Enhancing Gram-Based Fuzzy Keyword Search in Encrypted Data Using XGBoost Comparison to SVM for Increased Accuracy

This study sets out to explore and compare two widely used techniques, Support Vector Machines (SVM) and Extreme Gradient Boosting (XGBoost), in the context of Gram-Based Fuzzy Keyword Search applied to encrypted data. By leveraging a large dataset consisting of 21,875 samples, the research aims to boost the accuracy of this search method through various training and testing splits. With an average GPower of nearly 85% (with settings of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\alpha=0.05$</tex> and power <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$=0.85$</tex>), the study analyzes performance at two significance levels: <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\alpha$</tex> of 0.05 and 0.85. The findings revealed a <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{p}$</tex>-value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$p=0.664(p&gt;0.05)$</tex>, which suggests that there isn't a statistically significant difference between the two approaches. However, when it comes to accuracy, XGBoost takes the lead, achieving an impressive accuracy rate of 90 %, while SVM lags slightly behind at 87 %. In essence, when comparing Gram-Based Fuzzy Keyword Search using XGBoost with the same search using SVM, XGBoost clearly demonstrates its superiority with a 90 % recognition rate. This research not only highlights the effectiveness of XGBoost over SVM but also contributes to the broader goal of enhancing the accuracy of Gram-Based Fuzzy Keyword Search in the context of encrypted data.