SMARTies: Sentiment Models for Arabic Target Entities

TL;DR

This paper introduces SMARTies, a system for entity-level sentiment analysis in Arabic that leverages morphological and distributional semantic features to improve target and sentiment identification in complex posts.

Contribution

The paper presents a novel Arabic sentiment analysis system that combines morphological representations and semantic clustering, achieving significant performance improvements.

Findings

Morphological features enhance target and sentiment detection.

Distributional semantic clusters further improve accuracy.

System outperforms multiple baseline models.

Abstract

We consider entity-level sentiment analysis in Arabic, a morphologically rich language with increasing resources. We present a system that is applied to complex posts written in response to Arabic newspaper articles. Our goal is to identify important entity "targets" within the post along with the polarity expressed about each target. We achieve significant improvements over multiple baselines, demonstrating that the use of specific morphological representations improves the performance of identifying both important targets and their sentiment, and that the use of distributional semantic clusters further boosts performances for these representations, especially when richer linguistic resources are not available.