ShivamPandey Feb 23, 2021

This is an amazing book.Unsupervised Learning with Python contains comprehensive coverage of the mathematical foundations, algorithms, and practical implementations of unsupervised learning. This book provides machine learning approach to uncover patterns and trends in your data, and support sound strategic decisions for your business.The book bridges the gap between complex math and practical Python implementations. It involves Fundamental building blocks and concepts of unsupervised learning,Choosing the right algorithm for your problem & How to interpret the results of unsupervised learning.Market Basket Analysis & HotSpot Analysis were the stand outs for me. The concepts are explained in an easy to go smooth flowing pattern.I would love if the book could also provide hands on with code via video excercises. I would rate this book a 5 star & must read for individuals exploring the space of Unsupervised Machine Learning. Special call out to the author & publishers of the book. Thank you.

Amazon Verified review

Marleen Feb 19, 2021

The books starts basically right away with the first main topic: Clustering. Only a few pages of Supervised vs Unsupervised Learning serve as introduction. However, often times I find myself scanning through 3 chapters or so of intro or repetition before a book starts with what it actually is about, therefore this is a welcoming change and meant to be a positive feedback.Unsupervised Learning is known to be more difficult to implement and also to explain to e.g. Management, Audit etc. This book can definitely be used as a guide to understand various areas of unsupervised learning and within each area you will learn multiple methods that one could use as examples at work to explain a concept.I am fairly new to unsupervised learning and tend to use supervised learning as much as possible but this book definitely gives me a good base to try new things.I can see that for more advanced users this book might be limited and sometimes to high level, but for me this book has just the right depths. I also enjoyed the accompanying material and I liked that the books tried to put images wherever possible. Therefore I would rate it - for my use-cases - with 5 stars.

Amazon Verified review

Julian M. - DS and ML Advisor | MSc. [c] Feb 16, 2021

The book is written with a simple vocabulary, combining technical topics with accessible content for those inexperienced in Machine Learning. The examples are theoretical-practical and always written in an understandable and easily digestible format. When reading the algorithmic details, a continuous and coherent narration of the explained techniques is presented, always maintaining the general context of unsupervised learning.The theoretical framework begins with the explanation of well-known clustering techniques such as K-means, DBSCAN, and Hierarchical clustering. The deepening of the description of K-means allows extrapolating the theoretical aspects to other more complex procedures. Consecutively, dimensionality reduction techniques are detailed, emphasizing Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbor Embedding (t-SNE) as a visualization algorithm. Decomposition by eigenvector and eigenvalues is explained through Singular Value Decomposition. The Autoencoders topic explained after PCA has an understandable introduction about several activation functions and perceptron models, explaining Artificial Neural Networks (ANN) and Convolutional Neural Networks (CNN) in simple words with a theoretical grasp. Finally, the content is oriented towards (a) some components of natural language processing as an application example for techniques such as Topic Modeling, with emphasis on Latent Dirichlet Allocation (LDA); and (b) some components of inferential statistics such as the estimation of data distributions through Kernel Density Estimation and the use of Kernel functions for linear regression, among others.From a practical point of view, a notable aspect is the writers' ability to explain algorithmic methods from scratch and using Python libraries. Each technique is detailed with its practical component in Python, allowing to practice while the reader is learning the theory.My major concern is regarding the depth of the content. It is a book that allows one to grasp the general aspects of unsupervised learning as a workshop, as the title mentions. However, some Deep Learning topics, such as Convolutional Neural Networks, are not explained sufficiently rigorously. The interpretability and explainability of such models are partially ignored. If you are looking in the book for the depth of theory and open implementations, this book is not the place to be. Conversely, if looking for a tool that allows you to learn about applied unsupervised learning algorithms, implementing known techniques, and using existing libraries, the book is the right place for your learning. Although some code is incomplete for the sake of understanding and flow of the book, it can be found in the Github repository in more detail, which is referenced throughout the book.Overall, I enjoyed reading the book and highly recommend it. It covers a wide variety of unsupervised learning techniques. The balance between theory and practice is right for those looking to get a grasp and general overview of these Machine Learning topics.

Amazon Verified review

Darpan Jan 27, 2021

Pros:This book is written in etiquette and very attractive.Details and enough information about the methods used in unsupervised learning.This is the first time I am seeing you implement jupyter style code in each section and also output which is quite amazing.Every data scientist would like to know how this algorithm would help in real time applications where the actual data would come.I am glad that the author mentioned it.It's so amazing that visualization is also in the book which helps to understand algorithms statistically.Cons :One thing is it's length as I am reading the book chapter by chapter. It is becoming lengthy.I am not sure but every beginner has to learn each and every algorithm from scratch in terms of coding. If a person is not into the coding field, they would directly implement the library without understanding of how parameters are used.Less information about training and optimization.

Amazon Verified review

Amazon Customer Dec 18, 2020

I received this book as a sample to review and enjoyed reading through it. As the book points out, unsupervised learning techniques are often given less weight in ML texts which tend to favor supervised methods. It’s refreshing to see an entire book devoted to unsupervised learning.The book covers a wide variety of topics, starting from well known clustering algorithms (K-means, hierarchical clustering, DBSCAN) before moving into more advanced topics (autoencoders). It also covers a few dimensionality reduction techniques, as well as going into market basket and hotspot analysis. For each topic covered, the book gives multiple examples and presents some exercises for the reader (complete with solutions). The book covers a lot of topics, and I suspect that most readers will learn something from reading it - i definitely did.My main complaints with the book were that it’s treatment of theory behind some of the methods are a little hand-wavey. In many cases, the author gives a formula, but doesn’t go into much depth explaining why this formula is the correct one. When reading it, I often found myself consulting other sources for a more detailed explanation. At the same time, some of the code examples given were redundant; the balance between why and how felt a little off. In addition, there are several instances where the code given in the book is incomplete - only the first 20 or so lines are given - the reader needs to consult the books git repo to view the entire code, but this isn’t explicitly states in situ. This can be confusing the first few times one encounters it.Overall, I enjoyed the book and would recommend it. It covered a wide range of unsupervised learning topics and gave several good examples of each - complete with visualizations. While the book definitely favors “practice” over “theory,” there is enough theory to give the reader a general idea of what’s going on, which is probably sufficient for a book that covers so many techniques. The focus specifically on unsupervised learning fills an important niche making this book a good addition to any machine learning library.

Amazon Verified review

The Unsupervised Learning Workshop: Get started with unsupervised learning algorithms and simplify your unorganized data to help make future predictions

What do you get with eBook?

Key benefits

Description

Who is this book for?

What you will learn

Product Details

What do you get with eBook?

Product Details

Frequently bought together

Recommendations for you

Customer reviews

Filter reviews by

People who bought this also bought

About the 3 authors

FAQs

The Unsupervised Learning Workshop: Get started with unsupervised learning algorithms and simplify your unorganized data to help make future predictions

What do you get with eBook?

Contact Details

Billing Address

Key benefits

Description

Who is this book for?

What you will learn

Product Details

What do you get with eBook?

Contact Details

Billing Address

Product Details

Packt Subscriptions

Frequently bought together

Recommendations for you

Customer reviews

Filter reviews by

People who bought this also bought

About the 3 authors

FAQs