Published:
Some organizations are at an early stage of data collection while others have large teams of analysts who are continuously blocking fraudsters. In this post, we will discuss different ML frameworks that can be implemented depending on the stage of the organization. 
Published:
In this post, I will discuss the basic foundations that companies need to be ready with in order to build their first fraud detection system. Over the next few weeks, I will build upon these foundations and add layers of engineering and scientific systems that will help create the most advanced fraud detection systems. 
Published:
With constant changes in how we interact with external environment (e.g. how we shop, what media channels we consume, how we do banking, how we order food etc.), the types of fraud, scams and abuses are also emerging. People with malicious intent are always on the lookout for finding loopholes in systems and using those loopholes for personal gain. In this article, I discuss about a few types of common fraudulent, abusive and other malicious activities that individuals and organizations experience. 
Published:
Authors: Prince Grover, Zheng Li, Jianbo Liu, Jakub Zablocki, Hao Zhou, Julia Xu, Anqi Cheng. The Fraud Dataset Benchmark (FDB) is a compilation of publicly available datasets relevant to fraud detection. The FDB aims to cover a wide variety of fraud detection tasks, ranging from card not present transaction fraud, bot attacks, malicious traffic, loan risk and content moderation. The goal is to provide researchers working in the field of fraud and abuse detection a standardized set of benchmarking datasets and evaluation tools for their experiments.