A Computational Algorithm for Metrical Classification of Verse

TL;DR

This paper introduces a computational algorithm for classifying Sanskrit and English verses by metre, utilizing rules of chandas and euphonic conjunctions to improve accuracy in verse analysis.

Contribution

It presents a novel computational scheme and algorithms for metre classification of Sanskrit and English verses, including correction of euphonic conjunction errors.

Findings

Effective metre classification for Sanskrit and English verses

Incorporation of euphonic conjunction rules improves accuracy

Applicable to Unicode-encoded texts

Abstract

The science of versification and analysis of verse in Sanskrit is governed by rules of metre or chandas. Such metre-wise classification of verses has numerous uses for scholars and researchers alike, such as in the study of poets and their style of Sanskrit poetical works. This paper presents a comprehensive computational scheme and set of algorithms to identify the metre of verses given as Sanskrit (Unicode) or English E-text (Latin Unicode). The paper also demonstrates the use of euphonic conjunction rules to correct verses in which these conjunctions, which are compulsory in verse, have erroneously not been implemented.