AyaMakna: Qur'anic Semantic Intelligence System
AyaMakna is a web application that seeks to explore structural and thematic patterns in the Qur’an through computational semantic analysis. All 6,236 verses across 114 surahs are loaded, processed, and visualized as an interactive force graph. Each view mode exposes a different layer of meaning.
❓ Story Behind It: This Ramadan, while reading the Qur’an, I noticed something different in the translation format. The Al-Qur'an that I read presents the verses thematically. Since then, I realized that I had often learned the Qur’an linearly, reading translations verse by verse without figuring out the broader concepts. This approach made it difficult to see thematic relationships, conceptual patterns, and structural contrasts across verses and juz. Although many digital Qur’an platforms provide search functionality, and search engines offer broad access to knowledge, few platforms enable users to explore semantic connections in a visual and analytical way. 💡 Offered Solution: AyaMakna represents the verses from the Qur'an in a graph-based data model, mapping shared root words, concepts, actions, and contrasts into an explorable network. Instead of static reading, users can navigate meaning through connections, revealing patterns that are not immediately visible in most of formats. 🚀 How it Works: 1. Text Preprocessing & Linguistic Normalization → the corpus is tokenized and linguistically normalized, with each word tagged for part-of-speech and mapped to its morphological root and lemma. 2. Semantic Entity Extraction → extracted and categorized into conceptual, behavioral, and ontological layers. 3. Relationship Modeling → graph construction 4. Feature Engineering & Structural Metrics → frequency, centrality, and polarity are computed to quantify semantic importance and contrast dynamics. 5. Intelligence Layer → dynamic filtering into multiple analytical lenses, enabling focused exploration of roots, concepts, behaviors, and semantic oppositions. 6. Visualize → put them into node-edge graph to have better understanding in terms of relationship between verses. 🛠️ Tech Stack: - Frontend: React 18, TypeScript, Vite, D3.js, Tailwind CSS - Backend & Data: PostgreSQL (Supabase), RESTful API integration - AI-Assisted Development: Lovable.dev + Claude Code (Sonnet 4.6) - Skill: NLP, deep concept-code review, concept-to-product implementation, structured and analytical thinking. 🤔 What to Expect: AyaMakna demonstrates how richly layered scripture of Qur'an can be analysed with science approach into a knowledge system. This project highlights patterns such as shared roots connections, conceptual clustering, recurring contrasts, and thematic density. It shows how graph structures can enhance interpretability and exploratory learning. ✨ Why This Matters: This project reflects my approach to data: transforming rich and complex information into structured systems that enhance clarity and exploration. It bridges text analysis, semantic modeling, and user-centered visualization in a meaningful way. repo: https://github.com/umarfadhil/ayamakna/