The Multi Layer Auto Encoder Neural Network (ML-AENN) for Encryption and Decryption of Text Message
Achmad Fanany Onnilita Gaffar, Arief Bramanto Wicaksono Putra, Rheo Malani · 2019
Efficient key generation techniques require highly secure cryptosystems. The traditional key generation technique is very systematic that it is easy to attack. The Deep Learning algorithm is one of the research paths into the automated extraction of complex data representations (features) at a high level of abstraction. Auto Encoder Neural Network (AENN), one of the deep learning's architectures, play an essential role in unsupervised learning in deep architecture for learning transfer and other tasks. The AENN are simple learning networks that aim to convert inputs into outputs with the least amount of distortion. This study proposes the Deep Learning approach for text message encryption and decryption by using the Multi-Layer AENN where the outputs of each of the deepest layer as the secret-key generator and hash value generator. The result of this research showed that the proposed method has a high degree of confidentiality because each training always produces a different secret key. Furthermore, the tampered data/information can be seen from the presence of different hash values that occur.