Towards An Efficient Real-time Approach To Loan Credit Approval Using Deep Learning

The last decade has seen an important rise of data gathering, especially in the financial sectors. Banks are indeed one of the biggest producers of big data, as a matter of fact no other company than the bank has so much data gathered on its customers. Gathering and analyzing this data is a key feature for decision making, particularly in banking sector. One of the most important and frequent decision banks has to make, is loan approval. The challenge is to know how to build a proactive, powerful, responsible and ethical exploitation of personal data, to make loan applicant proposals more relevant and personalized. Machine learning is a promising solution to deal with this problem. Therefor, in the last years, many algorithms based on machine learning have been proposed to solve loan approval issue. However, these algorithms have not taken into consideration Real-time paradigm during processing. In this paper, we propose a Real-Time Binary classification model to deal with loan approval. Our proposed model is based on a deep neural network, and it permits to classify loan applicant as good or bad risk. Experimental results prove that our proposed Real-Time model, based on deep neural network, outperforms typical binary classifiers, in terms of precision recall and accuracy.