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

Description

Learning how to apply unsupervised algorithms on unlabeled datasets from scratch can be easier than you thought with this beginner's workshop, featuring interesting examples and activities Key Features Get familiar with the ecosystem of unsupervised algorithms Learn interesting methods to simplify large amounts of unorganized data Tackle real-world challenges, such as estimating the population density of a geographical areaBook Description Do you find it difficult to understand how popular companies like WhatsApp and Amazon find valuable insights from large amounts of unorganized data? The Unsupervised Learning Workshop will give you the confidence to deal with cluttered and unlabeled datasets, using unsupervised algorithms in an easy and interactive manner. The book starts by introducing the most popular clustering algorithms of unsupervised learning. You'll find out how hierarchical clustering differs from k-means, along with understanding how to apply DBSCAN to highly complex and noisy data. Moving ahead, you'll use autoencoders for efficient data encoding. As you progress, you'll use t-SNE models to extract high-dimensional information into a lower dimension for better visualization, in addition to working with topic modeling for implementing natural language processing (NLP). In later chapters, you'll find key relationships between customers and businesses using Market Basket Analysis, before going on to use Hotspot Analysis for estimating the population density of an area. By the end of this book, you'll be equipped with the skills you need to apply unsupervised algorithms on cluttered datasets to find useful patterns and insights. What you will learn Distinguish between hierarchical clustering and the k-means algorithm Understand the process of finding clusters in data Grasp interesting techniques to reduce the size of data Use autoencoders to decode data Extract text from a large collection of documents using topic modeling Create a bag-of-words model using the CountVectorizerWho this book is for If you are a data scientist who is just getting started and want to learn how to implement machine learning algorithms to build predictive models, then this book is for you. To expedite the learning process, a solid understanding of the Python programming language is recommended, as you'll be editing classes and functions instead of creating them from scratch. Table of Contents Introduction to Clustering Hierarchical Clustering Neighborhood Approaches and DBSCAN Dimensionality Reduction Techniques and PCA Autoencoders t-Distributed Stochastic Neighbor Embedding Topic Modeling Market Basket Analysis Hotspot Analysis Cover FM Copyright Table of Contents Preface Chapter 1: Introduction to Clustering Introduction Unsupervised Learning versus Supervised Learning Clustering Identifying Clusters Two-Dimensional Data Exercise 1.01: Identifying Clusters in Data Introduction to k-means Clustering No-Math k-means Walkthrough K-means Clustering In-Depth Walkthrough Alternative Distance Metric – Manhattan Distance Deeper Dimensions Exercise 1.02: Calculating Euclidean Distance in Python Exercise 1.03: Forming Clusters with the Notion of Distance Exercise 1.04: K-means from Scratch – Part 1: Data Generation Exercise 1.05: K-means from Scratch – Part 2: Implementing k-means Clustering Performance – Silhouette Score Exercise 1.06: Calculating the Silhouette Score Activity 1.01: Implementing k-means Clustering Summary Chapter 2: Hierarchical Clustering Introduction Clustering Refresher The k-means Refresher The Organization of the Hierarchy Introduction to Hierarchical Clustering Steps to Perform Hierarchical Clustering An Example Walkthrough of Hierarchical Clustering Exercise 2.01: Building a Hierarchy Linkage Exercise 2.02: Applying Linkage Criteria Agglomerative versus Divisive Clustering Exercise 2.03: Implementing Agglomerative Clustering with scikit-learn Activity 2.01: Comparing k-means with Hierarchical Clustering k-means versus Hierarchical Clustering Summary Chapter 3: Neighborhood Approaches and DBSCAN Introduction Clusters as Neighborhoods Introduction to DBSCAN DBSCAN in Detail Walkthrough of the DBSCAN Algorithm Exercise 3.01: Evaluating the Impact of Neighborhood Radius Size DBSCAN Attributes – Neighborhood Radius Activity 3.01: Implementing DBSCAN from Scratch DBSCAN Attributes – Minimum Points Exercise 3.02: Evaluating the Impact of the Minimum Points Threshold Activity 3.02: Comparing DBSCAN with k-means and Hierarchical Clustering DBSCAN versus