAGCN-NOA) for short-term Bitcoin market prediction. Then, pre-processed data are fed to the Large-kernel Attention Graph Convolutional Network (LAGCN) to effectively predict the bitcoin market in the short term. LAGCN does not express adaptive optimization strategies to determine optimal factors to effectively predict the Bitcoin market. Hence, the Nutcracker Optimization Algorithm (NOA) is employed to optimize the weight parameter of the Large-kernel Attention Graph Convolutional Network to predict the bitcoin market. Then the proposed BP-ST-LAGCN-NOA is implemented in Python, and the performance metrics like Accuracy, Precision, Recall, Specificity, F1-score, and ROC are analysed. The BP-ST-LAGCN-NOA model achieves 99.64 % accuracy, 99.21 % precision, 99.18 % recall, 98.14 % F1-Score and 98.47 % specificity, outperforming all baseline methods. The BP-ST-LAGCN-NOA model demonstrates superior accuracy and robustness in short-term Bitcoin market prediction, outperforming existing machine learning (ML) and deep learning (DL) approaches.
扫码关注我们
求助内容:
应助结果提醒方式:
