Scholarlink

The rapid expansion of data poses a significant challenge for analyzing sentiments. The importance of user-generated reviews highlights the need to carefully curate and evaluate text data to extract opinions. This research delves into the world of learning for sentiment analysis exploring models such as Bi-LSTM, CNN, and GRU along with various word embeddings like N-grams, Keras embedding, BERT, Roberta, CT-Bert, and Elmo. It effectively categorizes sentiments and detects sarcasm in Twitter and IMDB data making it a vital tool for applying learning to real-world sentiment analysis tasks.

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

Exploring Deep Learning and Word Embedding Techniques for Sentiment Analysis on Diverse Textual Data

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • B. N. Lavanya,
  • K. V. Anitha Rathnam,
  • K. Kiran,
  • Abhishek Appaji,
  • P. Deepa Shenoy,
  • K. R. Venugopal

摘要

The rapid expansion of data poses a significant challenge for analyzing sentiments. The importance of user-generated reviews highlights the need to carefully curate and evaluate text data to extract opinions. This research delves into the world of learning for sentiment analysis exploring models such as Bi-LSTM, CNN, and GRU along with various word embeddings like N-grams, Keras embedding, BERT, Roberta, CT-Bert, and Elmo. It effectively categorizes sentiments and detects sarcasm in Twitter and IMDB data making it a vital tool for applying learning to real-world sentiment analysis tasks.