[Paper Review] The Arabic Noun System Generation

Abdelhadi Soudi, Cavalli-Sforza, Violetta+1 more|arXiv (Cornell University)|May 17, 2024

Syntax, Semantics, Linguistic VariationArts and Humanities3 citations

TL;DR

This paper proposes a multiple-stem approach to Arabic noun plural generation within Lexeme-based Morphology, eliminating the need for complex truncation or deletion rules by pre-specifying plural affixes (uwna for masculine, aAt for feminine) in the lexicon. It demonstrates that broken plurals are better modeled as distinct stems rather than derived from roots, with validation through linguistic and statistical evidence and implementation in the MORPHE system using equivalencing nodes.

ABSTRACT

In this paper, we show that the multiple-stem approach to nouns with a broken plural pattern allows for greater generalizations to be stated in the morphological system. Such an approach dispenses with truncating/deleting rules and other complex rules that are required to account for the highly allomorphic broken plural system. The generation of inflected sound nouns necessitates a pre-specification of the affixes denoting the sound plural masculine and the sound plural feminine, namely uwna and aAt, in the lexicon. The first subsection of section one provides an evaluation of some of the previous analyses of the Arabic broken plural. We provide both linguistic and statistical evidence against deriving broken plurals from the singular or the root. In subsection two, we propose a multiple stem approach to the Arabic Noun Plural System within the Lexeme-based Morphology framework. In section two, we look at the noun inflection of Arabic. Section three provides an implementation of the Arabic Noun system in MORPHE. In this context, we show how the generalizations discussed in the linguistic analysis section are captured in Morphe using the equivalencing nodes.

Motivation & Objective

• To address the complexity of Arabic broken plural morphology, which traditionally relies on intricate truncation and deletion rules.
• To challenge the assumption that broken plurals are derived from singular forms or roots.
• To propose a more generalizable and linguistically coherent model of Arabic noun plurals using multiple stems.
• To implement the proposed system in the MORPHE morphological processor to validate its feasibility.
• To demonstrate that sound noun inflection requires pre-specification of plural affixes (uwna, aAt) in the lexicon.

Proposed method

• Adopting the Lexeme-based Morphology framework to model Arabic nouns with multiple stems instead of deriving plurals from roots.
• Pre-specifying the sound plural masculine (uwna) and feminine (aAt) affixes directly in the lexicon to support inflection.
• Using equivalencing nodes in the MORPHE system to represent morphological generalizations across noun paradigms.
• Evaluating the model using linguistic analysis and statistical evidence against root-derivation hypotheses.
• Structuring the system to avoid deletion and truncation rules by treating broken plurals as independent lexical entries.
• Applying the model to both broken and sound plural systems to ensure consistency across noun types.

Experimental results

Research questions

• RQ1 Can the Arabic broken plural system be modeled more effectively without relying on root-derivation or deletion rules?
• RQ2 Is there linguistic and statistical evidence supporting the independence of broken plurals from singular forms?
• RQ3 How can the sound plural affixes (uwna and aAt) be systematically encoded in a morphological lexicon?
• RQ4 To what extent can the multiple-stem approach generalize across Arabic noun paradigms?
• RQ5 How can the MORPHE system effectively represent the proposed morphological generalizations using equivalencing nodes?

Key findings

• The multiple-stem approach successfully eliminates the need for complex truncation and deletion rules in modeling Arabic broken plurals.
• Linguistic and statistical evidence supports the independence of broken plurals from singular forms, challenging the root-derivation hypothesis.
• Pre-specification of the affixes uwna (masculine) and aAt (feminine) in the lexicon is essential for accurate sound noun inflection.
• The implementation in MORPHE confirms that the proposed generalizations can be effectively captured using equivalencing nodes.
• The model demonstrates greater consistency and generalizability compared to traditional derivation-based approaches.
• The study provides a robust, rule-minimal framework for Arabic noun system generation that supports both broken and sound plurals.