Teaching Responsibility

LJMU Schools involved in Delivery:

Astrophysics Research Institute
Computer Science and Mathematics

Learning Methods

Lecture
Seminar
Workshop

Module Offerings

7022DATSCI-JAN-CTY

Aims

The module provides an introduction to algorithms for the analysis of complex data sets. This includes high-dimensional data sets which require pre-processing using efficient dimensionality reduction and visualisation techniques. Other complex data sets come in the form of large graphs (networks) or have been generated by complex underlying processes. Especially for the latter class of data sets the module aims to develop the key skill of working with experts from other domains.

Learning Outcomes

1.
Synthesize a combination of various machine learning and pre-processing techniques to explore large data sets.
2.
Apply efficient dimensionality reduction and distributed & parallel data processing.
3.
Evaluate different incremental machine learning approaches.

Module Content

Outline Syllabus:1. Key Linear Algebra Techniques (vectors, matrices, numerical calculation of eigenvalues and eigenvectors) 2. Dimensionality Reduction (statistical methods and random projections) 3. Network Analysis 4. Monte Carlo methods for analysis and simulation of complex systems (e.g. Supernova atmosphere). 5. Computational Bayesian statistics and Markov Chain Monte Carlo (MCMC) 6. Practical implementation of statistical and machine learning techniques using distributed and parallel data processing.
Module Overview:
This module aims to develop skills in big data analysis, including techniques of dimensionality reduction and the application of statistical and machine learning models. In addition it aims to develop the key skill of working with experts from other domains.

Assessments

Report
Exam