Identify where to apply unsupervised machine learning and how to do it in Python. Find meaning in unlabeled data through Python-based unsupervised machine learning.Key FeaturesConcrete examples that can jump-start the learner to solve a real problem (work-related, for example) right away.Help with implementation in Python -- libraries, best practices.An understanding of what kinds of problems can be solved by which unsupervised learning algorithm.Book DescriptionUnsupervised learning is useful and practical in situations where labelled data is not available. Unsupervised Learning with Python shows you 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. You will learn in detail various clustering methods, such as K-means, hierarchical clustering, and DBSCAN, and build algorithms from scratch using these methods. Then, you will learn about dimensionality reduction and its applications. You will also learn Principal Component Analysis (PCA), t-distributed Stochastic Neighbor Embedding (t-SNE) and Autoencoders in detail, and learn their implementations and their shortcomings. You will use sklearn to implement and analyse PCA on the Iris dataset. Then, you will use Keras to build autoencoder models for the CIFAR 10 dataset and visualize them using t-SNE. While studying the applications of unsupervised learning, you will explore mine trending topics in twitter or facebook, and build a news recommendation engine for readers.You will complete the book with several interesting activities, such as performing a market basket analysis and finding relations between different merchandises, using hotspot discovery and KDE algorithm to analyze crime data in London for this effort, and using the Apriori algorithm to study transaction data.What you will learnLearn where to apply unsupervised learning, with practical examples.Learn which unsupervised learning algorithm to apply to what problem.Gain practical experience solving various unsupervised learning problems in Python. Who This Book Is ForAudience should already know Python and wants to gain some hands-on experience in using Python for unsupervised machine learning. Our ideal audience would also have some experience with machine learning generally, as otherwise they may benefit from a more general course. Learning unsupervised machine learning will help our audience find structure in their data that they may not know exist.
Author: Eugene Chen