University Computing Project: Emotions Classification using Transformers (DistilBERT) and ML

Comprehensive University Report: Emotions Classification using Natural Language Processing (NLP) This in-depth university report details a practical implementation and comparative analysis for emotions classification in text data. It showcases two distinct yet powerful machine learning approaches: the cutting-edge DistilBERT transformer model (a deep learning model) and the foundational Logistic Regression algorithm. As part of my Computing studies at Aston University, this project provided extensive hands-on experience in: Natural Language Processing (NLP) techniques for text analysis on tweets. Deep Learning concepts with advanced transformer architectures (DistilBERT). Machine Learning fundamentals and model selection. Data preprocessing specifically for text-based datasets. Model training, evaluation, and performance comparison methodologies. Academic report writing for complex technical projects. What you'll find in this report: A clear problem definition and literature review on emotions classification. Detailed methodology covering data collection, preprocessing, and feature engineering. In-depth explanation and implementation details for both DistilBERT and Logistic Regression. A comprehensive analysis of model performance, including metrics, results, and comparative insights. Discussion of strengths, weaknesses, and potential future work for each approach. A well-structured and meticulously referenced example of a high-quality university-level Computing project report. Why this document is invaluable for other students: Excellent coursework example: Perfect for students tackling similar NLP, Machine Learning, Deep Learning, or Artificial Intelligence (AI) projects. Understand complex models: Gain practical insight into applying transformer models like DistilBERT. Compare different approaches: See a direct, side-by-side comparison of a deep learning model versus a classic ML algorithm. Save time: Use this as a reference for structuring your own reports, understanding key concepts, and identifying successful methodologies for text classification.