Enhancing the Identification of Fake Indian Currency Using Ensemble Decision Tree in Comparison with Ridge Linear Classification
R. Kishore Kumar, C. Nelson Kennedy Babu · 2024
The objective of this work is Identification of Fake Indian Currency using Ensemble Decision Tree in Comparison with Ridge Linear Classification for better accuracy. Materials and Methods: For predicting the fake currency, an Ensemble Decision Tree is used in comparison to a Ridge Linear Classification with different training and testing splits. The EDT classifier model, which is suggested as the recommended choice, is trained using 80% of the dataset, while the remaining 20% is reserved for testing purposes. For SPSS analysis, the outcome of two classifiers is categorized as two groups and each group consists of 10 outcome values under different functional operations; finally, it counts 20. During SPSS analysis, the parameters CI of 95% and G power of 0.85 are used. The dataset contains 1,050 unique values related to the currencies and the factors considered are size, color, special attributes. Result: The SPSS analysis performed with the outcome of group 1 and group 2 and achieved mean accuracy of 88.47% and 85.759% respectively. It shows that there is no statistical significance difference between the Novel EDT algorithm and Ridge Linear algorithm with p=0.203 (p>0.05). Conclusion: The suggested approach - Ensemble Decision Tree successfully attained better accuracy in predicting the fake currency in comparison with Ridge Linear Classification.