The TL;DR of EDA
summary
Automated EDA on a 200-wine dataset with 13 chemical/physical features uses pandas and quick seaborn plots, a profiling report (ydata/pandas-profiling style) and a KMeans pipeline that scans 2–11 clusters by silhouette score, selects 3 clusters, then runs t‑tests, PCA and explained variance (PC1 36%, PC2 20%). Clustering matches original class at ~96% and flags 12 outliers for manual review. A chain to an LLM auto-labels clusters (red/white/rosé) and extracts the top features driving each cluster.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 2:26What is EDA and why it matters
- 7:27Automated EDA libraries
- 11:51Auto pipeline for clustering
- 19:48AI for EDA analysis
- 23:26Summary and resources
related talks
38mTransforming Data to Unlock Its Latent Value
32mEDA Techniques for Data Analysis
32mData Transformation: A Framework for Exploratory Data Analysis
37mUsing Exploratory Data Analysis to Discover Patterns
02h 23mEffective Data Visualization
01h 38mEffective Data Visualization
34mDimensionally reducing data squeezing out the good stuff
25mA Python application to flag outliers in very high...