About the Client

The client is a payment services provider  

Business Challenges

  • The client was processing more than a half a million transaction per day 
  • Some of these transactions were fraudulent as reported by the card user
  • The mandate was to predict such transaction and block them 
  • The prediction had to work in a fraction of a second, so that the user experience would not be compromised. 
  • As the number of fraudulent transactions were very low (0.004%), it was very hard to arrive at a generalized solution 


Our solution was an AI based model that provided a confidence score for each transaction. A low score would mean that the transaction is suspicious. The solution included the following capabilities 

  • Sampling of data to reduce the data skew 
  • Feature engineering to add more decision parameters for the AI model 
  • Synthetic data extrapolation 
  • AI based transaction scoring model 


  • The client used this model along with a combination of other models that they had to make decisions on the card transactions