Fraud Detection
[wtm_use_case_cats]
A fraud detection system helps companies to decrease unidentified transactions and increase company revenue and brand value. Many different forms of fraud exists in E-commerce businesses such as Interception Fraud, Chargeback Fraud, Payment Fraud, Promotion Abuse or Sign-up Fraud.We can detect suspicious behavior by using data science techniques like matching algorithms, anomalies detection, classification and data mining. Signs of suspicious behavior could include a shipping address differing from a billing address, an unexpected international order or multiple orders of the same item.
Benefits for the company
Fraud has never been a new thing. It is especially obvious now, when the world is moving away from in-store purchases and the number of cash-free transaction increases
As the trend for E-commerce fraud rises , E-commerce fraud scenarios and malware become more subtle and harder to detect. E-commerce fraud protection has never been so important.
Feasability
High
Type of expertise/ AI domain
Hidden Markov Model (HMM), Bayesian Neural Network (BNN), Fuzzy Logic
Internal data required
Anomalies Detection, Data mining, Classification Analysis
With E-commerce sales estimated to reach $630 billion (or more) in 2020, an estimated $16 billion will be lost because of fraud