Information extraction for conversational systems in Indian languages - ARnekt IECSIL

Abstract

Data being the new source of wealth, mining intelligence from every possible units of it, has become today�s salient feature in many fields. Text data is not limited to one language and this has showcased its usability in creating multiple applications from various languages. Development of Indian languages is just getting better both in terms of resource and application specific. Information Extraction for Conversational Systems in Indian Languages - Arnekt IECSIL has taken its step in creating its own resource in Indian languages (Hindi, Kannada, Malayalam, Tamil and Telugu) for Named Entity Recognition (NER) and Information Extraction (IE) tasks. This overview paper will be detailing more on the existing Indian language corpora development and the steps taken for building our own corpus along with its statistics. � 2016 Association for Computing Machinery.