Book Image

The Supervised Learning Workshop - Second Edition

By : Blaine Bateman, Ashish Ranjan Jha, Benjamin Johnston, Ishita Mathur
Book Image

The Supervised Learning Workshop - Second Edition

By: Blaine Bateman, Ashish Ranjan Jha, Benjamin Johnston, Ishita Mathur

Overview of this book

Would you like to understand how and why machine learning techniques and data analytics are spearheading enterprises globally? From analyzing bioinformatics to predicting climate change, machine learning plays an increasingly pivotal role in our society. Although the real-world applications may seem complex, this book simplifies supervised learning for beginners with a step-by-step interactive approach. Working with real-time datasets, you’ll learn how supervised learning, when used with Python, can produce efficient predictive models. Starting with the fundamentals of supervised learning, you’ll quickly move to understand how to automate manual tasks and the process of assessing date using Jupyter and Python libraries like pandas. Next, you’ll use data exploration and visualization techniques to develop powerful supervised learning models, before understanding how to distinguish variables and represent their relationships using scatter plots, heatmaps, and box plots. After using regression and classification models on real-time datasets to predict future outcomes, you’ll grasp advanced ensemble techniques such as boosting and random forests. Finally, you’ll learn the importance of model evaluation in supervised learning and study metrics to evaluate regression and classification tasks. By the end of this book, you’ll have the skills you need to work on your real-life supervised learning Python projects.
Table of Contents (9 chapters)

Boosting

The second ensemble technique we'll be looking at is boosting, which involves incrementally training new models that focus on the misclassified data points in the previous model and utilizes weighted averages to turn weak models (underfitting models having a high bias) into stronger models. Unlike bagging, where each base estimator could be trained independently of the others, the training of each base estimator in a boosted algorithm depends on the previous one.

Although boosting also uses the concept of bootstrapping, it's done differently from bagging, since each sample of data is weighted, implying that some bootstrapped samples can be used for training more often than other samples. When training each model, the algorithm keeps track of which features are most useful and which data samples have the most prediction error; these are given higher weightage and are considered to require more iterations to properly train the model.

When predicting the output...