Fraud Analytics Using Descriptive, Predictive, and Social Network Techniques: A Guide to Data Science for Fraud Detection

by Bart Baesens, Veronique Van Vlasselaer, Wouter Verbeke

Detect fraud earlier to mitigate loss and prevent cascading damage Fraud Analytics Using Descriptive, Predictive, and Social Network Techniques is an authoritative guidebook for setting up a comprehensive fraud detection analytics solution. Early detection is a key factor in mitigating fraud damage, but it involves more specialized techniques than detecting fraud at the more advanced stages. This invaluable guide details both the theory and technical aspects of these techniques, and provides expert insight into streamlining implementation. Coverage includes data gathering, preprocessing, model building, and post-implementation, with comprehensive guidance on various learning techniques and the data types utilized by each. These techniques are effective for fraud detection across industry boundaries, including applications in insurance fraud, credit card fraud, anti-money laundering, healthcare fraud, telecommunications fraud, click fraud, tax evasion, and more, giving you a highly practical framework for fraud prevention.

It is estimated that a typical organization loses about 5% of its revenue to fraud every year. More effective fraud detection is possible, and this book describes the various analytical techniques your organization must implement to put a stop to the revenue leak.

Examine fraud patterns in historical data Utilize labeled, unlabeled, and networked data Detect fraud before the damage cascades Reduce losses, increase recovery, and tighten security The longer fraud is allowed to go on, the more harm it causes. It expands exponentially, sending ripples of damage throughout the organization, and becomes more and more complex to track, stop, and reverse. Fraud prevention relies on early and effective fraud detection, enabled by the techniques discussed here. Fraud Analytics Using Descriptive, Predictive, and Social Network Techniques helps you stop fraud in its tracks, and eliminate the opportunities for future occurrence.

382 pages, Kindle Edition

First published July 27, 2015

Reviews

[David · Rating 4 out of 5]

overall: this book is informative. the book comprises lots of stats technique to detect fraud from data.

[Lāsma Supe · Rating 5 out of 5]

The book is very well written and handful insight material for data scientists to understand business side behind business requirements on data modeling as well brilliant easy language material for business managers on advanced data science techniques including predictive data modeling and social network analysis.

[Altaix · Rating 2 out of 5]

This books feels more like a reference than a learning resource. Many different approaches are discussed, and almost all algorithmic ones could benefit from more business examples and less mathematical rigor.

[Jack Coates · Rating 4 out of 5]

Good book on fraud analytics

Several good ideas, but a bit heavy on the implementation specific math formulas for me. Still I enjoyed it and found it useful.

[Montgomery · Rating 4 out of 5]

Solid overview, but you need at least intermediate college-level math skills to interpret many of the concepts.

[~ · Rating 2 out of 5]

Decent survey for those unfamiliar with basic statistics and machine learning. Fraud detection, at least based on the descriptions here, is a straightforward classification problem. But the authors overlook an important consideration: fraud is an extremely rare event, but there is very discussion of the problems that highly imbalanced data present when using standard classification algorithms (the intro says this problem is covered in the fourth chapter but it's not really). It seems this would be extremely important to cover in great detail in a book specifically on fraud detection. Perhaps they wanted to keep the book more accessible, but this makes me wonder what we're supposed to get from this book that isn't covered more extensively in the dozens, if not hundreds, of other introductory 'data science' books (again considering that there's not much more domain knowledge in the book than you could quickly get from a Google search). Gold standard is the 'Elements of Statistical Learning' books, which come in varying levels of difficulty. I would recommend reading those and looking for more detailed texts on any of the subjects they don't cover (e.g. statistical network analysis).