# A Simple and Effective Approach for Fine Tuning Pre-trained Word Embeddings for Improved Text Classification

## TL;DR

This paper introduces a straightforward method for fine-tuning pre-trained word embeddings by incorporating class information, enhancing their discriminative power for text classification tasks across multiple datasets.

## Contribution

The proposed approach uniquely integrates class context into word embeddings during fine-tuning, improving their effectiveness for text classification.

## Key findings

- Significant improvement in classification accuracy across datasets
- Effective for both Arabic and English text classification
- Enhances word vector discriminability within classes

## Abstract

This work presents a new and simple approach for fine-tuning pretrained word embeddings for text classification tasks. In this approach, the class in which a term appears, acts as an additional contextual variable during the fine tuning process, and contributes to the final word vector for that term. As a result, words that are used distinctively within a particular class, will bear vectors that are closer to each other in the embedding space and will be more discriminative towards that class. To validate this novel approach, it was applied to three Arabic and two English datasets that have been previously used for text classification tasks such as sentiment analysis and emotion detection. In the vast majority of cases, the results obtained using the proposed approach, improved considerably.

---
Source: https://tomesphere.com/paper/1908.02579