مقالات
A new framework for detecting similar texts in Islamic Hadith Corpora
نویسندگان
Hossein Juzi, Ahmed Rabiei Zadeh, Ehsan Barati, Behrouz Minaei-Bidgoli
چکیده
Nowadays similarity detection is one of the most applicable aspects of text mining techniques. There are different methods for similarity detection. This paper presents a new system for text similarity detection in Islamic Large Hadith Corpus of Computer Research Center of Islamic Science (CRCIS). This system uses Ngram method and Cosine measure for similarity detection. According to evaluation result, computer-based similarity detection systems can be more efficient than the previous related work in text similarity detection. We have obtained a 97% F-Score of similarity detection for Hadith texts. We hope that our system enables researches to find the unified Hadiths and to detect that how one large Hadith is divided into several small pieces of Hadith in different traditional Hadith books. This system would be very fruitful for many researches in the area of Hadith and the Holy Qur’an investigations.
کلیدواژهها
Text similarity, similarity detection, Hadith, prophetic traditions, text mining in Islamic texts