A Named Entity Recognition System for Malayalam Using Conditional Random Fields
A P Ajees, Sumam Mary Idicula · 2018
Named Entity Recognition(NER) is the process of classifying elementary units in a text document into predefined categories such as person, location, organization, etc. It is one of the major steps in the analysis of natural language text. Extracting relevant information from the text is the ultimate goal of NER systems. Various methods are applied for NER in different languages. In this paper, we propose a Named Entity Recognition system for Malayalam using Conditional Random Fields. CRFs are probabilistic graphical models for sequence labeling. The system makes use of different features such as words, preceding words, following words, suffixes of words, etc. Training is conducted on a corpus of 20615 sentences. The proposed system performs with an accuracy of 92.3%.