Data Mining for Direct Marketing and Finance
Course Co-ordinator: Nicos Pavlidis (A54a, email@example.com)
Term Taught: Lent
Pre-requisites: MSCI 100 or MSCI101 or MSCI110 or equivalent basic maths & stats courses given by other departments.
How do financial institutions predict whether a person will pay back a mortgage, credit card or loan they grant? How do marketeers predict the likelihood of a customer to react to a shopping catalogue or a special discount in a direct mailing? How does AMAZON determine which products to present to you in the hope of up-selling or cross-selling? Data mining has now been formally established as a discipline of its own right to support managerial decision making in predictive and descriptive modelling in business, bringing together simple algorithms (that often do not draw upon rigorous statistics, but work with impure data) and real datasets.
This module ‘Data Mining for Direct Marketing and Finance’ develops further the students’ modelling skills on synthetic and empirical data by showing simple statistical methods and introducing novel methods from artificial intelligence and machine learning. Although an understanding of basic statistics may allow you to better appreciate the working of models in regression and classification tasks, the course is focussed on the application of algorithms on real datasets using software, and making real predictions. The skills and knowledge enable you to pursue jobs in the area of Business & Marketing Analytics in the role of a Business Analyst or Consultant who often apply these methods and techniques.
The course will also include a series of workshops in which you will learn how to use the SAS Enterprise Miner software for data mining (a software skill much sought after in the job market) and how to use it on real datasets in a real world scenario.
The module will cover a range of data mining methods including logistic regression, decision trees, and artificial neural networks, and develop SAS Enterprise Miner software skills & techniques.
Course Objectives and Learning Outcomes
Subject-specific learning outcomes:
When you have completed this course you should be able to understand:
- General modelling concepts in relation to complex models.
- How to use a wide range of advanced data mining methods to handle and filter data of different types and for different applications.
- How to structure an SAS Enterprise Miner model to deal with complexity and large datasets.
Contact Hours: 20 hours of lectures and 8 workshop hours
There will be one 2 hour lecture per week and one tutorial, in the form of computer lab workshops.
The individual group work (50%) will consist of a small scale data mining exercise including data analysis, exploration, transformation, modelling of different classification methods and evaluating them. Alternatively, an essay on topics relevant to data mining may be set as coursework.
The group coursework (50%) will consist of participating in a current or past competition style exercise, applying classification or clustering methods on real datasets.
Both assignments will involve building multiple models following the complete data mining process to answer a practical management science application.
For all coursework that is submitted to the department, you must attach and sign a document where you declare that you have read and are aware of the university regulations on collusion and plagiarism. All coursework should be submitted by the deadline. Late work will be penalised by one full grade for submission of up to three days and will be awarded a mark of zero after that. Extensions may in certain exceptional circumstances be granted but you have to contact the lecturer/tutor before the deadline with a valid reason. Valid reasons for extensions are normally restricted to severe illness and family bereavement.
Reading and Lecture Notes
Lecture notes will be provided for all lectures and will be posted on the course web board before the lectures. The recommended book for this course is "Introduction to Data Mining" by Tan, Steinbach, and Kumar. For SAS Enterprise Miner, a reference book is "Data Mining Using SAS Enterprise Miner" by Matignon. The book is available as e-book from the library.
The undergraduate secretary for the Management Science department is in A68 Management School. Her office hours are 10 - 12 and 2.30 - 4.30.
Lecture notes, Tutorial questions and answers, coursework assignments and all other information related to this course will be posted on the web board of the course.
There is also a departmental web board giving answers to frequently asked questions at: