# ELE 3912 Machine Learning for Business Using R

## ELE 3912 Machine Learning for Business Using R

In finance, marketing, consulting, public policy, science, you name it, people who can make sense of large and complex data sets are in high demand. Machine learning refers to a vast set of tools that help us in making sense of data. This course gives an introduction to various machine learning methods, concentrating on those that fall under the umbrella of supervised learning. The course contains both model based and algorithm based machine learning approaches, contrasting the advantages and limitation of the two. While the course contains some mathematical statistics meant to enable the students to peek into various algorithms, the focus is on doing things. The thing we do is to make sense of big data sets, and in so doing we will use the programming language R.

During the course students shall learn:

- To think critically about the possibilities and limitations of machine learning.
- When to use algorithm based approaches to data analysis, and when a model based approach might be called for.
- To understand the differences between predicting and explaining phenomena, and how this relates to choice of methodology.
- The importance of uncertainty quantification.
- How to identify challenges related to big data.

After completed course students shall be able to:

- Implement a large set of machine learning methods in R.
- Be able to evaluate and compare the performance of different methods.
- Produce graphics illustrating one's findings and predictions, as well as the performance of ones methods.

Students will understand that in many situations machine learning tools are a prerequisite for decision making, but also be critically aware of its limitations and underlying assumptions.

- Introduction to R.
- What is machine learning.
- Linear regression, including dimension reduction methods.
- Classification. Logistic regression, linear discriminant analysis, naive Bayes, and K nearest neighbours.
- Bootstrapping and other resampling methods.
- Tree-based methods. Bagging, random forests, boosting.
- Data cleaning.

The course consists of a combination of lectures and problemsolving in class using R.

The students are expected to do all the homework, though it is not mandatory.

Higher Education Entrance Qualification

**Disclaimer**

Deviations in teaching and exams may occur if external conditions or unforeseen events call for this.

Basic courses in Mathematics and Statistics.

Exam category | Weight | Invigilation | Duration | Support materials | Grouping | Comment exam |
---|---|---|---|---|---|---|

Exam category:Submission Form of assessment:Written submission Exam code:ELE 39121 Grading scale:ECTS Grading rules:Two examiners Resit:Examination when next scheduled course | 50 | No | 2 Week(s) | Group/Individual (1 - 3) | A 2 week take home exam, concentrating of the analysis of one or more data sets. Will also contain some theory. All exams must be passed to obtain a final grade in the course. | |

Exam category:Submission Form of assessment:Written submission Exam code:ELE 39122 Grading scale:ECTS Grading rules:Two examiners Resit:Examination when next scheduled course | 50 | Yes | 3 Hour(s) | - BI-approved exam calculator
| Individual | Regular school exam. All exams must be passed to obtain a final grade in the course. |

All exams must be passed to get a grade in this course.

Activity | Duration | Comment |
---|---|---|

Teaching | 36 Hour(s) | |

Feedback activities and counselling | 36 Hour(s) | Homework sessions, problemsolving under supervision, etc. |

Student's own work with learning resources | 45 Hour(s) | |

Submission(s) | 30 Hour(s) | |

Examination | 53 Hour(s) | A 2 week take home exam and final exam. |

A course of 1 ECTS credit corresponds to a workload of 26-30 hours. Therefore a course of 7,5 ECTS credit corresponds to a workload of at least 200 hours.