Fraud Protection and Anomaly Detection
Fraudsters use compromised accounts and stolen payment instruments to commit payment fraud. They also attempt other to steal from merchants with account takeover, free-trial abuse, fake product reviews, warranty fraud, refund fraud, reseller fraud, abuse of program discounts, etc. This is a growing issue which are getting visible in most merchants profitability including in Microsoft’s online stores.
Microsoft has many years of experience protecting their own e-commerce businesses from fraud and this expertise are now being put into this tool to benefit Microsofts online stores and all other Merchants. Insights found across all using Fraud Protection will collectively help identify fraud patterns faster and reduce fraud losses. Read more
Request preview here: https://dynamics.microsoft.com/en-us/ai/fraud-protection/
Cognitive Service Anomaly detection
If the full Monty of Dynamics 365 Fraud Protection is too much or to specific for you needs Microsoft Cognitive service just released, in preview, an multi purpose anomaly detection that can run on all time series data. This can easily be added to you app with No machine-learning background required.
There are many types of time-series data—and no one algorithm fits them all. Anomaly Detector looks at your time-series data set and automatically selects the best algorithm from the model gallery to ensure high accuracy for your specific scenario: business incidents, monitoring IoT device traffic, managing fraud, responding to changing markets, etc.
Anomaly detection can be used in two ways either “latest data point in” or in “batch upload” see more in documentation.
Deep dive blog post on Cognitive Service Anomaly detection coming up if a lot of you give this a like or comment.