en_GB
Hold Ctrl-tasten nede. Trykk på + for å forstørre eller - for å forminske.

BST240_1

Big Data Analysis for Social Sciences

This is the study programme for 2019/2020. It is subject to change.


Big Data Analysis is all about using computer-assisted methods to analyze large amounts of data. Using the R statistical programming language, methods such as topic modeling and sentiment analysis will be explored. But, as a course without formal prerequisites, also the general form and function of (R) code will be discussed. Although most of the methods discussed in the course are applicable to many kinds of data, there will be a specific focus on text data, such as newspaper articles and online news.

Learning outcome

After completion of this course students will be able to understand and explain the main concepts related to machine learning and automated text analysis. They will also be able to apply this knowledge to develop research questions and designs that are suitable for use with machine learning methods. Finally, students will be able to write code in R/RStudio to manage their data and conduct simple machine learning tasks.

Contents

The first part of the course consists of a general introduction into (coding with) R, and a theoretical introduction to the different kinds of machine learning methods available. At the end of the first part, students will write a (graded) research proposal to illustrate their understanding of the main concepts in big data analysis, and relate these concepts to a subject of their personal interest.
During the second phase of the course, students will be placed into small groups, based on their interest, and develop a single research proposal out of their separate proposals. They will then use the methods learned in part one to answer the research question(s) in this proposal and write a report, which will be graded.
The course will consist of the following subjects:
  • Data structures (csv, json, databases)
  • Introduction to R
  • Introduction to RStudio
  • Preprocessing
  • Supervised learning
  • Unsupervised learning

Note that the maximum number of students able to participate in this course is limited to 12. Also, students should have successfully completed the Quantitative Methods course before they can take part in this course. Students that are eligible are admitted to the course based on their time of registration (first come, first served)
A well-functioning laptop is required. Recommended system requirements are at least an Intel Core i3 or equivalent, and at least 4GB RAM.

Required prerequisite knowledge

BSS300 Quantitative research methods

Exam

Individual research proposal and Joint research report
Weight Duration Marks Aid
Individual research proposal3/5 A - F
Joint research report2/5 A - F
Individual research proposal (60%)
Joint research report (40%)

Course teacher(s)

Course coordinator
Erik de Vries

Method of work

  • Lectures
  • Groupwork
  • Individual work
  • Supervisory group meetings

Literature



This is the study programme for 2019/2020. It is subject to change.

Sist oppdatert: 13.11.2019

History