Machine Learning based Attack on Certain Encryption Schemes

Deep learning provides a very promising approach to attack cryptographic implementations. In this paper, a deep learning-based attack on text encrypted with RC4 Cipher is shown. Deep Learning algorithms, a popular approach to deal with complex problems are utilized to learn the structure of the text from training examples so that an unknown similar text can be decrypted successfully. The attack is performed on a section of the ENG-FRENCH dataset. The results of deep learning algorithms are 74.83% on seen words or similar words but don’t give good accuracy on unseen or unfamiliar words which it has never seen in training data.