I will predict credit card fraud and anomaly detection
Data Scientist and AI Pratitioner who loves talking to data!
About this Gig
I will develop a predictive model to alert against credit cards that have a high probability of being fraudulent in the future (next 3 months as compared to the identification date). This gig is valid for one use case at a time, meaning I will work on any specific problem of your choice and deliver that. For other problems, you will need to hire again on the same gig
Technology:
Jupyter Notebook
•
Tableau
Analysis type:
Predictive analysis
Expertise:
Prediction
•
Anomaly detection
•
Statistics
Programming language:
Python
•
SQL
FAQ
Have I worked on Fraud and Risk Analytics?
Yes, I have worked extensively on that. With 6+ years of experience working in the data analytics industry, I have helped payment gateway clients to set up their Risk and Fraud Analytics related tools and predictive models to help them catch fraud at an early stage.
What is the accuracy that can be expected from the predictive model?
Well, this depends on a variety of things, such as the data present, machine capabilities, etc. However, if everything is good, you can expect an accuracy upwards of 25%. Minumum accuracy commitment is 15%.
