Predicting Stress in Sanskrit Texts: A Deep Learning Approach to Sentiment Analysis.

Sanskrit, one of the world's oldest languages, grammar plays major role in language translation, involving the structural arrangement of sentences through specific guidelines. Recently, there has been growing interest in the analysis of Sanskrit, particularly in translating the Bhagavad Gita into various languages. However, there is a lack of work validating the excellence of these English conversions. Advances in verbal models motorized by deep learning have not solitary facilitated conversions but also enhanced the sympathetic of languages and manuscripts through sentimentality examination. Despite these advancements, natural language processing (NLP) tasks such as machine conversion and sentimentality examination for Sanskrit have not been fully discovered, largely due to the scarcity of available data. To address these challenges, we present a sentiment analysis to predict the stress of Sanskrit texts using deep learning technique. We first perform the text preprocessing with the help of NLP transformer that considers word sequences to ascertain the accurate interpretation of words. Next, the XLNet based feature extraction is used to extracts meaningful features from the preprocessed results. We design the modified hyper spherical searching (MHSS) algorithm is used to selects the optimal features to reduce the dimensions. A dynamic dual-layer Q-learning (DDQL) model is present for sentiment analysis to classify the sentiments and predict the stress form Sanskrit text. We authenticate the effectiveness of the planned perfect using selected chapters and verses from the Bhagavad Gita across different translations. The proposed framework has demonstrated superior performance compared to existing methods for translation and sentiment classification. [ABSTRACT FROM AUTHOR]