Introduction to Data Analytics using R

Module outline

This module covers the principle concepts and techniques of data analytics and how to apply them to large-scale data sets. Students develop the core skills and expertise needed by data scientists, including the use of techniques such as linear regression, classification and clustering. The module will show you how to use the popular and powerful data analysis language and environment R to solve practical problems based on use cases extracted from real domains.

Aims

To study advanced aspects of big data analytics, applying appropriate machine learning techniques to analyse big data sets, assessing the statistical significance of data mining results, and using the open-source tool R to perform basic data mining tasks on big data.

Syllabus

• Introduction to big data analytics: big data overview, data pre-processing, concepts of supervised and unsupervised learning.
• Basic statistics: mean, median, standard deviation, variance, correlation, covariance.
• Linear regression: simple linear regression, introduction to multiple linear regression.
• Classification: logistic regression, decision trees, SVM.
• Ensemble methods: bagging, random forests, boosting.
• Clustering: K-means, K-medoids, Hierarchical clustering, X-means.
• Evaluation and validation: cross-validation, assessing the statistical significance of data mining results.
• Selection of advanced topics such as: scalable machine learning, big data related techniques, mining stream data, social networks.
• Tools: R.

Prerequisites

Experience with a modern programming language.

Timetable

All dates and timetables are listed in the programme handbooks of individual programmes.

Coursework

Several pieces practical exercises involving learning and mining big data sets using the tool R.

Assessment

Recommended reading

• An Introduction to Statistical Learning: With Applications in R: Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani.