A Machine Learning Approach For Opinion Holder Extraction In Arabic Language

TL;DR

This paper introduces a novel approach for extracting opinion holders in Arabic news texts, overcoming the lack of robust parsers by using a comprehensive feature set and machine learning techniques, achieving over 54% F-measure.

Contribution

It presents the first approach for opinion holder extraction in Arabic without relying on lexical parsers, using a new feature set and semi-supervised learning methods.

Findings

Achieved 54.03 F-measure in opinion holder extraction

Developed and released a new Arabic opinion mining corpus and lexicon

Demonstrated effectiveness of feature-based models without structural parsers

Abstract

Opinion mining aims at extracting useful subjective information from reliable amounts of text. Opinion mining holder recognition is a task that has not been considered yet in Arabic Language. This task essentially requires deep understanding of clauses structures. Unfortunately, the lack of a robust, publicly available, Arabic parser further complicates the research. This paper presents a leading research for the opinion holder extraction in Arabic news independent from any lexical parsers. We investigate constructing a comprehensive feature set to compensate the lack of parsing structural outcomes. The proposed feature set is tuned from English previous works coupled with our proposed semantic field and named entities features. Our feature analysis is based on Conditional Random Fields (CRF) and semi-supervised pattern recognition techniques. Different research models are evaluated via cross-validation experiments achieving 54.03 F-measure. We publicly release our own research outcome corpus and lexicon for opinion mining community to encourage further research.