MAPROC at AHaSIS Shared Task: Few-Shot and Sentence Transformer for Sentiment Analysis of Arabic Hotel Reviews

TL;DR

This paper presents a few-shot learning approach using Sentence Transformers for sentiment analysis of Arabic hotel reviews in Moroccan and Saudi dialects, achieving competitive results despite limited annotated data.

Contribution

It introduces a data-efficient SetFit framework for Arabic dialect sentiment analysis, demonstrating its effectiveness in a low-resource domain.

Findings

Achieved 73% F1 score on the shared task dataset.

Showed the viability of few-shot learning for dialectal Arabic sentiment analysis.

Ranked 12th among 26 participants in the shared task.

Abstract

Sentiment analysis of Arabic dialects presents significant challenges due to linguistic diversity and the scarcity of annotated data. This paper describes our approach to the AHaSIS shared task, which focuses on sentiment analysis on Arabic dialects in the hospitality domain. The dataset comprises hotel reviews written in Moroccan and Saudi dialects, and the objective is to classify the reviewers sentiment as positive, negative, or neutral. We employed the SetFit (Sentence Transformer Fine-tuning) framework, a data-efficient few-shot learning technique. On the official evaluation set, our system achieved an F1 of 73%, ranking 12th among 26 participants. This work highlights the potential of few-shot learning to address data scarcity in processing nuanced dialectal Arabic text within specialized domains like hotel reviews.