Data Types and Analytical Workflows

The Data Science Lifecycle (DSLC) describes research at a broad level, but day-to-day analysis depends on two essentials:
understanding data types and following a clear workflow.

Data Types

Every dataset contains different kinds of values called data types, which determine how information can be processed.
The most common ones are:

• Numeric – numbers that can be measured or counted (e.g., years, quantities).
  
• Text (strings) – words or characters (e.g., names, titles).
  
• Categorical – labels or groups (e.g., genre, material type).
  
• Boolean – true or false values (e.g., published or not).
  
• Date and time – temporal information (e.g., publication year).

Python handles these through its main data libraries, such as, NumPy (numeric data), Pandas (all data types in tables), and Matplotlib (for visualizing them).

Analytical Workflow

Once data types are understood, analysis follows four main steps:

Import → Clean → Analyze → Visualize

1. Import – Load data from files, databases, or web sources using tools like pandas.read_csv().
   
2. Clean – Fix errors, handle missing values, and ensure data types are correct.
   
3. Analyze – Explore patterns and relationships using NumPy and Pandas.
   
4. Visualize – Present results clearly with charts and plots using Matplotlib.

This process is iterative — each step can lead back to another as new questions arise.
It captures the dynamic rhythm of working with data: importing, refining, exploring, and communicating insights.

Chapter exercisesWorked answers after an attempt