The rapid expansion of big data in modern computing environments has created unprecedented opportunities and challenges in data analysis and knowledge discovery. Traditional data mining techniques often fail to address the scalability and efficiency requirements of large-scale datasets characterized by high volume, velocity, and variety. This study presents an in-depth exploration of scalable and efficient data mining techniques for big data analytics, focusing on modern computational frameworks and advanced machine learning approaches. The integration of distributed computing, parallel processing, and hybrid machine learning models has significantly enhanced the ability to process and analyze large datasets in real time. Additionally, optimization strategies such as dimensionality reduction, feature selection, and model tuning have improved computational efficiency and predictive performance. The findings highlight that scalable data mining frameworks, including cloud-based and distributed architectures, play a critical role in enabling efficient big data analytics. However, challenges related to data heterogeneity, computational complexity, and data privacy remain significant concerns. This study emphasizes the need for adaptive, efficient, and scalable solutions to address the evolving demands of big data environments and supports the development of intelligent systems capable of real-time decision-making.