مقالات
A framework for detecting Holy Quran inside Arabic and Persian texts
نویسندگان
Mohsen Shahmohammadi, Toktam Alizadeh, Mohammad Habibzadeh Bijani, Behrouz Minaei
چکیده
This paper presents how to design and implement the Quranic intelligent engine to detect Quranic verses in the texts automatically. Process area of this system is in the scope of text mining processes and its operations are beyond the usual multiple patterns matching for reasons are explained in the paper. A new algorithm based on indexing text and patterns is designed in implementation of this system in which the main idea is to map text and patterns to some numerical arrays and process on them rather than the text. This algorithm detects Quranic verses in two stages. In the first stage, using a new index based exact multiple patterns matching algorithm, Quranic words in the text are detected and are converted to numerical arrays. In the second stage a filter is designed that by search on the arrays is able to detect the indexing sequence between arrays and determine whether these words are a part of Quran. Experimental results show that processing on numerical values rather than the text has a significant impact on increasing the performance of algorithm to be faster and more precise for detecting holy Quran phrases inside the texts.
کلیدواژهها
text mining, exact multiple pattern matching, intelligent detection of Quranic verse