Introduction The “ALPHA” application is a Machine-Learning based desktop application which provides an easier route for bank employees to accept and process any loan application. Which is usually a tedious ordeal where the details of the applicant must be entered manually on to the bank database by the employees which merely involves typing out the information on the application forms. Bank employees can scan the loan application of a customer, instead of manually entering the data onto their computer screen, save it in an image format, like .png, .jpeg, etc. Then the OCR-Optical Character Recognition can be used to populate the text boxes that were to be entered from the information on the application-form. The officer can change any content if there are any errors and then upload the information. The uploaded information gets stored in a database, which may be retrieved by the authorized personnel to scrutinize the validity of the information. The data is then loaded into a Machine-Learning algorithm that predicts the worthiness of the applicant to avail the loan. The Manager then can approve or reject the loan application with the help of the prediction made by the algorithm. This application proved to be user-friendly, users with limited computer knowledge can use this application without any discrepancies. It avoids the tedious process of data entry and makes the job easier for the bank employees.
Software requirements
The software tools used for designing and developing the application includes: ● HTML, CSS, JavaScript for web design ● MySQL and MongoDB/NOSQL for database management. ● Flask as framework for Python. ● Tesseract. ● OpenCV2 for Image Processing.
ScreenShots
Conclusions
The ALPHA Loan Processing is a useful tool for both employee and manager in a bank. The application benefits the bank as well as the customers as the processing of the application is not delayed and avoids unnecessary overheads while processing loan requests. The Application provides a solution to the tedious task of the data entry of the customers through OCR and assists the management in deciding whether a loan should be granted. The application comes with the added benefit that it is easy to use and anybody with minimal computer knowledge will be able to use the application without any hassle. It also provides an automatic email system where an email is sent to a customer as soon as the manager clicks on the approve or reject button. Overall all features that were planned were implemented successfully in the application and is working as anticipated.