A Survey of Automatic Urdu Language Processing

Most of the research in last few decades has focused on automatic natural language processing (NLP) in English, European and East Asian languages. But unfortunately South Asian languages especially Urdu have received less attention. In this paper we present a survey regarding classification of Urdu language. The main goal of this survey is to present briefly about the material available on Urdu NLP, with the aim to also allow researchers to develop new techniques. So this survey contains the initial attempts in the development of Urdu language. First we give brief description about Urdu language and its written system. Secondly, we introduce Urdu language corpus. Then we compare this corpus to other East Asian languages and discuss which type of problems one has to face while building Urdu corpus. Thirdly, we summarize different linguistic analysis of domain: part-of-speech tagging, parsing and named entity recognition. Overall goal of this paper is to provide a structure which would provide the foundation for future research Urdu language processing techniques