Subject Description Form

Subject Code COMP417

Subject Title Data Warehousing and Data Mining Techniques in Business and Commerce

Credit Value 3

Level 4

Pre-requisite / Pre-requisite: COMP311

Co-requisite/
Exclusion Co-requisite/Exclusion: Nil

Objectives
This subject aims at equipping students with the latest knowledge and skills to:
• Create a clean, consistent repository of data within a data warehouse for large
corporations;
• Utilize various techniques developed for data mining to discover interesting
patterns in large databases;
• Use existing commercial or public-domain tools to perform data mining tasks
to solve real problems in business and commerce;
• Expose students to new techniques and ideas that can be used to improve the
effectiveness of current data mining tools.

Intended Learning Upon completion of the subject, students will be able to:
Outcomes
Professional/academic knowledge and skills

(a) understand why there is a need for data warehouse in addition to traditional
operational database systems;

(b) identify components in typical data warehouse architectures;

(c) design a data warehouse and understand the process required to construct one;

(d) understand why there is a need for data mining and in what ways it is different
from traditional statistical techniques;

(e) understand the details of different algorithms made available by popular

commercial data mining software;

(f) solve real data mining problems by using the right tools to find interesting
patterns;

(g) understand a typical knowledge discovery process such as CRISP-DM;

(h) obtain hands-on experience with some popular data mining software.

Attributes for all-roundedness

(i) solve real-world problems in business and commerce using data mining and
data warehousing tools;
 (j) learn independently and search for relevant information to write reports to
recommend appropriate data warehousing and data mining tools.

(k) Solve complex problems individually or in groups and develop group work
skills directly and indirectly.

Alignment of Programme Outcomes:

Programme Outcome 1: This subject contributes to having students practice their

writing skills with project document and report writing.

Programme Outcome 4: This subject contributes to developing student critical

thinking through assignments, tutorial and lab exercises on solving problems.
They will also practice more in written assignments and projects involving real
cases in business and commerce.

Programme Outcome 7: This subject contributes to team work with group-based

project for students to practice team spirit.

Subject Synopsis/ Duration of

Indicative Syllabus Topic Lectures /
Tutorials
1. Introduction to data warehousing and data mining 3
Introduction to data warehousing and data mining; possible
application areas in business and finance; definitions and
terminologies; types of data mining problems.
2. Data warehousing 3
Data warehouse and data warehousing; data warehouse and
the industry; definitions; operational databases vs. data
warehouses.
3. Data warehouse architecture and design 3
Data warehouse architecture and design; two-tier and three-
tier architecture; star schema and snowflake schema; data
characteristics; static and dynamic data; meta-data; data
marts.
4. Data Replication and Online Analytical Processing 6
Data replication, data capturing and indexing, data
transformation and cleansing; replicated data and derived
data; Online Analytical Processing (OLAP);
multidimensional databases; data cube.
5. Data mining and knowledge discovery 3
Data mining and knowledge discovery, the data mining
lifecycle; pre-processing; data transformation; types of
problems and applications.
6. Association rules 3
Mining of association rules; the Apriori algorithm; binary,
quantitative and generalized association rules; interestingness
measures.
7. Classification 6
Classification; decision tree based algorithms; Bayesian
approach; statistical approaches, nearest neighbor approach;
neural network based approach; genetic algorithms based
 technique; evaluation of classification model.

8. Clustering 6
Clustering; k-means algorithm; hierarchical algorithm;
Condorset; neural network and genetic algorithms based
approach; evaluation of effectiveness.
9. Sequential data mining 3
Sequential data mining; time dependent data and temporal
data; time series analysis; sub-sequence matching;
classification and clustering of temporal data; prediction.
10. Other techniques 6
Computation intelligence techniques; fuzzy logic, genetic
algorithms and neural networks for data mining.
Total 42

Laboratory Experiment:
Duration of
Topic
Laboratory
1. Knowledge discovery lifecycle using CRISP-DM 2
2. Discover Association rules and sequential patterns using 2
Clementine 2
3. Discover Classification rules using Clementine 1
4. Discover Clusters using Clementine
Total 7

Case Study:
• Application of data mining techniques to solve real business problems.
• Attributes leading to success and failure of data warehousing projects tutorials
when appropriate.

