This dataset contains 50,000 movie reviews collected from the Internet Movie Database (IMDb) platform and is intended for natural language processing (NLP), sentiment analysis, and machine learning research. Each review is labelled with a binary sentiment category: positive or negative. The dataset is structured in CSV format with the following attributes: review – textual movie review contentsentiment – sentiment label (positive or negative) The dataset can be used for: Sentiment classificationText mining and preprocessingDeep learning and NLP model trainingOpinion mining researchBenchmarking supervised learning algorithms This dataset is suitable for academic research, educational projects, and experimental studies involving machine learning, artificial intelligence, and computational linguistics.