| Instructor |
Professor Alan
Montgomery
( alm3@andrew.cmu.edu) |
| Teaching assistant |
Mr. Jason Ma |
| Syllabus |
Data mining focuses on identifying patterns using past transactions to
discover relationships. By its very nature electronic commerce is able
to generate large amounts of information and data mining methods are
quite helpful for managers in turning this information into knowledge
which in turns can be used to make better decisions. These quantitative
methods have the potential to dramatically change decision making in
many areas of business. For example ideas like interactive marketing,
customer relationship management, and database marketing are pushing
companies to utilize the information they collect about their customers
in order to make better marketing decisions.
This module focuses on how data mining techniques can be applied to
solve managerial problems in marketing and electronic commerce. The
emphasis is on understanding and applying existing techniques using
computer software tools. The set of data mining techniques and marketing
problems that can be studied is immense; therefore our strategy will be
to focus on popular techniques like decision trees, logistic regression,
linear regression, and text processing methods. Each of these techniques
is applied to a specific case study in which students will be asked to
solve a business problem using the specified approach. The objective is
for students to be able to generalize their experience in these settings
to other problems using the same technique. |
| Objectives |
- Understand the data mining process
- Introduce and understand a set of data mining techniques
- Illustrates these techniques to specific case studies
for solving business and e-commerce problems
|
| Learning outcomes |
Upon completion of this module you will understand the role of data mining
in addressing many problems that arise in electronic commerce. Our focus
is on learning a core set of principles that underlies many data mining
techniques, and consider their applications to specific problems like:
analyzing clickstream data from web traffic, market segmentation,
predicting customer profitability and retention, and analyzing text data
from keyword search and social media. We consider a compatible set of
techniques to analyze these problems, such as cluster analysis, linear
regression, regression trees, and logistic regression. The focus of this
course is on the application and interpretation of the techniques, as
opposed to implementing the algorithms. |
| Prior knowledge expected |
It is expected that students understand that this module makes extensive
use of statistics and mathematics. |
| Topics covered |
- Understanding Data and the Data Mining Process
- Exploring and Visualizing Data
- Data Based Decision Making
- Market Segmentation of Customers
- Predictive Modeling
- Identifying Profitable Customers
- Overfitting and Evaluating Models
- Data Mining Techniques for Prediction
- Working with Unstructured Datasets
- Business Strategy for Employing Data Science
|
| Teaching format |
10 three-hour lectures with prescribed readings to provide important
background information for the lectures. Students are asked to complete
a considerable amount of computer and analysis work. |
| Assessment |
- In-course assessment (65%)
- Class Participation (10%)
- Three case assignments (55%)
- Final examination (35%)
Date: 11 August 2015 (Tue)
Time: 7:30pm - 9:30pm
Venue: TBA
Note: The textbook for this module is permitted during the
examination along with a one-page sheet (A4) of notes.
This note sheet may be typed or hand-written on both sides,
but the type must be legible (7 point or larger)
|
| Course materials |
Prescribed textbook: Foster Provost and Tom Fawcett, Data Science for
Business: What you need to know about data mining and data-analytic
thinking, O'Reilly Media; 1 edition (August 16, 2013) |
| Enrolment |
No add or drop after 30 June 2015. |
| Class quota |
Class size will be maintained at 45. |