Practical Big Data Analytics

Cover
Copyright and Credits
Packt Upsell
Contributors
Table of Contents
Preface
Chapter 1: Too Big or Not Too Big
	What is big data?
		A brief history of data
			Dawn of the information age
			Dr. Alan Turing and modern computing
			The advent of the stored-program computer
			From magnetic devices to SSDs
	Why we are talking about big data now if data has always existed
		Definition of big data
			Building blocks of big data analytics
	Types of Big Data
		Structured
		Unstructured
		Semi-structured
	Sources of big data
		The 4Vs of big data
	When do you know you have a big data problem and where do you start your search for the big data solution?
	Summary
Chapter 2: Big Data Mining for the Masses
	What is big data mining?
		Big data mining in the enterprise
			Building the case for a Big Data strategy
			Implementation life cycle
			Stakeholders of the solution
			Implementing the solution
	Technical elements of the big data platform
		Selection of the hardware stack
		Selection of the software stack
	Summary
Chapter 3: The Analytics Toolkit
	Components of the Analytics Toolkit
	System recommendations
		Installing on a laptop or workstation
		Installing on the cloud
	Installing Hadoop
		Installing Oracle VirtualBox
		Installing CDH in other environments
	Installing Packt Data Science Box
	Installing Spark
	Installing R
		Steps for downloading and installing Microsoft R Open
	Installing RStudio
	Installing Python
	Summary
Chapter 4: Big Data With Hadoop
	The fundamentals of Hadoop
		The fundamental premise of Hadoop
		The core modules of Hadoop
			Hadoop Distributed File System - HDFS
			Data storage process in HDFS
		Hadoop MapReduce
			An intuitive introduction to MapReduce
			A technical understanding of MapReduce
			Block size and number of mappers and reducers
		Hadoop YARN
			Job scheduling in YARN
			Other topics in Hadoop
				Encryption
				User authentication
				Hadoop data storage formats
			New features expected in Hadoop 3
	The Hadoop ecosystem
	Hands-on with CDH
		WordCount using Hadoop MapReduce
		Analyzing oil import prices with Hive
			Joining tables in Hive
	Summary
Chapter 5: Big Data Mining with NoSQL
	Why NoSQL?
		The ACID, BASE, and CAP properties
			ACID and SQL
			The BASE property of NoSQL
			The CAP theorem
		The need for NoSQL technologies
			Google Bigtable
			Amazon Dynamo
	NoSQL databases
		In-memory databases
		Columnar databases
		Document-oriented databases
		Key-value databases
		Graph databases
		Other NoSQL types and summary of other types of databases 
	Analyzing Nobel Laureates data with MongoDB
		JSON format
		Installing and using MongoDB
	Tracking physician payments with real-world data
		Installing kdb+, R, and RStudio
			Installing kdb+
			Installing R
			Installing RStudio
	The CMS Open Payments Portal
		Downloading the CMS Open Payments data
		Creating the Q application
			Loading the data
			The backend code
		Creating the frontend web portal
	R Shiny platform for developers
		Putting it all together - The CMS Open Payments application
		Applications
	Summary
Chapter 6: Spark for Big Data Analytics
	The advent of Spark
		Limitations of Hadoop
		Overcoming the limitations of Hadoop
		Theoretical concepts in Spark
			Resilient distributed datasets
			Directed acyclic graphs
			SparkContext
			Spark DataFrames
			Actions and transformations
			Spark deployment options
			Spark APIs
		Core components in Spark
			Spark Core
			Spark SQL
			Spark Streaming
			GraphX
			MLlib
		The architecture of Spark
		Spark solutions
	Spark practicals
		Signing up for Databricks Community Edition
	Spark exercise - hands-on with Spark (Databricks)
	Summary
Chapter 7: An Introduction to Machine Learning Concepts
	What is machine learning?
		The evolution of machine learning
	Factors that led to the success of machine learning
	Machine learning, statistics, and AI
	Categories of machine learning
		Supervised and unsupervised machine learning
			Supervised machine learning
				Vehicle Mileage, Number Recognition and other examples
			Unsupervised machine learning
	Subdividing supervised machine learning
	Common terminologies in machine learning
	The core concepts in machine learning
		Data management steps in machine learning
			Pre-processing and feature selection techniques
				Centering and scaling
			The near-zero variance function
			Removing correlated variables
			Other common data transformations
			Data sampling
			Data imputation
			The importance of variables
		The train, test splits, and cross-validation concepts
			Splitting the data into train and test sets
			The cross-validation parameter
				Creating the model
	Leveraging multicore processing in the model
	Summary
Chapter 8: Machine Learning Deep Dive
	The bias, variance, and regularization properties
	The gradient descent and VC Dimension theories
	Popular machine learning algorithms
		Regression models
		Association rules
			Confidence
			Support
			Lift
		Decision trees
		The Random forest extension
		Boosting algorithms
		Support vector machines
		The K-Means machine learning technique
		The neural networks related algorithms
	Tutorial - associative rules mining with CMS data
		Downloading the data
		Writing the R code for Apriori
		Shiny (R Code)
		Using custom CSS and fonts for the application
		Running the application
	Summary
Chapter 9: Enterprise Data Science
	Enterprise data science overview
	A roadmap to enterprise analytics success
	Data science solutions in the enterprise
		Enterprise data warehouse and data mining
		Traditional data warehouse systems
			Oracle Exadata, Exalytics, and TimesTen
			HP Vertica
			Teradata
			IBM data warehouse systems (formerly Netezza appliances)
			PostgreSQL
			Greenplum
			SAP Hana
		Enterprise and open source NoSQL Databases
			Kdb+
			MongoDB
			Cassandra
			Neo4j
		Cloud databases
			Amazon Redshift, Redshift Spectrum, and Athena databases
			Google BigQuery and other cloud services
			Azure CosmosDB
		GPU databases
			Brytlyt
			MapD
		Other common databases
	Enterprise data science – machine learning and AI
		The R programming language
		Python
		OpenCV, Caffe, and others
		Spark
		Deep learning
		H2O and Driverless AI
		Datarobot
		Command-line tools
		Apache MADlib
		Machine learning as a service
	Enterprise infrastructure solutions
		Cloud computing
		Virtualization
		Containers – Docker, Kubernetes, and Mesos
		On-premises hardware
		Enterprise Big Data
	Tutorial – using RStudio in the cloud
	Summary
Chapter 10: Closing Thoughts on Big Data
	Corporate big data and data science strategy
	Ethical considerations
	Silicon Valley and data science
	The human factor
		Characteristics of successful projects
	Summary
Appendix: External Data Science Resources
	Big data resources
	NoSQL products
	Languages and tools
	Creating dashboards
	Notebooks
	Visualization libraries
	Courses on R
	Courses on machine learning
	Machine learning and deep learning links
	Web-based machine learning services
	Movies
	Machine learning books from Packt
	Books for leisure reading
Other Books You May Enjoy
	Leave a review - let other readers know what you think
Index