To calculate the term frequency of each term on each documents. For example, "spywar business.001 1".
To filter out terms that appear less than 3 times across all documents.
To get the top 10 most frequent terms in all documents.
For each class, find the top 5 most frequent terms.
Implementing a simple Kmeans clustering for 2D points.
To calculate the TF-IDF score for the BBC dataset.