Building an end to end Machine Learning Model to identify a fraudulent transaction in Real-time and Batch modes for UPI.Implementation of traditional and Neural Networks models based on diversified Modus Operandis(MOs).

I was working as a Data Science Intern, we a team of 5 interns working on a SVM model, with different kernels. The project was to make the web application for Intraday Stock Price Prediction on two real-time datasets using Machine Learning. We have implemented multiple kernels to achieve higher accuracy. The algorithm that we used was SVM(SVR) of the polynomial kernel and degree 4.

‘Code In Place’ is a worldwide program and a 100% fee waived offered by “Stanford University” once a year. A limited number of students are selected after verifying an application and the performance in the tasks provided. The program is about one and a half months long and committing a minimum of 10 hours per week and with a guide allotted for a group of 8-10 students from several countries randomly, to have a more personalized learning experience. The program includes lecture videos for every week, an hour long live session with the group members, and a guide allotted, followed by assignments.