This chapter focuses on uncertain transactional databases (UTDB), which represent collections of transactions containing items with associated probabilities, reflecting the inherent uncertainty in real-world data. We formally define uncertain transactions, patterns, and their expected support in the context of uncertain data. The chapter also covers techniques for creating synthetic uncertain transactional databases, converting structured data into UTDBs, and deriving statistical details to understand the data’s characteristics better. A significant portion of the chapter is dedicated to the challenges of frequent pattern discovery, specifically addressing the limitations of the downward closure property in uncertain data and introducing algorithms like TUBE-P for efficient pattern mining. Practical Python code examples demonstrate how these methods can be implemented to analyze uncertain transactional data.