Mathematical Structuring, Encoding and Algorithmic Processing of Hindi / Sanskrit Poems

One cannot ignore the importance of natural languages as the basic means of communication among humans. These languages generally evolve in humans without proper planning. They are unlike constructed and formal languages which are used to develop programs for computing machines. Therefore, natural languages pose numerous challenges in human-computer interaction. Poems play a major role in enhancing the expressive power of a natural language. Due to its very nature, it is difficult for an ordinary human to write poems, and poets are generally not computer scientists. Hence, it is very difficult to make a computer understand poems. Sanskrit and Hindi (along with its dialect Avadhi) have very rich literature in the form of poems. After Mandarin Chinese, Spanish and English; Hindi has the highest number of native speakers in the world. More than 50% population of India read, write and speak Hindi. In this paper, a few algorithms are presented that implement the concept of encoding of Hindi poems and detect various poetic devices present in them, like Ardhali, Chaupaiya etc. The results show that the encoding of Ramcharitmanas is 100% correct, as per the algorithms presented in this paper whereas other forms of poetic devices yield more than 95% accuracy when it comes to automatic detection using these algorithms.