Challenges in Morphological Analysis of Tamil Biomedical Texts

The purpose of a Morphological analyser is to explore the internal structure of the word and retrieve grammatical features and properties of a morphologically inflected word. Breaking down these amalgamated words is in itself a challenging job in the field of Natural Language Processing. The complexity further increases when the analysis is done on a more ancient and morphologically rich dataset like Tamil Siddha Medicinal documents. In this paper we list the different challenges we faced when trying to explore the syntactic and semantic features of Tamil Siddha texts for building a Tamil Biomedical NER. We also highlight the different fine tuning that was carried out on the analyser to overcome some of the difficulties and possible changes that can be done to improve the accuracy of the analyser in the given domain.