Sentiment Analysis Using Natural Language Processing

This project presents the development of a sentiment analysis dashboard to analyse and visualize player reviews from platforms like Steam. The dashboard leverages advanced natural language processing (NLP) techniques such as BERT for review sentiment classification and InstructABSA for aspect-based sentiment analysis (ABSA), fine-tuned with datasets of Steam reviews. The dashboard provides insights through data visualization features such as sentiment gauges, word clouds, and aspect-based sentiment trends. These tools empower developers to identify areas for improvement, while helping players make informed decisions about game purchases. Despite challenges like computational demands and linguistic complexities (e.g., sarcasm detection), the project demonstrates the potential of sentiment analysis in enhancing decision-making and the importance of a player-driven approach to game development. Future improvements could include database integration for historical analysis and optimization for scalability.