Development of Hierarchical Part of Speech Tagset

Today is world of natural language processing. The book emphasis on development of tagset for Dravidian languages Kannada in particular. Description of Kannada within the framework of computational linguistics is itself a legitimate research objective, because it meaningfully occupies the space between descriptions offered by grammars of the traditional type and those with restrictive theoretical commitments. The book gives deep insight about how to achieve functional automated annotation system for Kannada . A Hierarchical Part of Speech (HPOS) tagset is developed based on syntax. POS tagger serves as the fundamental building block for NLP researches. Expert Advisory Group on Language Engineering Standards (EAGLES) [1] is taken as guideline with modifications as required for Kannada Language. Designing the tagset for Kannada is challenging due to its inherent complexword formation process in Kannada. Contingent feature which is a special feature for Kannada,voice modifiers, causative feature, reflexive features, clitics information, and semantic information like countable/uncountable features are all considered in our tagset design A total of 171 tag sets are developed.

Autorentext

Dr. Bhuvaneshwari C Melinamath is working currently as a Professor & Head of the Department of Computer Science SVERI, College of Engineering, Pandharapur, Maharastra, has an experience of 27 years in teaching in Engineering College and received Best Faculty award under research Innovation bt CSI Chapter in Technofest 2018 at IIT Bombay.