Understanding Machine Learning with R

This course walks through the process of creating a machine learning prediction solution. The course introduces and uses R, the primary language for Machine Learning.
Course info
Rating
(271)
Level
Beginner
Updated
Feb 17, 2016
Duration
1h 25m
Table of contents
Description
Course info
Rating
(271)
Level
Beginner
Updated
Feb 17, 2016
Duration
1h 25m
Description

In this course, you will learn how developers and Data Scientists use Machine Learning to predict events based on data. Specifically, how to format your problem to be solvable, where to get data, and how to combine that data with algorithms to create models that can predict the future. Throughout this course we will use R, one of the best known Machine Learning languages. No previous R experience is required.

About the author
About the author

Jerry Kurata is a Solutions Architect at InStep Technologies.

More from the author
Deep Learning with Keras
Intermediate
2h 33m
26 Mar 2018
TensorFlow: Getting Started
Beginner
2h 37m
3 May 2017
Section Introduction Transcripts
Section Introduction Transcripts

Course Overview
Hi, my name is Jerry Kurata and welcome to my course, Understanding Machine Learning with R. These days, machine learning is all around us, from helping doctors diagnose patients to detecting fraudulent credit card transactions. We likely encounter machine learning applications a dozen times a day and may not even realize it. It silently scans our inbox for spam emails and makes sure we see an ad on every web page for those shoes we looked at last week. This course will introduce you to machine learning and the technology behind it. You will see why companies are in such a rush to use machine learning to grow their business and increase profits. You will learn how developers and data scientists use machine learning to predict events based on data, specifically, how to format your problem to be solvable, where to get data, and how to combine that data with algorithms to create models that can predict the future. Throughout this course, we'll use the R language, the best known machine learning language. We'll utilize R and its libraries to make it easy to build and test machine learning solutions. However, you don't need prior R experience. In this course, we will learn by doing and the R code you will use will be explained in the examples. By the end of this course, you'll know the how, when, where, and why of building a machine learning solution. You will have the skills you need to transform a one-line problem statement into a tested prediction model that solves the problem. I look forward to you joining me on this journey of Understanding Machine Learning with R from Pluralsight.

Understanding Machine Learning With R
Hi, I'm Jerry Kurata. Welcome to the Pluralsight course on Understanding Machine Learning with R. In this course, you'll learn how to apply machine learning to solve problems that are difficult, and some might say impossible to solve with standard coding techniques. In this module, we'll provide some basic information about machine learning. This includes examples of machine learning, a definition of machine learning, and importantly, how machine learning differs from traditional programming. We'll go over the two basic types of machine learning, supervised and unsupervised. We'll see each of these types in action, which will clarify how they differ, and when each type of machine learning should be used. After that, we'll review the contents of this course, and the skills you need, and do not need, for this course. We'll finish with a brief discussion of how machine learning fits into the larger subject of data science.

Training the Model
Hi, I'm Jerry Kurata. Welcome back to the Pluralsight course Understanding Machine Learning with R. In previous modules, we covered the workflow steps of asking the right question, where we defined our solution statement, preparing data, in which we obtained raw data and transformed it into data we could use for training, and selecting the algorithm, where we selected the initial algorithm we will train and evaluate. In this module, we'll put the pieces together and train the algorithm we selected with the data we prepared. When we are done with this training process, we will have a model that can predict if a flight will be delayed. In this module, we'll get a detailed understanding of the training process, introduce the Caret package which can make the training and evaluation process easier, then go back to R and train our algorithm with our DOT delay data and produce a train model. A good definition of machine learning training is letting specific data teach a Machine Learning algorithm to create a specific forecast model. Notice the use of the term "specific. " Data drives the training. If data changes over time or new data is used, in many cases we need to go back and retrain. And we want to retrain if the data changes. Retraining will ensure that our model can take advantage of the new data to make better predictions and also verify the algorithm can still create a high-performance model with the new data.

Testing Your Model's Accuracy
Hello I'm Jerry Kurata. Welcome back to the Pluralsight course on Understanding Machine Learning with R. In previous modules we went through the workflow steps of defining the solution statement, getting our data, and selecting an initial algorithm. In the last module we produced a model trained with our training data. In this module, we will evaluate this trained model and see how well it can predict if a flight will be delayed. In this module we will evaluate our trained model by using a set of test data. Remember this test data was not used to train the model so it should give us an accurate estimate of the real world performance of our model. This evaluation will provide us with a series of results that we can use to decide if the performance of the model is acceptable. The results will also give us some ideas on how we might revise the workflow steps to improve performance. Throughout the evaluation process we need to keep in mind that statistics only provide us with data. We are the ones that interpret this data and determine if it is good or bad. And, we need to define good and bad in the context of how we will use our model. But, enough theory for now, let's go back to R and evaluate the model.