Mean Shift Algorithm to Determine Customer Segmentation in Online Store Sales

Ryan Reliovani (Teknik Informatika, UIN Sunan Gunung Djati Bandung)
Nina Nadia Syafitri Husein (Teknik Informatika, UIN Sunan Gunung Djati Bandung)
Kamal Zaki Abdurrafi (Teknik Informatika, UIN Sunan Gunung Djati Bandung)
Cecep Rafqi Al Husni (Teknik Informatika, UIN Sunan Gunung Djati Bandung)
Muhammad Azka Khowarizmi (Teknik Informatika, UIN Sunan Gunung Djati Bandung)

Gunung Djati Conference Series
Vol. 3 (2021): Mini Seminar Kelas Data Mining 2020

Market segmentation is one of the most important things for a business or business, with market segmentation a shop or company can see the purchasing power, needs and customers of customers. The purpose of this study was to determine the value of customer segmentation in an online shop based in the UK where the main sales are unique gifts for various events where the shop's customers are wholesalers from various countries. Data mining with clustering techniques is used in this study. The algorithm used to build clusters is the Mean Shift algorithm, with an estimated bandwidth value of  1.55, the quantile value = 4, epsilon = 4% and n_samples = 5000, there are 3 clusters visualized using a scatter plot model.