Seminario: " Automatic Identification of Rhetorical Elements in Classical Arabic Poetry"

Interviene: Ali Hussein (University of Haifa)

Coordinación: María Ángeles Gallego (ILC-CSIC)

Abstract: This presentation introduces a rule-based automatic framework—the Rhetorical Element Identifier (REI)—for detecting rhetorical elements in Classical Arabic poetry. The framework is proposed as a provisional solution until a sufficiently large annotated corpus becomes available to support machine learning approaches. Manually identifying such features is not only time-consuming but also requires the expertise of highly specialised scholars.

To address this challenge, REI combines natural language processing techniques with a carefully structured rule-based system, allowing for gradual refinement as more annotated data becomes available. This hybrid approach represents a foundational step toward building a corpus large enough for machine learning applications in this domain.

The study focuses on 20 rhetorical elements, all recognised as central by classical Arabic rhetoricians, and demonstrates highly promising results. The project was led by Ali Hussein, a specialist in Classical Arabic poetry, in collaboration with colleagues from the Department of Information Systems, who developed the technical components of the framework.