Clustering Data

K-Nearest Neighbors (KNN)

Topic modelling with k-means clustering

K-means clustering videos

• Siehe auch Buch: “Machine Learning with Python Cookbook” Chapter 15
• Very nice visuell explanation → here
• K-Means has two major problems:
  • Number of clusters must be known
  • Doesn't handle outliers
    • But there's a solution! → DBSCAN

K-means clustering groups similar data points into clusters (e.g. customer segmentation, image compression) by finding the optimal number and location of cluster centers.

<aside> 💡 Useful for discovering patterns in unlabeled data.

</aside>

How to find appropriate value of k?

https://twitter.com/rasbt/status/1628427006386884614

Intuition

Lass uns K-Means Clustering anhand einer intuitiven Analogie erklären:

K-Means ist wie das Organisieren einer Party Stell dir vor, du organisierst eine große Party und musst Tische für die Gäste aufstellen:

Grundprinzip: