An AI-Driven Method for Detecting Fake Reviews through Feature Engineering

This research aims to develop a model capable of distinguishing between genuine and fraudulent reviews, thereby assisting customers in avoiding online scams. Businesses also stand to gain, as enhanced trust can lead to increased sales. The study focuses on refining the prediction system for identifying fake reviews by utilizing real-time datasets from Amazon to train the model. Various machine learning algorithms, including Random Forest, AdaBoost, and Naïve Bayes, will be employed for classification purposes. The effectiveness of each algorithm will be evaluated using a confusion matrix. A detection process will be implemented to ascertain the authenticity of reviews through feature engineering. By leveraging Natural Language Processing (NLP) to extract significant features from the text, the research will facilitate the detection of review spam.