Scammer Detection in App Reviews Using Comment Analysis aims to improve the credibility of app review systems by identifying fake or fraudulent reviews through NLP and machine learning. Fake reviews negatively impact app trust and rankings, so this project builds an automated model to classify reviews as genuine or scam. Using a labeled Kaggle dataset containing review text, scores, user details, timestamps, and authenticity labels, the system performs text cleaning, sentiment analysis, and behavioral pattern detection. Features such as TF-IDF, linguistic cues, review frequency, and metadata trends are extracted to train models like Logistic Regression, Random Forest, XGBoost, SVM, and LSTM. Model performance is evaluated using accuracy, precision, recall, F1-score, and ROC-AUC, with SHAP used for interpretability. The final output includes a web-based tool for real-time scam detection, insights into scam patterns, and an academic report. This hybrid ML–NLP approach enhances trust and supports developers in filtering fake reviews.