Named entity recognition in Odia language: a rule-based approach
Amrita Anandika, Sujata Chakravarty, Bijay Kumar Paikaray
Abstract
NLP can be defined as a process of automatic handling of human spoken languages by computers. NLP is a vast research field linked to multiple areas like artificial intelligence, machine translation, information retrieval etc. NER is an information extraction (IE) process concerned with extracting named entities (NE) from raw data and categorising these NEs into some predefined classes. The process of recognising and extracting NEs from unstructured corpus or data is an essential task for solving different complications in various research fields such as, question answering, summarisation system, information extraction, machine learning, semantic web search, bio-informatics, video annotation and many more. For this research work a classical methodology, i.e., rule-based approach is used to construct a system for automatic identification of NEs from tourism domain data. The system considers words with their repetition and without their repetition and acquires 83% and 71% accuracy respectively.