Fine-Tuning Fake Review Detection Models with Genetic Algorithms

Online customer reviews have a strong effect on decisions in all aspects (other customers or business management). Fake reviews could mislead these decisions and thus negatively affect business profit. Machine learning has been used to detect fake reviews; however, most machine learning models need parameter tuning. This paper investigates the effectiveness of genetic algorithms (GA) in improving false review detection (FRD) techniques. The methodology involves applying GA to optimize the hyper-parameters of multiple machine learning classifiers, including Random Forest (RF), Decision Trees (DT), k-Nearest Neighbors (KNN), Support Vector Machines (SVM), and eXtreme Gradient Boosting (XGBoost). Through an iterative evaluation of different combinations of parameters, the study identifies the optimal configurations that maximize performance metrics. The key findings indicate that the RF and GA approach achieved the highest accuracy, outperforming other models. The results highlight the critical role of hyperparameter optimization in improving the detection of fake reviews. Performance evaluation was done using different metrics such as accuracy, precision, recall, and F1-score.