Online reviews hold significant influence over customer purchasing decisions, often serving as a deciding factor when selecting a product. A higher proportion of positive reviews tends to drive a surge in sales but this can lead to unethical practices that manipulate consumer trust. These deceptive tactics not only promote unlawful sales growth but also mislead customers into purchasing subpar or unsuitable products. To exploit this phenomenon, spammers frequently fabricate fake reviews, either by creating fraudulent user profiles or by posting repetitive, misleading comments. In this paper, authors present a classification model designed to identify deceptive reviews within a dataset collected from amazon.in. Paper involves an in-depth exploration of data, its preparation, and development of an effective Support Vector Machine (SVM) classifier. By refining various aspects of data and training SVM model. Goal is to achieve a high level of accuracy in detecting fake reviews, ultimately enhancing reliability of online review platforms.