Real-Time Bitcoin Cost Identification to Improve Efficiency Using Lasso Regression in Comparison with Decision Tree

Enhancing the Bitcoin cost prediction to improve accuracy using a decision tree in comparison with Lasso regression. For trend confirmations, we often utilize the financial phrases support and resistance to the prediction. Materials and Methods: The price of bitcoin is forecasted using a straightforward prediction technique called Lasso regression. The effectiveness of each model's performance on the task was evaluated, and the findings were examined. The goal of testing so many different models was to examine how the fundamental hypotheses of each of them varied. We used Lasso regression together with the best parameters to weigh the input variables because it is a reliable technique. Lasso regression is a Model based on a sample T-test with a value of p is 0.000 (Independent sample t-test
$\mathrm{p} < 0.05$
). This indicates that there is a statistically significant difference between the two algorithms. The sample size calculation is done with a G-power pretest of 80%, threshold 0.05% and CI - 95%. Result: This research study indicated that using Novel Lasso Regression compared over Decision Tree with improved accuracy is 93.06% shows a better accuracy in the prediction of Bitcoin. Conclusion: Using Lasso regression compared over Decision Tree with improved accuracy.