CUSAT_TEAM@IECSIL-FIRE-2018: A Named Entity Recognition System for Indian Languages

Named Entity Recognition is the process of classifying the elementary units in a text document into meaningful categories such as person, location, organization, etc. It is a significant preprocessing step in the semantic analysis of natural language text. There is an enormous growth of Indian language content on various media types such as websites, blogs, email, chats, etc. over the past decade. Automatic processing of this huge unstructured data is a challenging task especially when the companies are interested to ascertain public view on their products and processes. NER is one of the subtasks of Information Extraction. Extracting structured information from the natural language text is the ultimate goal of IE systems. Different methods are proposed and experimented for NER. In this paper, we propose a Named Entity Recognition system for Indian languages using Conditional Random Fields. Training and testing are conducted using the shared corpus provided by ’ARNEKT-IECSIL 2018’ competition organizers. The evaluation results show that the proposed system is able to outperform most of the reported methods in the competition.