Much of the data we encounter has an inherent order. Think of words in a sentence, stock prices over time, or sensor readings from a device. This type of information, where the sequence matters, is known as sequential data. Standard machine learning models often treat inputs independently, which isn't suitable when the order contains valuable information.

This first chapter establishes the groundwork for understanding and working with sequential data. We will look at:

• What constitutes sequential data, providing concrete examples like text and time series.
• The key properties that define sequences, such as temporal dependencies and variable lengths.
• Why traditional feedforward networks fall short for sequence processing tasks.
• Methods for converting raw sequential data into numerical formats that machine learning models can understand.
• An overview of typical problems addressed using sequence models, including prediction, classification, and generation.

By the end of this chapter, you will have a clear understanding of the nature of sequential data and why specialized approaches are required, setting the stage for learning about Recurrent Neural Networks in subsequent chapters.