Skip to content

This notebook aims to find groups of people in the OkCupid social network using dimensionality reduction and clustering.

Notifications You must be signed in to change notification settings

robertomancebom/OkCupidClustering

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 

Repository files navigation

OkCupid Clustering

Open In Collab

This notebook is an exercise that tries to visualize different OkCupid profiles and group them into clusters. In order to achieve that, we used different tools and algorithms as tf-idf Vectorizer, UMAP, K-Means or HDBSCAN among others.

Here there are some of the results of the analysis but you can find more on the notebook.

Most Relevant Words using tf-Idf vectorizer

relevant words

Dimensionality Reduction with UMAP

HDBSCAN Clustering

Clustering Model Silhouette Comparison

About

This notebook aims to find groups of people in the OkCupid social network using dimensionality reduction and clustering.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published