This website uses cookies. By using this website, you consent to the use of cookies in accordance with the NTU Privacy and Cookies Statement.
For more information about cookies and how you can disable them, visit our Privacy and Cookies page.

Continue

Online Courses: MBA

Nottingham Trent University

Telephone UK: 0800 032 1180 Intl: +44 (0)115 941 8419
admissions@online.ntu.ac.uk

Practical Machine Learning Methods for Data Mining

Overview and aims
The digital revolution has made data easy to capture digitally and inexpensive to store. The rate at which data is being stored is growing at a phenomenal rate with databases typically doubling in size every 20 months. As a result, many businesses are struggling to analyse and make sense of this vast collection of data. This is particularly true for the following domains:

  • Business and finance: identifying and interpreting buyer/seller behaviour patterns, risk assessment data, product distribution patterns, foreign exchange rates, stock indexes and prices, interest rates, credit card data, fraud
  • Internet: extracting useful information from online systems, such as company websites, social media and blog sites, to support personal or business decisions
  • Health care: interpreting diverse diagnostic information stored by hospital management systems and biological/biometric data to inform patient care and drug development
  • Telecommunications: Calling patterns and fault management systems

This module therefore aims to facilitate you to develop the core knowledge and skills of a data scientist. You will be exposed to a number of so called ‘machine learning’ techniques that are able to automatically discover useful patterns in data. A practical approach to teaching machine learning is adopted using the open source data mining software package called WEKA. The practical emphasis will help you to develop an intuitive grasp of the sophisticated mathematical ideas that underpin this challenging but fascinating subject.

Module content
The module is designed to develop you as a data scientist who is able to work competently with a variety of different data sets to extract, interpret and present meaningful information.

Subsequently, the content is organised as follows:

  • The lifecycle of knowledge discovery in databases (KDD)
  • Understanding and transforming your data
  • The basic components of machine learning models
  • Model fitting, model selection and model evaluation
  • Common machine learning methods for classification, prediction and clustering
    • Decision tree learning (ID3 and variants thereof)
    • Naïve Bayes
    • Simple linear models (linear regression, least means squared)
    • Artificial neural networks
    • Simple K-means clustering

The module takes a practical approach to teaching machine learning and you will therefore use the data mining package WEKA (www.cs.waikato.ac.nz/ml/weka) to put theory into practice using realistic case studies and data. WEKA is a full industrial-strength implementation of the techniques taught on this module and will enable you to gain a deeper understanding of them, appreciating their strengths and applicability to various problem domains. Fundamentally, the module will provide you with opportunities to obtain the computer science skills necessary to analyse large volumes of data to extract meaningful information and patterns for business, government and society.

Learning outcomes
After studying this module you should be able to:

  • critically discuss the principles and ideas underpinning the current practice of data mining
  • prepare data for data analytics using appropriate data pre-processing and transformation techniques
  • propose and evaluate one or more suitable machine learning models for analysing real-world data

This module will help you gain the skills and qualities to:

  • use appropriate software tools for collating, pre-processing and visualising data
  • use appropriate software tools in the construction of an appropriate machine learning solution to a data analytics problem
  • communicate effectively using written formats appropriate to the target audience

Back to MBA curriculum