Gzip versus bag-of-words for text classification

TL;DR

This paper compares gzip compression and bag-of-words methods for text classification, demonstrating that bag-of-words can match or outperform gzip in accuracy and efficiency.

Contribution

It shows that bag-of-words approaches are competitive with gzip compression for text classification, offering similar or better results with greater efficiency.

Findings

Bag-of-words achieves comparable or better accuracy than gzip.

Bag-of-words is more computationally efficient.

Bag-of-words can be a preferable alternative to gzip for text classification.

Abstract

The effectiveness of compression in text classification ('gzip') has recently garnered lots of attention. In this note we show that `bag-of-words' approaches can achieve similar or better results, and are more efficient.