标题：Application of Computational Stylistics and Text Mining Techniques to Identify and Compare Salient Features of Different English Translations of the Holy Quran
期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2020
卷号：20
期号：2
页码：147-154
出版社：International Journal of Computer Science and Network Security
摘要：The presence and role of religion in human life and history is an undisputable phenomenon. The religious text provides foundation for understanding the meaning of life and offers answers of fundamental questions related to human life. Religious text is claimed to be revealed in a particular language on some holy person and later, it was translated and interpreted in multiple languages. Since the text can have multiple meanings, the translation and interpretation depend on the background, style and understandings of the person who is translating and interpreting the text. Quran, the Holy book of the Muslims is a unique literary masterpiece in terms of its linguistics and style. More than forty English translations of the original Arabic text appeared over the past two centuries. These translations of the Holy Quran vary widely in terms of linguistics and style. Various translators adopted different strategies in translating the Quranic Arabic to English language. Comparative studies by linguistic re-searchers identified differences among these translated works in terms of translating Arabic figures of speech, homonyms, metaphors, polysemous words, and overall loyalty and preciseness to the Arabic text. The syntactic and semantic features of each translation also varies greatly. Existing comparative studies are mostly limited in nature with regard to the number of translation aspects com-pared and the size of corpus used for the study. These limitations are primarily due to the linguistic complexity and the sheer large volume of translated works of the Quranic text. The rise of computing power as well as artificial intelligence and machine learning advancements have resulted in crossing many limits. Text mining with its set of linguistic, statistical and machine learning techniques can help overcome the limitations faced by the traditional linguistic researchers. In this paper, computational stylistics and text mining techniques were utilized to perform a detailed comparative study on a large number of available translations of the Holy Quran. The quantification of literary works to measure style and compare translations was performed using internationally recognized metrics. Cosine similarity was used to discover textual level of similarity.
关键词：Text Mining; Computational stylistics; Cosine Similarity; Religious text; Holy Quran
