#topic-modeling

Question 1: Single choice

What is the main difference between topic modelling a collection of documents and clustering the same documents with K-Means?

K-Means gives each document one cluster; a topic model gives it a mix of topics1
Topic modelling is supervised and needs topic labels, while K-Means is unsupervised2
K-Means outputs named topics such as "sports", while a topic model only outputs cluster ids3
Topic models work only on short texts, K-Means only on long ones4