The case study is the classification of fraudulent and legit websites based on text mining technique. Several attempts are done by the candidate with the purpose to maximize the integrity of the classifier. The case-study is divided in shallow and deep learning approaches for the classification task. In the shallow learning approach, the term's frequency, Cialdini's principles and sentiment analysis features are extracted from the corpus text and are submitted to the classifiers. A different approach is the using the spam email dataset as training set such that it is possible to find a relationship between scam and spam case. In deep learning approach, email dataset is used for the training phase and the websites' dataset is used for the transfer learning technique, on neural networks architecture with different layers..