• Homa Rizvi
• Farheen Siddiqui
• Dr. Nikhat Akhtar
• Shamim Ahmad
• Dr. Yusuf Perwej

The increasing impact of artificial intelligence in natural language processing (NLP) has rendered emotion identification from textual data an essential element in applications including sentiment analysis, customer feedback assessment, mental health evaluation, and human-computer interaction. Social networking platforms have become a vital medium for expressing emotions globally owing to the fast proliferation of the Internet age. Numerous individuals use written information, images, audio, and video to convey their emotions or perspectives. Conversely, text communication via web-based networking platforms is quite overwhelming. A substantial volume of unstructured data is produced on the Internet every second as a result of social media platforms. The data must be analyzed as swiftly as it is created to understand human psychology, achievable via sentiment analysis, which identifies polarity in words. Conventional models often fail to accurately represent the complex semantic relationships and emotional nuances inherent in textual and emoji communication. Our strategy utilizes BERT's capacity to store profound contextual links and BiLSTM's proficiency in modeling sequential patterns, therefore improving emotion identification accuracy. We evaluate our approach using two reputable benchmark Yelp datasets, which constitute a substantial text emotion identification corpus of 30,000 phrases. The text-based Yelp dataset allows the model to acquire intricate linguistic patterns and emotional signals inherent in written language. By incorporating these varied modalities, our method promotes a more thorough and flexible emotion categorization system. The integration of contextual embeddings and sequential learning significantly enhances robustness, making emotion recognition more accurate and dependable.

Emotion Recognition, Feature Extraction, Long Short-Term Memory (LSTM), BERT, Deep Learning, NLP, BiLSTM, Sentiment Analysis, Yelp Dataset.

" An NLP-Driven Approach to Improved Emotion Recognition in Textual Data Utilizing the Yelp Dataset", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 10, page no.a354-a360, October-2025, Available :http://www.jetir.org/papers/JETIR2510045.pdf

" An NLP-Driven Approach to Improved Emotion Recognition in Textual Data Utilizing the Yelp Dataset", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 10, page no. ppa354-a360, October-2025, Available at : http://www.jetir.org/papers/JETIR2510045.pdf