• Thanushree G.P
• Mr. Anil Kumar C
• Halima Sania Samreen
• Prajwal S
• Yashwanth B

A very complicated and fundamental mortgage endorsement system that banks depend on to a great extent for their earnings and profitability is considered the backbone of the banking industry. An advanced statistical approach was used introduced into the process through utilization historical data about past applicants to this system and how they paid off their debt and more efficiency. Three different statistical methods discussed in this paper provide insight into the ways predicting the approval of mortgage applications: Decision Trees and Naïve Bayes. To test these models, we use an openly accessible dataset that contains income, credit history, loan sizes used by potential borrowers among other attributes, with performance being evaluated in forms of accuracy, specificity, prediction accuracy for minority groups and overall efficacy. In relation to our study findings, the boost-based decision tree method outperformed all other approaches generating excellent results on all test data metrics. Finally, we have provided an importance table which identifies some specific features whose presence or absence impact loan approval rate predictions most strongly. This leads us to conclude that boosting based decision tree is a robust predictor of loan approval thereby making it best recommended tools in banking industry position as well as position as a powerful tool for predicting loan approval rates.

Loan approval, Decision tree, Naive bayes, Data sets, Training, Testing, Prediction

"FORECASTING LOAN SUITABILITY WITH MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.b841-b847, April-2024, Available :http://www.jetir.org/papers/JETIR2404195.pdf

"FORECASTING LOAN SUITABILITY WITH MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppb841-b847, April-2024, Available at : http://www.jetir.org/papers/JETIR2404195.pdf