AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Machine Learning Models for Fraud Detection
The overall impact of fraud involved is around like five trillion dollars. And machine learning models are kind of ideal to fraud detection. The data set is really by unbalanced what I mean by that is that let's say you have 1000 observation. In those 1000 observation you will just see around 10 or 12 fraud and rest all will be not fraud. So it will be very difficult to kind of train a good model a model will give you high accuracy but it might not be able to model those fraud cases quite well.