Book Image

Applied Unsupervised Learning with Python

By : Benjamin Johnston, Aaron Jones, Christopher Kruger
Book Image

Applied Unsupervised Learning with Python

By: Benjamin Johnston, Aaron Jones, Christopher Kruger

Overview of this book

Unsupervised learning is a useful and practical solution in situations where labeled data is not available. Applied Unsupervised Learning with Python guides you in learning the best practices for using unsupervised learning techniques in tandem with Python libraries and extracting meaningful information from unstructured data. The book begins by explaining how basic clustering works to find similar data points in a set. Once you are well-versed with the k-means algorithm and how it operates, you’ll learn what dimensionality reduction is and where to apply it. As you progress, you’ll learn various neural network techniques and how they can improve your model. While studying the applications of unsupervised learning, you will also understand how to mine topics that are trending on Twitter and Facebook and build a news recommendation engine for users. Finally, you will be able to put your knowledge to work through interesting activities such as performing a Market Basket Analysis and identifying relationships between different products. By the end of this book, you will have the skills you need to confidently build your own models using Python.
Table of Contents (12 chapters)
Applied Unsupervised Learning with Python

Activity 1: Implementing k-means Clustering

Scenario: You are asked in an interview to implement a k-means clustering algorithm from scratch to prove that you understand how it works. We will be using the Iris dataset provided by the UCI ML repository. The Iris dataset is a classic in the data science world and has features that are used to predict Iris species. The download location can be found later in this activity.

For this activity, you are able to use Matplotlib, NumPy, scikit-learn metrics, and pandas.

By loading and reshaping data easily, you can focus more on learning k-means instead of writing dataloader functionality.

Iris data columns are provided as follows for reference:

['SepalLengthCm', 'SepalWidthCm', 'PetalLengthCm', 'PetalWidthCm', 'species']

Aim: To truly understand how something works, you need to build it from scratch. Take what you have learned in the previous sections and implement k-means from scratch in Python.

Please open your favorite editing platform and try the following:

  1. Using NumPy or the math package and the Euclidean distance formula and write a function that calculates the distance between two coordinates.

  2. Write a function that calculates the distance from centroids to each of the points in your dataset and returns the cluster membership.

  3. Write a k-means function that takes in a dataset and the number of clusters (K) and returns the final cluster centroids, as well as the data points that make up that cluster's membership. After implementing k-means from scratch, apply your custom algorithm to the Iris dataset, located here:


    This dataset was downloaded from It can be accessed at

    UCI Machine Learning Repository []. Irvine, CA: University of California, School of Information and Computer Science.

  4. Remove the classes supplied in this dataset and see if your k-means algorithm can group the different Iris species into their proper groups just based on plant characteristics!

  5. Calculate the Silhouette Score using the scikit-learn implementation.

Outcome: By completing this exercise, you will gain hands-on experience of tuning a k-means clustering algorithm for a real-world dataset. The Iris dataset is seen as a classic "hello world" type problem in the data science space and is helpful for testing foundational techniques on. Your final clustering algorithm should do a decent job of finding the three clusters of Iris species types that exist in the data, as follows:

Figure 1.21: Expected plot of three clusters of Iris species


The solution for this activity can be found on page 306.