Arabic Metaphor Sentiment Classification Using Semantic Information

TL;DR

This work addresses the absence of tools for sentiment analysis of Arabic metaphors by leveraging the Arabic Metaphor Corpus (AMC) and the Arabic Semantic Tagger (AraSAS). It introduces two sentiment classification tools—one that excludes metaphor and one that incorporates metaphor via semantic tags—evaluated with precision, recall, and $F$-score, along with an optimization variant for full semantic tagging. Results show that the non-metaphor approach can achieve higher $F$-scores in many categories, while metaphor-aware methods yield gains in some short-token categories but generally lower overall performance, highlighting token-length effects and the need for richer features to capture metaphor-related nuances. The study demonstrates the potential of semantic tagging to enable metaphor sentiment analysis and points to practical pathways for autonomous metaphor detection with reduced reliance on pre-annotation, paving the way for improved Arabic sentiment analysis in online texts.

Abstract

In this paper, I discuss the testing of the Arabic Metaphor Corpus (AMC) [1] using newly designed automatic tools for sentiment classification for AMC based on semantic tags. The tool incorporates semantic emotional tags for sentiment classification. I evaluate the tool using standard methods, which are F-score, recall, and precision. The method is to show the impact of Arabic online metaphors on sentiment through the newly designed tools. To the best of our knowledge, this is the first approach to conduct sentiment classification for Arabic metaphors using semantic tags to find the impact of the metaphor.

Figures (5)

• Figure 1: Sentiment classification using AraSAS
• Figure 2: Sentiment classification using AraSAS
• Figure 3: Sentiment detection tool
• Figure 4: Sentiment classification and sentiment score for metaphor
• Figure 5: Metaphor sentiment classification tool