Comparative evaluation assesses sentiment classification models across online review and tweet datasets, highlighting the utility of embedding fusion for polarity and sarcasm detection.
In todays changing landscape the abundance of unprocessed data plays a vital role. It is crucial to curate this dataset and conduct a thorough analysis of the sentiments it captures. The aim of this research study is to explore learning techniques such as Bi LSTM, CNN and GRU for sentiment analysis. It also involves an in-depth examination of datasets, including restaurant reviews from platforms like Yelp and tweets related to Covid 19. The study further investigates word embedding methods ranging from Word2Vec, Keras, Glove, N-gram to more complex approaches, like Bert, Roberta, CT-Bert, FastText and Elmo. The ultimate objective is to categorize data into positive, negative or neutral sentiments by evaluating their underlying polarity. Furthermore, the study delves into the aspect of detecting sarcasm in text and aims to evaluate the level of sarcasm in different instances.
No takes yet. Share an insight, caveat, or question.
N et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: