## Fletcher // w8d1

**Winter 2015**

**03/02/2015**

### Planned schedule and activities

**9:00 am**: I may or may not have drunk 15 coronas last night

**9:15 am**: Moar Unsupervised Learning: Other Clustering Algorithms

**11:00 am**: Additional Unsupervised Learning Challenges

**12:00 pm**: Lunch

**1:30 pm**: Fletcher!

**5:00 pm**: Free as a bird

### Lecture Notes

Slides for Other Clustering Algorithms

#### Reading

Hierarchical Clustering

Hierarchical Clustering Tutorial

Hierarchical Agglomerative Clustering with different linkages

Ward

DBSCAN

Mean Shift

Mean Shift Math

Spectral Clustering

Spectral Clustering Math

Clustering overview in sklearn

Cluster Analysis

Euclidean Distance

Manhattan Distance

Cosine Distance

Jaccard Distance

Scipy distance metrics list

#### Unsupervised Learning Challenges 2

Use the same code for your previous clustering challenges. Repeat each challenge (except the inertia curves, since only the KMeans implementation gives a quick way of calculating that.) However, this time, try (both) Agglomerative Clustering and DBSCAN instead of KMeans. For text clustering, use cosine distance.