The rapid growth of mobile wallet usage has led to a sharp increase in fraudulent transactions, making fraud detection in portable wallets a pressing concern. Accurately detecting fraud is difficult because transaction data is complicated and unbalanced. Conventional rule-based systems are less flexible and frequently provide large false positive rates along with poor accuracy. Effective feature selection is crucial to the performance of Machine Learning (ML) models, notwithstanding their increased detection rates. Redundancy and noise are introduced by high-dimensional data, which lowers model performance and raises computing costs. The advantages of hybrid feature selection are frequently overlooked in current research, particularly when it comes to portable wallet fraud detection. By combining Random Forest Importance, LASSO Regression, Recursive Feature Elimination (RFE), and Mutual Information (MI) with resampling to solve class imbalance, this study fills that gap. Our approach provides a more reliable and effective solution for safe portable wallet fraud detection by removing superfluous features, increasing accuracy, and reducing computing cost. The model becomes faster and more effective when superfluous characteristics are eliminated because this reduces the computational effort. By concentrating just on the most instructive data, it increases accuracy. By addressing class imbalance and combining several selection strategies, the hybrid approach guarantees robustness. All things considered, this leads to a scalable and safe fraud detection system for transactions using mobile wallets. Our results show that a successful feature selection approach improves fraud detection accuracy, which in turn improves operational effectiveness and financial security. 

The rapid rise of mobile technology paired with the steady growth of the internet, has led to a massive increase in the amount of user generated content, such as online consumer reviews, accessible through the browser. As the volume of user-generated content continues to rise, it becomes increasingly important to develop sophisticated methods for performing sentiment analysis on the texts collected from users, especially those that have been generated in relation to restaurants and similar types of service establishments. In this paper, we will present a new approach to sentiment analysis which incorporates Latent Dirichlet Allocation topic models, Term Frequency- Inverse Document Frequency vector representations and XGBoost Classifiers into a unified framework. Unlike conventional implementations, this study integrates probabilistic topic distributions from LDA with multi-level n-gram TF-IDF features and evaluates their combined impact using XGBoost for enhanced classification performance. Using three distinct n-gram levels (unigrams, bigrams, and trigrams), we will evaluate various aspects of text-based data including common linguistic patterns and sentiment trends. Higher-order n-grams were included to capture contextual dependencies beyond single-word features. Overall, our results demonstrate that the performance of our proposed framework is superior to traditional corpus-based models on multiple evaluation metrics, including: classification accuracy 96.07%, classification sensitivity 95.43%, classification specificity 97.12% and F1-Score 96.16%.