Number of unknown and rare diseases are increas-ing day by day. There are only 2000 drugs and more than 35000 known diseases. Existing drugs can be used to treat new diseases.It is known as drugrepositioning.Since drug discovery is a costly and time consuming process computational methods are necessary. Machine learning methods can be applied for drug repositioning. A deep-neural network is trained with drug sim-ilarity , disease similarity and known drug-disease associations. Once trained, the neural network can be used to predict new drug-disease associations. The neural network gives new drug-disease associations with its corresponding probabilities.

Keywords : Machine learning, Bioinformatics, Drug reposi-tioning.