Spotting Fake Reviews: How Our Platform Identified Repetitive Language, Extreme Ratings & Suspicious Timing