20240037571. CLASSIFYING FRAUD INSTANCES IN COMPLETED ORDERS simplified abstract (Maplebear Inc.)

From WikiPatents
Jump to navigation Jump to search

CLASSIFYING FRAUD INSTANCES IN COMPLETED ORDERS

Organization Name

Maplebear Inc.

Inventor(s)

Sean Cashin of San Francisco CA (US)

Camille Van Horne of San Francisco CA (US)

Maksim Golivkin of Oakland CA (US)

Benjamin Peyrot of San Francisco CA (US)

CLASSIFYING FRAUD INSTANCES IN COMPLETED ORDERS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240037571 titled 'CLASSIFYING FRAUD INSTANCES IN COMPLETED ORDERS

Simplified Explanation

An online concierge system for fraud detection in customer orders fulfilled by a picker. The system involves a picker client device with a picker application that sends a list of items related to a fulfilled customer order. Additionally, a transaction log containing a list of items purchased is received from the retailer's inventory system. The system compares these lists to identify any unmatched items. A pretrained fraud detection model, trained on historical data with labeled instances of fraud or non-fraud, is applied to the unmatched items to assess the likelihood of fraud. If the likelihood surpasses a predefined threshold, the item is flagged as a fraudulent instance. This determination is then sent to an auditor client device for further action.

  • The system involves an online concierge system for fraud detection in customer orders fulfilled by a picker.
  • A picker client device with an installed picker application is used to send a list of items related to a fulfilled customer order.
  • A transaction log containing a list of items purchased is received from the retailer's inventory system.
  • The system compares the lists of items to identify any unmatched items.
  • A pretrained fraud detection model, trained on historical data with labeled instances of fraud or non-fraud, is applied to the unmatched items.
  • If the likelihood of fraud surpasses a predefined threshold, the item is flagged as a fraudulent instance.
  • The determination of fraud is sent to an auditor client device for further action.

Potential Applications

  • Fraud detection in customer orders fulfilled by a picker.
  • Enhancing security and trust in online retail transactions.

Problems Solved

  • Identifying fraudulent instances in customer orders.
  • Improving the accuracy and efficiency of fraud detection in online retail transactions.

Benefits

  • Reduces the risk of fraudulent transactions.
  • Enhances customer trust and satisfaction in online retail.
  • Improves the overall security of online retail transactions.


Original Abstract Submitted

an online concierge system for fraud detection in customer orders fulfilled by a picker. a picker client device, with an installed picker application, sends a first list of items related to a fulfilled customer order. separately, a transaction log containing a second list of items purchased is received from the retailer's inventory system. these lists are compared to identify any unmatched items. a pretrained fraud detection model, trained on historical data with labeled instances of fraud or non-fraud, is applied to the unmatched items to assess the likelihood of fraud. if this likelihood surpasses a predefined threshold, the item is flagged as a fraudulent instance. this determination is then sent to an auditor client device for further action.

Retrieved from "http://wikipatents.org/index.php?title=20240037571._CLASSIFYING_FRAUD_INSTANCES_IN_COMPLETED_ORDERS_simplified_abstract_(Maplebear_Inc.)&oldid=30809"