Centroid estimation based on symmetric KL divergence for Multinomial text classification problem

TL;DR

This paper introduces a novel centroid estimation method for text classification using symmetric KL-divergence, leading to significant performance improvements over traditional classifiers on standard datasets.

Contribution

The paper proposes a new centroid estimation technique based on symmetric KL-divergence specifically for multinomial text classification, enhancing classification accuracy.

Findings

Significant improvement over traditional classifiers

Effective centroid estimation method

Validated on multiple standard datasets

Abstract

We define a new method to estimate centroid for text classification based on the symmetric KL-divergence between the distribution of words in training documents and their class centroids. Experiments on several standard data sets indicate that the new method achieves substantial improvements over the traditional classifiers.