Machine Learning Quick Reference

Quick and essential machine learning hacks for training smart data models

Rahul Kumar

• Publisher: Packt Publishing
• Year: 2019
• ISBN:
  9781788831611
• Edition: 1
• Topic: Computer Science
• Subtopic: Artificial Intelligence (AI) & Semantics
• Index: Computer Science

Your hands-on reference guide to developing, training, and optimizing your machine learning models

Key Features

• Your guide to learning efficient machine learning processes from scratch
• Explore expert techniques and hacks for a variety of machine learning concepts
• Write effective code in R, Python, Scala, and Spark to solve all your machine learning problems

Book Description

Machine learning makes it possible to learn about the unknowns and gain hidden insights into your datasets by mastering many tools and techniques. This book guides you to do just that in a very compact manner.

After giving a quick overview of what machine learning is all about, Machine Learning Quick Reference jumps right into its core algorithms and demonstrates how they can be applied to real-world scenarios. From model evaluation to optimizing their performance, this book will introduce you to the best practices in machine learning. Furthermore, you will also look at the more advanced aspects such as training neural networks and work with different kinds of data, such as text, time-series, and sequential data. Advanced methods and techniques such as causal inference, deep Gaussian processes, and more are also covered.

By the end of this book, you will be able to train fast, accurate machine learning models at your fingertips, which you can easily use as a point of reference.

What you will learn

• Get a quick rundown of model selection, statistical modeling, and cross-validation
• Choose the best machine learning algorithm to solve your problem
• Explore kernel learning, neural networks, and time-series analysis
• Train deep learning models and optimize them for maximum performance
• Briefly cover Bayesian techniques and sentiment analysis in your NLP solution
• Implement probabilistic graphical models and causal inferences
• Measure and optimize the performance of your machine learning models

Who this book is for

If you're a machine learning practitioner, data scientist, machine learning developer, or engineer, this book will serve as a reference point in building machine learning solutions. You will also find this book useful if you're an intermediate machine learning developer or data scientist looking for a quick, handy reference to all the concepts of machine learning. You'll need some exposure to machine learning to get the best out of this book.

Natural Language Processing

How fast has the world been changing? Well, technology and data have been changing just as quickly. With the advent of the internet and social media, our entire outlook on data has changed. Initially, the scope of most data analytics revolved around structured data. However, due to so much unstructured data being pumped in through the internet and social media, the spectrum of analytics has broadened. Large amounts of text data, images, sound, and video data are being generated every second. They contain lots of information that needs to be synthesized for business. Natural language processing is a technique through which we enable a machine to understand text or speech. Although unstructured data has a wide range, the scope of this chapter will be to expose you to text analytics.

Structured data is typically made up of fixed observations and fixed columns set up in relational databases or in a spreadsheet, whereas unstructured data doesn't have any structure, and it can't be set up in a relational database; rather, it needs a NoSQL database, example, video, text, and so on.

In this chapter, you will learn about the following topics:

• The document term matrix
• Different approaches to looking at text
• Sentiment analysis
• Topic modeling
• The Bayesian technique

Text corpus

A text corpus is text data that forms out of a single document or group of documents and can come from any language, such as English, German, Hindi, and so on. In today's world, most of the textual data flows from social media, such as Facebook, Twitter, blogging sites, and other platforms. Mobile applications have now been added to the list of such sources. The larger size of a corpus, which is called corpora, makes the analytics more accurate.

Sentences

A corpus can be broken into units, which are called sentences. Sentences hold the meaning and context of the corpus, once we combine them together. Sentence formation takes place with the help of parts of speech. Every sentence is separated from other sentences by a delimiter, such as a period, which we can make use of to break it up further. This is called sentence tokenization.

Words

Words are the smallest unit of corpuses and take the shape of sentences when we put them in order by following the parts of speech. When we break down the sentences into words, it is called word tokenization.

Bags of words

When we have text as input data, we can't go ahead and work with raw text. Hence, it's imperative for that text input data to get converted into numbers or vectors of numbers, in order to make it usable for a number of algorithms.

A bag of words model is one of the ways to make the text usable for the algorithms. Essentially, it is a representation of text that works on the occurrence of words in the document. It has nothing to do with the structure, order, and location; this model only looks for the count of the words as a feature.

The thought process behind this model is that having similar content means having a similar document.

The different steps to be taken in the bag of words model are as follows:

• Building the corpus: In this step, the documents are collected and combined together to form a corpus. For example, the famous song from the TV series Friends has been used here as a corpus:

I will be there for you
When the rain starts to pour
I will be there for you
Like I have been there before
I will be there for you

Let's consider each line of th...

Table of contents

• Title Page
• Copyright and Credits
• About Packt
• Contributors
• Preface
• Quantifying Learning Algorithms
• Evaluating Kernel Learning
• Performance in Ensemble Learning
• Training Neural Networks
• Time Series Analysis
• Natural Language Processing
• Temporal and Sequential Pattern Discovery
• Probabilistic Graphical Models
• Selected Topics in Deep Learning
• Causal Inference
• Advanced Methods
• Other Books You May Enjoy