Data science is the art of finding the story within data: patterns, relationships, and rules that help you explain what’s happening and predict what might happen next. This comes up everywhere—from biology and medicine to economics and business. Real datasets are usually messy, incomplete, and noisy, so data science is really a problem-solving game: ask the right question, choose a model, test your idea, and check whether your conclusion actually holds water.
Modern data science is tightly connected to AI. Many AI systems learn from large datasets to sort things into categories, make predictions, and produce useful output.
In this course, we’ll both learn the concepts and get our hands dirty with real tools. We will:
- Learn the core Python toolkit for data analysis, create data visualizations, and solve “data-scientist style” problems where you spot patterns and make predictions.
- Then we move into modern AI by building a neural network and learning what’s happening inside it: layers, weights, activation functions, loss functions, and training with gradient-based methods.
We’ll treat the math conceptually—enough to understand what’s going on and use it well, without long technical derivations. - If time allows, we’ll finish up with an introduction to computer vision and explore a neural network trained on a classic image dataset.
Prerequisites: Python coding—comfortable with conditionals, loops, and lists.