.png){kind=logo}
According to Eknath Easwaran, M.K. Gandhi, and Purohit Swami's analysis of the quality of Bhagavad Gita English translations, machine learning and other artificial intelligence (AI) approaches have achieved enormous success in scientific and technological tasks like determining how protein molecules form, and identifying faces in a crowd. The use of these methodologies in the humanities, on the other hand, is yet to be substantially explored. But, what can AI teach us about philosophy and religion? They employed deep learning AI algorithms to analyse English versions of the Bhagavad Gita, an ancient Hindu scripture written initially in Sanskrit, as a starting place for such an investigation.
They investigated sentiment and semantics in translations using BERT, a deep learning language model. Despite considerable differences in language and sentence construction, they discovered that emotional and meaning-related patterns were largely comparable in all three. This study paves the way for the application of AI-based tools to compare translations and assess sentiments in a variety of texts.
The Bhagavad Gita is religious and intellectual literature in Hinduism. It was written almost 2,000 years ago, has been translated into over 100 languages, and has piqued the interest of western scholars since the eighteenth century. The 700-verse composition is part of the broader Mahabharata epic, which describes the events of an ancient battle believed to have taken place near modern-day Delhi in India at Kurukshetra. The Bhagavad Gita text recounts a discourse between Lord Krishna and a nobleman named Arjuna. They debate whether a soldier must go to war for ethical and moral reasons if he or she has close friends or relatives on the opposite side. The scripture was critical in laying the groundwork for Hinduism.
There have been numerous English translations of this holy book, however, there is little evidence to support their excellence. Translations of poems and songs not only disrupt the rhythm and rhyming structures but can also lead to semantic content loss. In 1785, the first of numerous English translations of the Bhagavad Gita was produced. In their study, they employed a deep learning language method to assess 3 selected Bhagavad Gita translations (from Sanskrit to English) using semantic and emotional analyses, which aid in the assessment of translation quality. They used BERT, a Google-developed pre-trained language model. They fine-tuned the algorithm even further by utilising a human-labeled training dataset based on Twitter tweets that captures ten different attitudes. These sentiments (optimistic, thankful, empathetic, pessimistic, anxious, sad, annoyed, denial, surprised, and joking) were adapted from their prior study of social media attitudes during the COVID-19 pandemic's beginning.
The three translations they looked at had extremely varied vocabulary and structure, but the language model identified similar feelings in different chapters of the translations. According to their model, the most often stated emotions are optimism, irritation, and surprise. Furthermore, the model demonstrated how the overall sentiment polarity shifts (from negative to positive) during Arjuna and Lord Krishna's discourse. Arjuna is initially gloomy, but as Lord Krisha teaches him Hindu philosophy, he becomes hopeful. Krishna's sentiments demonstrate how, with philosophic knowledge of dharma and mentorship, a disturbed mind can find clarity to make the right decisions in times of conflict. Their model's one restriction is that it was taught on Twitter data, so it identifies "joking" as popular sentiment. It inappropriately attaches this name to various passages of the Bhagavad Gita. Humour is nuanced and culturally bound, and understanding it would be like asking too much of their model at this point. Different translators chose different words to explain the same principles considering the nature of Sanskrit literature; in reality, Bhagavad Gita is a song with rhyme and rhythm, with the varying dates of the versions.
Their findings suggest the potential utility of AI-based tools for comparing translations and assessing sentiments in a variety of texts. This technique can also be used to analyse emotions portrayed in entertainment material. Another possible application is assessing films and songs to inform parents and officials about the appropriateness of content for youngsters.
Top 10 Crypto Data and Research Sites Investors Should Know
Google can Improve your Beauty in Virtual Meetings, Even When the Connection is Poor
Metaverse as a Service will Take Web3 Motive to Next Level
Only Cryptographic Algorithms Can Stop Quantum Computers from Data Stealing
Top 10 Machine Learning Skills that Land you in a High Paying Job
Enterprise Data Science can do Much More to Your Business than You Think
Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp
_____________
Disclaimer: Analytics Insight does not provide financial advice or guidance. Also note that the cryptocurrencies mentioned/listed on the website could potentially be scams, i.e. designed to induce you to invest financial resources that may be lost forever and not be recoverable once investments are made. You are responsible for conducting your own research (DYOR) before making any investments. Read more here.
