Scholarlink

The goal of this study is to advance sentiment analysis for Marathi, a language that is not as well-represented in natural language processing (NLP). Sentiment analysis for widely spoken languages like English and Spanish has advanced significantly, but Marathi still faces difficulties because there aren’t enough resources and datasets available. In order to overcome these restrictions, this study uses machine learning (ML) and natural language processing (NLP) models created especially for analysing Marathi text on social media sites like Instagram, Twitter, and YouTube. This study’s addition of a correction mechanism to the sentiment analysis pipeline is a noteworthy breakthrough. The informal language of social media messages frequently contains errors that our method successfully corrects. Through the utilization of a customized dataset enhanced with specialized lexicons and root words, the suggested method produces accurate content summaries and enhances sentiment identification. Developing a strong sentiment analysis tool for Marathi fills a crucial need in the field of multilingual natural language processing (NLP) and opens the door for comparable developments in other regional languages.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Sentiment Analysis of Historical Narratives: A Comparative Study of ML and NLP Models on Indian Leaders’ Data

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sarthak Deshpande,
  • Akshay Patil,
  • Pradip Pandhare,
  • Nikhil Wankhede,
  • Rushali A. Deshmukh

摘要

The goal of this study is to advance sentiment analysis for Marathi, a language that is not as well-represented in natural language processing (NLP). Sentiment analysis for widely spoken languages like English and Spanish has advanced significantly, but Marathi still faces difficulties because there aren’t enough resources and datasets available. In order to overcome these restrictions, this study uses machine learning (ML) and natural language processing (NLP) models created especially for analysing Marathi text on social media sites like Instagram, Twitter, and YouTube. This study’s addition of a correction mechanism to the sentiment analysis pipeline is a noteworthy breakthrough. The informal language of social media messages frequently contains errors that our method successfully corrects. Through the utilization of a customized dataset enhanced with specialized lexicons and root words, the suggested method produces accurate content summaries and enhances sentiment identification. Developing a strong sentiment analysis tool for Marathi fills a crucial need in the field of multilingual natural language processing (NLP) and opens the door for comparable developments in other regional languages.