Learning Word Representations for Tunisian Sentiment Analysis

TL;DR

This paper investigates the use of unsupervised word representations and neural network architectures for sentiment analysis of Tunisian dialect social media text, achieving competitive results without handcrafted features.

Contribution

It introduces a deep learning approach utilizing word2vec and BERT embeddings with CNN and BiLSTM models for Tunisian sentiment analysis, a resource-scarce language.

Findings

Unsupervised embeddings perform well on Tunisian sentiment data

Deep neural networks achieve comparable results to other languages

No handcrafted features are needed for effective sentiment classification

Abstract

Tunisians on social media tend to express themselves in their local dialect using Latin script (TUNIZI). This raises an additional challenge to the process of exploring and recognizing online opinions. To date, very little work has addressed TUNIZI sentiment analysis due to scarce resources for training an automated system. In this paper, we focus on the Tunisian dialect sentiment analysis used on social media. Most of the previous work used machine learning techniques combined with handcrafted features. More recently, Deep Neural Networks were widely used for this task, especially for the English language. In this paper, we explore the importance of various unsupervised word representations (word2vec, BERT) and we investigate the use of Convolutional Neural Networks and Bidirectional Long Short-Term Memory. Without using any kind of handcrafted features, our experimental results on two publicly available datasets showed comparable performances to other languages.