Semantic and sentiment analysis of selected Bhagavad Gita translations using BERT-based language framework

  title={Semantic and sentiment analysis of selected Bhagavad Gita translations using BERT-based language framework},
  author={Rohitash Chandra and Venkatesh Kulkarni},
  journal={IEEE Access},
It is well known that translations of songs and poems not only breaks rhythm and rhyming patterns, but also results in loss of semantic information. The Bhagavad Gita is an ancient Hindu philosophical text originally written in Sanskrit that features a conversation between Lord Krishna and Arjuna prior to the Mahabharata war. The Bhagavad Gita is also one of the key sacred texts in Hinduism and known as the forefront of the Vedic corpus of Hinduism. In the last two centuries, there has been a… 

Artificial intelligence for topic modelling in Hindu philosophy: Mapping themes between the Upanishads and the Bhagavad Gita

This paper uses advanced language models such as BERT to provide topic modelling of the Upanishads and the Bhagavad Gita and finds that topics generated by the BERT-based models show very high coherence when compared to the conventional models.

An ensemble deep learning classifier for sentiment analysis on code-mix Hindi–English data

This work ensembles a multilingual variant of RoBERTa and a sentence-level embedding from Universal Sentence Encoder to identify the sentiments of these code-mixed tweets with higher accuracy, and optimises the classifier's performance by using the strength of both for transfer learning.



A comprehensive survey on machine translation for English, Hindi and Sanskrit languages

A comprehensive survey of MTS in general and for English, Hindi and Sanskrit languages in particular is presented and the availability of MT language modeling tools, parsers data repositories and evaluation metrics is tabulated.

Multi-Word Expressions in the Spanish Bhagavad Gita, Extracted with Local Grammars Based on Semantic Classes

The theory and application of Local Grammars is described, and the very first results are discussed in detail.

A statistical probe into the word frequency and length distributions prevalent in the translations of Bhagavad Gita

A statistical study has been conducted on Bhagavad Gita. Four measures have been derived for the original text in Sanskrit and its translations in Hindi, English and French. First, word frequency

COVID-19 sentiment analysis via deep learning during the rise of novel cases

The results indicate that the majority of the tweets have been positive with high levels of optimism during the rise of the novel COVID-19 cases and the number of tweets significantly lowered towards the peak, and the predictions generally indicate that although the majority have been optimistic, a significant group of population has been annoyed towards the way the pandemic was handled by the authorities.

Attention-based LSTM for Aspect-level Sentiment Classification

This paper reveals that the sentiment polarity of a sentence is not only determined by the content but is also highly related to the concerned aspect, and proposes an Attention-based Long Short-Term Memory Network for aspect-level sentiment classification.

Knowledge Representation in Sanskrit and Artificial Intelligence

This article demonstrates that a natural language can serve as an artificial language also, and that much work in AI has been reinventing a wheel millenia old.

Translating for singing: the theory, art and craft of translating lyrics

The value of this volume lies not in its coverage of theory or history – I’ll come back to that later – but in practicalities.

Living the Bhagavad Gita at Gandhi’s Ashrams

The Bhagavad Gita is a philosophical Hindu scripture in which the god Krishna imparts lessons to the warrior prince Arjuna about sacred duty (dharma) and the path to spiritual liberation (moksha).

On Stopwords, Filtering and Data Sparsity for Sentiment Analysis of Twitter

The results show that using pre-compiled lists of stopwords negatively impacts the performance of Twitter sentiment classification approaches and the dynamic generation of stopword lists appears to be the optimal method to maintaining a high classification performance while reducing the data sparsity and shrinking the feature space.