@inproceedings{barbadikar-kulkarni-2024-automatic,
    title = "Automatic {S}anskrit Poetry Classification Based on K{\={a}}vyaguṇa",
    author = "Barbadikar, Amruta  and
      Kulkarni, Amba",
    editor = "Lalitha Devi, Sobha  and
      Arora, Karunesh",
    booktitle = "Proceedings of the 21st International Conference on Natural Language Processing (ICON)",
    month = dec,
    year = "2024",
    address = "AU-KBC Research Centre, Chennai, India",
    publisher = "NLP Association of India (NLPAI)",
    url = "https://preview.aclanthology.org/Add-Cong-Liu-Florida-Atlantic-University-author-id/2024.icon-1.13/",
    pages = "109--119",
    abstract = "K{\={a}}vyaguṇa denotes the syntactic and phonetic attributes or qualities of Sanskrit poetry that enhance its artistic appeal, commonly classified into three categories: M{\={a}}dhyurya (Sweetness), Oja (Floridity), and Pras{\={a}}da (Lucidity). This paper presents the K{\={a}}vyaguṇa Classifier, a machine learning module, designed to classify Sanskrit literary texts into three distinct guṇas, by employing a diverse range of machine learning algorithms, including Random Forest, Gradient Boosting, XGBoost, Multi-Layer Perceptron and Support Vector Machine. For vectorization, we employed two methods: the neural network-based Word2vec and a custom feature engineering approach grounded in the theoretical understanding of K{\={a}}vyaguṇas as described in Sanskrit poetics. The feature engineering model significantly outperformed, achieving an accuracy of up to 90.6{\%}"
}