قرآنی علوم میں کمپیوٹیشنل طریقہ: موضوعاتی،لسانی اور کراس ریسرچ کیلیےٹی ایف-ای ڈی ایف(TF-IDF) اور ارا بیرٹ (ARABERT)الگورتھم کا تقابلی تجزیہ

Computational Methods in Quranic Studies: Comparative Analysis of TF-IDF and ARABERT Algorithms for Thematic, Linguistic and Cross-Sectional Research

Abstract

The Quranic text, as the central source of Islamic knowledge and guidance, requires precise analytical tools for thematic classification, linguistic analysis, and the cross-referencing of verses (Ayat). Traditional manual methods, while scholarly authoritative, are time-consuming and subject to interpretive variation. This study proposes to investigate the application of computational methods, specifically Term Frequency–Inverse Document Frequency (TF-IDF) and Bidirectional Encoder Representations from Transformers for Arabic (AraBERT), as analytical aids for Quranic research. Using Quranic text from the Tanzil corpus and TF-IDF analysis conducted through Voyant Tools, the findings were evaluated against classical Tafsir literature, including the works of Ibn Kathir and Al-Tabari. TF-IDF demonstrated 62–80% agreement with manual scholarly findings, depending on the task type, while AraBERT, validated through existing literature, achieved 80–88% accuracy through context-aware semantic embeddings. A hybrid approach combining both methods is proposed to improve analytical performance, potentially achieving 84–89% accuracy. This study provides a practical roadmap for researchers in Quranic Studies to incorporate computational methods with minimal technical barriers while establishing a quantitative framework for future algorithmic research in Islamic textual analysis.