Developing a Computational Mantiq Knowledge Model for Mantiq Learning and Argument Analysis: The Knowledge-Representation Foundation of iMANTIQ

About the Project

Large Language Models (LLMs) can generate fluent explanations, but linguistic fluency does not ensure structurally valid reasoning or transparent rule use. In Mantiq education, key constructs such as *tasawwur, tasdiq, qadiyyah, muqaddimah, natijah, qiyas, burhan,* and *mughalatah* are not yet routinely represented in a machine-interpretable form. This study therefore aims to develop a Computational Mantiq Knowledge Model (CMKM) for Mantiq learning and argument analysis. This study uses a Design Science Research approach to analyze selected Mantiq sources, extract and normalize terminology, define concepts and semantic relations, specify computational rules, and develop provenance structures. The model is organized into six layers: terminology, concepts, propositions, argument roles, rules, and provenance. Domain experts will review construct definitions, relations, and rules, while annotation guidelines will support consistent corpus development. The expected outcome is a validated first version of the CMKM, including an ontology/taxonomy, rule specification, terminology mapping, annotation guideline, and provenance structure for later integration with NLP and symbolic reasoning. The CMKM is intended to provide a structured knowledge layer for transparent Mantiq learning and argument analysis while preserving expert review and source traceability. It also establishes a reusable foundation for future corpus development, explainable interfaces, retrieval, and controlled neuro-symbolic reasoning experiments in educational settings. Video link : https://youtu.be/r1SUHhCtlVk?si=0X_vhv9veiyTqog8