Islam and Data Science Roundup

In “OpenITI MAKHZAN: An Open Annotated Dataset of Arabic, Persian, Ottoman Turkish, and Urdu Print and Manuscript Data” (Journal of Open Humanities Data), Jonathan Parkes Allen (University of Maryland) and others introduce OpenITI MAKHZAN, a “large aggregation of Arabic-script ground truth and evaluation data drawn from a wide variety of Persian, Arabic, Ottoman Turkish, and… CONTINUE READING

In “Sacred or Synthetic? Evaluating LLM Reliability and Abstention for Religious Questions” (arXiv), Farah Atif (Mohamed Bin Zayed University of AI) and others “introduce a novel benchmark FiqhQA focused on the LLM generated Islamic rulings explicitly categorized by the four major Sunni schools of thought, in both Arabic and English. Unlike prior work, which either… CONTINUE READING

In “HATFormer: Historic Handwritten Arabic Text Recognition with Transformers” (arXiv), Adrian Chan (independent scholar) and others “propose HATFormer, a transformer-based encoder-decoder architecture that builds on a state-of-the-art English HTR model. By leveraging the transformer’s attention mechanism, HATFormer captures spatial contextual information to address the intrinsic challenges of Arabic script through differentiating cursive characters, decomposing visual… CONTINUE READING

In “The Affective Algorithm: Mapping the Emotional Architecture of Fatimid Geniza Petitions (Part 2)” (Digital Orientalist), Abdulqadir Haidermota (Aljamea-tus-Saifiyah) observes that, “for the historian of the Mediterranean, Geniza documents are rightly celebrated for their detailed preservation of the material world—offering up the price of flax, the lineage of a merchant, or the date of a… CONTINUE READING

In “The Affective Algorithm: Mapping the Emotional Architecture of Fatimid Geniza Petitions (Part 1)” (Digital Orientalist), Abdulqadir Haidermota (Aljamea-tus-Saifiyah) observes that, “for the historian of the Mediterranean, Geniza documents are rightly celebrated for their detailed preservation of the material world—offering up the price of flax, the lineage of a merchant, or the date of a… CONTINUE READING

In “Implementation of a transformer-based question answering model in KutubBot for the Kutubut Tis’ah Hadith Corpus” (EPJ Web of Conferences), Ana Tsalitsatun Ni’mah (Universitas Trunodjoyo Madura) and others “develop an artificial intelligence-based chatbot, called kutubBot, to answer questions related to Kutubut Tis’ah. This system employs a Transformer-based Question Answering approach to comprehend user questions and… CONTINUE READING

In “Between Hadith and Algorithms: The Epistemic Transformation of Farāiḍ by Era Artificial Intelligence” (International Conference on Islam, Law, and Society Proceedings), Mughniatul Ilma (UIN Kiai Ageng Muhammad Besari Ponorogo) and Husna Ni’matul Ulya (UIN Kiai Ageng Muhammad Besari Ponorogo) seek to “reinterpret the relationship between the revocation of knowledge (naz’ al-‘ilm) in hadith and… CONTINUE READING

In “Learning Word Embeddings from Glosses: A Multi-Loss Framework for Arabic Reverse Dictionary Tasks” (Proceedings of The Third Arabic Natural Language Processing Conference), Engy Ibrahim (Alexandria University) and others “address the task of reverse dictionary modeling in Arabic, where the goal is to retrieve a target word given its definition. The task comprises two subtasks:… CONTINUE READING

In “Video Games and the Work of Islamic Architectural History” (International Journal of Islamic Architecture), Glaire D. Anderson (University of North Carolina at Chapel Hill) discusses “why video games are significant for making Islamic art and history widely accessible” and how “video games and scholarly work in Islamic art, architecture, and cultural heritage can and… CONTINUE READING

In “Maghrebi Dialects—Arabic Bidirectional Translation: An Improved Transformer with Transfer Learning” (Natural Language Processing), Jihad R’baiti (Mohammed VI Polytechnic University) and others “present a hybrid approach for translating the Maghrebi dialects into/from Modern Standard Arabic (MSA). The approach takes advantage of the strengths of the transformer architecture and the BERT language model for transfer learning… CONTINUE READING