Quranic Cross-Lingual Information Retrieval Optimization using Hexadecimal Conversion Algorithm

Abstract

The electronic Quran text has a huge amount of words and lack of transliteration for machine understanding which drop the query search performance of the Digital Quran. The Holy Quran consists of 30 juzu, 60 hizb, 114 chapters, 6236 verses, 77439 words and 320015 letters. Therefore, this research proposed the Quran Hexadecimal (QuHex) model as the solution to represent the words of the Holy Quran by using hexadecimal conversion algorithm to improve machine readability at the presentation layer. The result is 3 different languages into were translated into hex value. Instead of storing into 3 different dictionaries, we used only 1 dictionary in third party language (machine). This research also improves the space usage as the reduction of space size is around 47-54 %. This research hypothesizes that the higher the reduction of space usage, the faster the retrieval time of CLIR. This article highlights the proposed Quranic Cross-Lingual Information Retrieval (Q-CLIR) model and Hexadecimal Conversion Algorithm. Currently, QuHex is still undergoing a continuous prototyping development and will implement the 2nd phase in creating interpreter for any word in a hex value. The implementation strategy and suggestions for the future works are also given.