Semantic search engine for the holy Quran
Ahmed, Omar (2024)
2024
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2024051010827
https://urn.fi/URN:NBN:fi:amk-2024051010827
Tiivistelmä
Fetching verses from the holy Quran is not an easy process, even using Google or ChatGPT will not help that much in this task and the user will get a ton of articles or untrusted quality data because ChatGPT still can make mistakes. This tool is important especially for Muslims to search verses by using the natural language and for everyone to know more about the Quran.
From theory to practice, the biggest challenge was which path needed to be used to solve this task. From using pre-trained models to topic modeling to word2vec that was the key point to the solution with cosine similarity. In the end, a web UI was built on top of that, that takes user queries and returns relevant Quranic verses.
From theory to practice, the biggest challenge was which path needed to be used to solve this task. From using pre-trained models to topic modeling to word2vec that was the key point to the solution with cosine similarity. In the end, a web UI was built on top of that, that takes user queries and returns relevant Quranic verses.