Hybrid-Deep-Learning-model

Thus is a deep learning's model in which the purpose was to build a model through which we can cluster the potential customers fraud detector according the mean internal neuron distance of each observation in the dataset Implementation of SOM : Self Organizing Map to cluster the potential fraud detectors i.e = those who got the approval of the bank and those who did not the approval and then comes the implementation of Supervised Deep learning Implementation Of ANN : to predict the probability of fraud for each customer