Zikr: An AI-Driven Islamic Social Platform
Loading...
Date
2025
Authors
Mohammad Raed
Mohammad Matar
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Abstract
This project presents Zikr an innovative platform designed to support Muslims worldwide
in their religious journey by combining social community features with advanced Artificial
Intelligence. Unlike traditional Islamic apps that focus on a single functionality, Zikr inte-
grates a vibrant social media environment—enabling users to create posts, comment, like,
and interact—while also providing AI-powered tools for religious education and engagement.
The platform’s core AI components include: (1) a Retrieval-Augmented Generation (RAG)
chatbot that answers Islamic Fatawa questions in Arabic and English, leveraging a dataset
scraped from the IslamQA website to ensure accurate, authentic responses; and (2) a K-Means
clustering model that highlights Quranic topics with distinct colors, enhancing the recitation
experience. Additionally, Zikr features a personalized content suggestion algorithm for the
social feed, utilizing a fuzzy logic system to recommend posts based on user interests and
activity, thus fostering meaningful engagement and community building.
The methodology involved two main tracks: application development and AI integration.
The application was built using a Flutter frontend and a Django backend, with features such as
user authentication, profile management, and a robust social module. The AI track included
dataset collection and preprocessing, model selection and training, and seamless integration
with the platform. The suggestion algorithm was designed to balance personalization with
content diversity, ensuring users are exposed to both relevant and varied topics.
The main challenges addressed were the scarcity of high-quality Islamic datasets, the
complexity of Quranic language, and the need for a safe, distraction-free social space. By
overcoming these, Zikr offers a unique, holistic solution that advances the use of AI in Islamic
education and community interaction. The project demonstrates the potential of combining
modern AI techniques with user-centric design to create impactful digital tools for the Muslim
community.
Keywords: Islamic AI, Social Community, Quran Topic Modeling, Fatawa Chatbot, Fuzzy
Suggestion Algorithm, RAG, Arabic NLP, Flutter, Django, K-Means Clustering