Enhanced Confix Stripping Stemmer And Cosine Similarity For Search Engine in The Holy Qur’an Translation

Abstract

The Holy Qur'an is a Muslim guidelines book. The Holy Qur'an has been translated into many languages to make it easier to understand the meaning of every verse. The lack of understanding of the meaning of the verse on the Holy Qur'an translation needs a misinterpretation or a detailed explanation. Along with technology development, many interpretations of the Holy Qur'an digital version have been equipped with the searching system. However, there are still difficulties in finding an interpretation of The Holy Qur’an in accordance with the discussion or topic because each surah has similarities and differences in a particular topic. This research is purposed a search engine system using Information Retrieval that allows users to find relevant information in the Holy Quran translation based on the information (query) entered by the user. The text processing in this research used Enhanced Confix Stripping Stemmer to find accurate basic words and Cosine Similarity to calculate the similarity of queries and database. Based on the system testing results using Enhanced Confix Stripping Stemmer stemming, the precision is 71.96%, and the ability to find all the relevant verses (recall) is 98.64%, whereas without using stemming, the precision is 82.95% and the recall of 76.70%.