Experiences in building the Nepali WordNet-insights and challenges

Machine translation in Nepali language is in an infant stage in comparison to other scheduled languages of India like Hindi, Sanskrit, Tamil, etc. One of the major reasons behind this is the non-availability of rich lexical resources in Nepali. The Nepali WordNet is thus an endeavour to prepare a rich lexical resource for the Nepali Language for effective machine translation and to facilitate the development of Information and Communication Technologies in Nepali. The endeavour is inspired by the famous English WordNet and the Hindi WordNet. In the present paper we discuss some of the preliminaries involved in this attempt like the expansion approach of WordNet creation, the linguistic challenges involved, WordNet creation tool interface and the synsets’ storage structure. We also discuss some of the special characteristics of the Nepali language.