Data-driven innovation is a developmental approach based on the foundations of Data Mining and Knowledge Discovery (KDD), which develops meaningful insights and trends within massive and complex datasets. The importance of scalable and intelligent data mining processes cannot be overestimated in the context of all areas, such as healthcare, finance, education, and cybersecurity, with the constantly growing volume, variety, and speed of data. In this paper, the basics and sophisticated methods in data mining are addressed, such as machine learning, deep learning, and hybrid approaches, preprocessing, evaluation, and model explainability. This discussion addresses critical challenges, including data quality, scalability, and privacy, and also focuses on the present and future applications of KDD. Also, the paper outlines the trends in the future of automated discovery, ethical AI, and real-time data stream mining. This paper provides a current review of the dynamic nature of data mining and knowledge discovery and summarizes the new literature.