Segmenting Customers with Machine Learning
summary
Workshop shows end-to-end customer segmentation using real Olabox orders: data cleaning, feature engineering from recent order sequences (orders as items, not static customer profiles), scaling and plotting, then clustering with k-means (choose k, caveats about nonconvex shapes) and DBSCAN (detects noise/outliers). Code runs in a provided Docker container and uses pandas & scikit-learn. Covers how many clusters made sense for their dataset, how features changed results, and how a pipeline can be deployed to label incoming orders.
AI-written summary · about
chapters (5)
- 0:00Introduction and motivation
- 7:01Machine learning basics
- 16:00Data preparation essentials
- 22:48Applying ML at work
- 23:55Q&A and closing
related talks
27mUnlocking Product Growth With Big Data, Data Science And Machine Learning
35mCustomer Segmentation in Python
22mReal-world applications of machine learning: Beyond the hype
56mMachine Learning in E‑commerce
50mBuilding Intelligent Mobile Apps using Machine Learning
22mFrontiers of Machine Learning from the Lens of Retail
49mUnderstanding Customer Journeys with Deep Learning
01h 20mClustering Data