Answers are grounded in this paper and cite their sources.
This study is intended to provide a novel random forest-based strategy for software bug detection and contrast its efficiency to the conventional K-Nearest Neighbor (KNN) technique. The primary purpose is to analyze the accuracy and efficiency of the suggested method in discovering and predicting software defects minimizing the security threats. This study compares the performance of two software bug prediction methods: the K-Nearest Neighbor (KNN) method and a novel random forest-based approach. The publicly available Eclipse Bug Data dataset is used, which contains information on software projects and their associated bug reports, including features such as code complexity, lines of code, and number of developers involved. To calculate the effectiveness of both methods, the dataset is randomly split into two groups of 10 samples each. The KNN method is trained and tested using the first group, while the novel random forest-based approach is trained and tested using the second group. Python and the scikit-learn machine learning package are employed to implement both strategies. To ensure the validity and reliability of the results, the study is conducted with a statistical power (G-power) of 0.85, an alpha level of 0.5, a beta level of 0.2, and a confidence interval of 95%. The experimental findings demonstrated that the suggested random forest-based approach achieved an accuracy of 78.59%, while the KNN method obtained an accuracy of 75.59%. These outcomes indicate that the novel approach outperformed the KNN method, demonstrating its superiority for software bug prediction reducing the security threats. The results were statistically significant values with 0.000 (p<0.05). The analysis reveals that the random forest-based method is preferable for handling high-dimensional data and missing data, both of which are found frequently in software bug prediction.
No takes yet. Share an insight, caveat, or question.
Bharath et al. (2024) studied this question.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: