Predict Gas Guzzlers using a Neural Net Model on the MPG Data Set

Coursera Project Network

Complete a random Training and Test Set from one Data Source using an R function.

Practice data distribution using R and ggplot2.

Apply a Neural Net model to the Data and examine the results by building a Confusion Matrix.

Clock2 Hours
Comment Dots英語

In this 1-hour long project-based course, you will learn how to (complete a training and test set using an R function, practice looking at data distribution using R and ggplot2, Apply a Neural Net model to the data, and examine the results using a Confusion Matrix. Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.


Random ForestData ScienceData AnalysisMachine Learning



  1. Task 1: In this task the Learner will be introduced to the Course Objectives, which is to how to execute a Neural Network using the NeuralNet R package on the MPG data set. There will be a short discussion about the Interface and an Instructor Bio.

  2. Task 2: The Learners will get experience looking at the data using ggplot2. This is important in order for the practitioner to see the balance of the data, especially as it relates to the Response Variable.

  3. Task 3: The Learner will get experience creating Testing and Training Data Sets. There are multiple ways to do this and the Instructor will go over two of them in this Task.

  4. Task 4: The Learner will get experience with the syntax of the Neuralnet package in R by building out a neural net model. There will be a short discussion on the differences between the predict function in R and compute with the Neuralnet package as well.

  5. Task 5: The Learner will get experience evaluation models in this Task. The Confusion Matrix will be discussed as the evaluation metric of choice for the specific problem. The conclusion of the course will use the two evaluation metrics see how well the model performed on the test data set.