Order Brushing Detection

Overview

Order Brushing Detection is a data analysis project created as a solution to Week 1 of the Shopee Code League.

The challenge focused on identifying abnormal ordering behaviour in e-commerce data that could indicate order brushing — a practice where sellers generate artificial orders to inflate product or shop performance.

Technical approach

The project focuses on:

• Loading and processing e-commerce order data
• Analysing order patterns between shops and buyers
• Identifying shops with suspicious ordering behaviour
• Identifying buyers associated with potential order brushing
• Using Python and Pandas for data processing and analysis

Task

The challenge required identifying:

1. Shops suspected of conducting order brushing.
2. Buyers suspected of conducting order brushing for each identified shop.

Problem context

Order brushing can be used to create artificial orders in order to inflate a seller’s or product’s ratings and potentially improve its position in search results.

The analysis uses the provided order_brush_order.csv dataset to identify patterns that may indicate this behaviour.

Lessons learned

This project gave me experience working with e-commerce data and using Python and Pandas to analyse it. I also learned how to look for patterns in data and use them to identify potentially unusual behaviour or abnormal activity.