Gaussian Mixture Model
This code implement the learning step of Gaussian Mixture Models, a statistical clustering model.
- Input: 20-news-groups processed documents, with each document is represented by a list of its words.
- Output: Cluster them into different groups.
- Ideas: Suppose each document is a multivariate random variable generated from a random process. Decode that process, we will find the way to group them. The details of this method is described in file document.pdf