(IEEE, 2011) H Abdel Ghafour, Hesham; El-Bastawissy, Ali; A Heggazy, Abdel Fattah
String matching algorithms play a vital & crucial role in many applications such as search engines, object identification, hand written recognition, name searching in large databases, data cleansing, and automatic spell checking. Many algorithms have been developed to measure string similarity but most of them designed mainly to handle Latin-based languages. In this paper, we propose a new algorithm for Arabic string matching which takes into consideration the unique features of the Arabic language and the different similarity levels of the Arabic letters such as phonetic similarity and character form similarity in addition to keyboard distance.
