A comprehensive analysis and sensitivity model for emotion detection in text
Pancham Singh, Amita Asthana, Beerbal Solanki, Swati Tomar, Mrignainy Kansal, Gagan Tyagi · 2025
People use a variety of communication channels, including text, voice, and gestures, to convey a wide range of emotions in their daily interactions. The problem of automatically identifying these emotions from textual information has important practical ramifications. This study focuses on natural language processing (NLP) methods for textual emotion identification, particularly in Twitter tweets. The study uses machine learning, sentiment analysis, tokenization, and other techniques with a focus on using APIs such as the Twitter API. With an emphasis on the use of unsupervised learning algorithms and data expansion methodologies, the focal issue examines relevant work in sentiment analysis on social media. The suggested approach tackles issues including tokenization, sentiment analysis, and noise reduction in order to categorize tweets into distinct moods. The front-end development using Flask provides a user-friendly interface to visualize real-time Twitter data based on user-defined keywords. The study achieved an overall accuracy rate of 92% in classifying human moods across a diverse dataset.