Lexico-Semantic Relations for Punjabi WordNet

Abstract

WordNet is a lexical structure composed of synsets. Synsets are linked by different lexical and semantic relations like hypernymy, hyponymy, meronymy, troponymy, holonymy, entailment, antonym and gradation etc. India is a multilingual country where people speak many different languages, which results in communication barrier. This acted as a motivation behind the building of IndoWordNet. IndoWordNet is a linked structure of WordNets of major Indian languages from Indo-Aryan, Dravidian and Sino-Tibetan families. Many Indian languages are using the expansion approach to develop their WordNets from the Hindi WordNet. Punjabi is the language of hundreds of millions of people in India, and is the religious language of all Punjabis around the world. Surprisingly, little work has been done in the field of computerized language and lexical resources for this language. It is therefore worthy to build up a Punjabi lexical resource (WordNet) that can discover the richness of Punjabi language. This thesis comprises of six chapters. Chapter 1 provides the brief overview of WordNet, challenges in natural language processing, role of WordNet in natural language processing, basic WordNet principle and approaches for WordNet creation. Chapter 2 describes the various constituents of WordNet and relations used in development of WordNet. The related work done in this field in the form of English WordNet and Hindi WordNet are also presented in this chapter. In chapter 3 problem statement is specified along with the objectives defined for thesis. A methodology to achieve those objectives is also given in chapter 3. Chapter 4 describes approach towards building a lexical resource in Punjabi language. Punjabi WordNet is being developed from Hindi WordNet using the expansion approach. The process of creation of synsets for Punjabi WordNet from Hindi WordNet, proposed database design for storage of lexical and semantic relations, porting of text files to designed database are also presented in this chapter. Algorithms to create the lexical relations like antonymy, gradation, compounding and conjunction etc. for Punjabi WordNet have also been discussed in this chapter. Chapter 5 demonstrates the working of various tools like lexical creation tool and web application for Punjabi WordNet. The conclusion and future scope of work presented in this thesis is described in chapter 6. Different tools for the creation of lexical relations like antonymy, compounding, conjunction, gradation and web application for Punjabi WordNet has been developed during this thesis work.