NORMA eResearch @NCI Library

Loan Under writing prediction using Deep learning techniques

Pusapati, Pranavi (2022) Loan Under writing prediction using Deep learning techniques. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


One of the major flaws of the human-based system is that humans are naturally prone to make mistakes. On the other hand, the digital system is far more accurate and efficient than human activities. This will definitely improve the efficiency of the system. Machine learning algorithms detect and detect errors in system information in a very short amount of time compared to the time it takes a person to perform a task. In addition to diagnosing errors in a timely manner, ML algorithms also eliminate the need for third-party operations. In fact, a flawless system means being able to test accurate information, accurate information and highly efficient customer service.

Lenders can incorporate machine learning into existing workflows to take advantage of it without completely redesigning their business processes. Algorithmic errors and the ability to handle lost data will help those lenders securely authenticate previously rejected applicants - applicants. Therefore, ML has great potential to improve efficiency in the customer acquisition process. The algorithm also helps in identifying applicants with high false criteria - successfully turning bad candidates into good people. In the research study exploratory data analysis has been conducted along with data analysis, data visualization, numerical values analysis and checking data point correlation using heatmaps and data visuals. The loan prediction is forecasted using neural network model with and without dropout function for the neural network. The model evaluated using the classification report which include accuracy score, precision value, F1-Score and recall value along with the model losses.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Deep learning model; Loan underwriting system; risk-based credibility analysis; Target Attribute; Dropout function; classification report; accuracy score; precision value; F1-Score; model losses
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
H Social Sciences > HG Finance > Credit. Debt. Loans.
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 01 Mar 2023 11:36
Last Modified: 01 Mar 2023 17:35

Actions (login required)

View Item View Item