Internet venues and online reviews have spread from specialized trade journals to end consumers. In addition to affecting local economies, this development has also had an impact on the cultural legacies of the nations. Deep neural networks with transformer-style topologies are frequently used in these models. Reshuffling can be easily understood by language models for further investigation and operations; the foundation for spotting deceptive reviews was then developed using two broadly used language models, BERT and ELECTRA. The performance attained is impressive, with an exactness and F1 score of close to 95%. This was obtained by segregating the dataset into training and testing sets, with the distribution of 80% for training, 20% for testing, 90% for training, and 10% for testing.