k-means and Hierarchical Clustering Summary Chapter 4: Dimensionality Reduction Techniques and PCA Introduction What Is Dimensionality Reduction? Applications of Dimensionality Reduction The Curse of Dimensionality Overview of Dimensionality Reduction Techniques Dimensionality Reduction Principal Component Analysis Mean Standard Deviation Covariance Covariance Matrix Exercise 4.01: Computing Mean, Standard Deviation, and Variance Using the pandas Library Eigenvalues and Eigenvectors Exercise 4.02: Computing Eigenvalues and Eigenvectors The Process of PCA Exercise 4.03: Manually Executing PCA Exercise 4.04: scikit-learn PCA Activity 4.01: Manual PCA versus scikit-learn Restoring the Compressed Dataset Exercise 4.05: Visualizing Variance Reduction with Manual PCA Exercise 4.06: Visualizing Variance Reduction with scikit-learn Exercise 4.07: Plotting 3D Plots in Matplotlib Activity 4.02: PCA Using the Expanded Seeds Dataset Summary Chapter 5: Autoencoders Introduction Fundamentals of Artificial Neural Networks The Neuron The Sigmoid Function Rectified Linear Unit (ReLU) Exercise 5.01: Modeling the Neurons of an Artificial Neural Network Exercise 5.02: Modeling Neurons with the ReLU Activation Function Neural Networks: Architecture Definition Exercise 5.03: Defining a Keras Model Neural Networks: Training Exercise 5.04: Training a Keras Neural Network Model Activity 5.01: The MNIST Neural Network Autoencoders Exercise 5.05: Simple Autoencoder Activity 5.02: Simple MNIST Autoencoder Exercise 5.06: Multi-Layer Autoencoder Convolutional Neural Networks Exercise 5.07: Convolutional Autoencoder Activity 5.03: MNIST Convolutional Autoencoder Summary Chapter 6: t-Distributed Stochastic Neighbor Embedding Introduction The MNIST Dataset Stochastic Neighbor Embedding (SNE) t-Distributed SNE Exercise 6.01: t-SNE MNIST Activity 6.01: Wine t-SNE Interpreting t-SNE Plots Perplexity Exercise 6.02: t-SNE MNIST and Perplexity Activity 6.02: t-SNE Wine and Perplexity Iterations Exercise 6.03: t-SNE MNIST and Iterations Activity 6.03: t-SNE Wine and Iterations Final Thoughts on Visualizations Summary Chapter 7: Topic Modeling Introduction Topic Models Exercise 7.01: Setting up the Environment A High-Level Overview of Topic Models Business Applications Exercise 7.02: Data Loading Cleaning Text Data Data Cleaning Techniques Exercise 7.03: Cleaning Data Step by Step Exercise 7.04: Complete Data Cleaning Activity 7.01: Loading and Cleaning Twitter Data Latent Dirichlet Allocation Variational Inference Bag of Words Exercise 7.05: Creating a Bag-of-Words Model Using the Count Vectorizer Perplexity Exercise 7.06: Selecting the Number of Topics Exercise 7.07: Running LDA Visualization Exercise 7.08: Visualizing LDA Exercise 7.09: Trying Four Topics Activity 7.02: LDA and Health Tweets Exercise 7.10: Creating a Bag-of-Words Model Using TF-IDF Non-Negative Matrix Factorization The Frobenius Norm The Multiplicative Update Algorithm Exercise 7.11: Non-negative Matrix Factorization Exercise 7.12: Visualizing NMF Activity 7.03: Non-negative Matrix Factorization Summary Chapter 8: Market Basket Analysis Introduction Market Basket Analysis Use Cases Important Probabilistic Metrics Exercise 8.01: Creating Sample Transaction Data Support Confidence Lift and Leverage Conviction Exercise 8.02: Computing Metrics Characteristics of Transaction Data Exercise 8.03: Loading Data Data Cleaning and Formatting Exercise 8.04: Data Cleaning and Formatting Data Encoding Exercise 8.05: Data Encoding Activity 8.01: Loading and Preparing Full Online Retail Data The Apriori Algorithm Computational Fixes Exercise 8.06: Executing the Apriori Algorithm Activity 8.02: Running the Apriori Algorithm on the Complete Online Retail Dataset Association Rules Exercise 8.07: Deriving Association Rules Activity 8.03: Finding the Association Rules on the Complete Online Retail Dataset Summary Chapter 9: Hotspot Analysis Introduction Spatial Statistics Probability Density Functions Using Hotspot Analysis in Business Kernel Density Estimation The Bandwidth Value Exercise 9.01: The Effect of the Bandwidth Value Selecting the Optimal Bandwidth Exercise 9.02: Selecting the Optimal Bandwidth Using Grid Search Kernel Functions Exercise 9.03: The Effect of the Kernel Function Kernel Density Estimation Derivation Exercise 9.04: Simulating the Derivation of Kernel Density Estimation Activity 9.01: Estimating Density in One Dimension Hotspot Analysis Exercise 9.05: Loading Data and Modeling with Seaborn Exercise 9.06: Working with Basemaps Activity 9.02: Analyzing Crime in London Summary Appendix Index