Teaching/Learning This subject consists mainly of class lectures and laboratory sessions. For the
Methodology class lectures, various cases will be presented to help student understand why there
is a need for data warehouse to be built and why data mining is important for
modern day business intelligence. Students will be given time to participate in
discussions when the cases are presented.

All assignments and projects will also be given in the form of different cases
collected so as to allow students to learn more about how data warehouse and data
mining can be and have been used in real business environment. For the projects
and assignments, students are expected to learn independently and think critically
with minimize guidance. They are expected to practice their writing kills through
project documentations and report writing. As students will work in teams on the
project, they are expected to also learn to work with each other collaboratively.

During laboratory sessions, students will be introduced to popular software

products from Oracle and IBM that can support the building of data warehouses
and the mining of them. Students are expected to solve real data mining problems
by using the right tools to find interesting patterns.
Assessment
Methods in Specific % Intended subject learning outcomes to be assessed
Alignment with assessment weighting (Please tick as appropriate)
Intended Learning methods/tasks
Outcomes a b C d e f g h i j k

1. Assignments 55%     

2. Project        

3. Examination 45%         

Total 100 %

The assessment consists of written assignments, a group project and an examination. For
the assignments and projects, they are designed to ensure that students are able to achieve
the learning outcomes intended for this subject. They are expected to tackle a number of
cases drawn from different application areas in business and commerce so that they can
understand why there is a need for data warehouse in addition to traditional operational
database systems and why data mining is important for modern-day business intelligence.
In addition, students will learn through the questions and cases, when a particular data
warehouse architecture or when a particular data mining algorithm is useful and should be
used. Questions in the assignments are expected to help students learning the details of
the data mining algorithm and the use of popular data mining software. They are also
expected to use such popular tool as Oracle Warehouse Builder to construct data
warehouses. For the projects, students are expected to work in groups of three to four to
tackle a real case involving the design of a data warehouse or the use of data mining to
mine very large data bases. They are expected to learn how real-world problems in
business and commerce should be tackled using real-world tools as Oracle’s Warehouse
Builder or IBM’s Clementine data mining system. They are expected to learn
independently and search for relevant information to write reports to recommend
appropriate data warehousing and data mining tools. Students are expected to practice
their writing skills with project document and report writing. They will learn to develop
critical thinking and team work skills.

Student Study Class contact:

Effort Required
 Lecture 42 Hrs.

 Laboratory 7 Hrs.

Other student study effort:

 Assignments and case studies 40 Hrs.

 Projects and research 20 Hrs.

Total student study effort 109 Hrs.

Reading List and

References Reference Books:
1. Chan, K.C.C., Course Notes and Lab Manuals for COMP417, 2009.
2. Inmon, W.H., Strauss, D., and Neushloss, G., DW 2.0: The Architecture
for the Next Generation of Data Warehousing, Morgan Kaufmann, 2008.
3. Golfarelli, M., and Rizzi, S., Data Warehouse Design: Modern Principles
and Methodologies, McGraw-Hill, 2009.
4. Rokach, L., and Maimon, O., Data Mining with Decision Trees: Theory
and Applications, World Scientific, 2008.
5. Witten, I.H., Frank, E., Hall, M.A., Data Mining, Third Edition: Practical
Machine Learning Tools and Techniques, Morgan Kaufmann, 2011.
6. Westphal., C., Data Mining for Intelligence, Fraud & Criminal
Detection: Advanced Analytics & Information Sharing Technologies,
CRC Press, 2008.
7. Cox, E., Fuzzy Modeling and Genetic Algorithms for Data Mining and
Exploration, Morgan Kaufmann, 2005.
8. Liu, B., Web Data Mining: Exploring Hyperlinks, Contents, and Usage
Data, Springer, Berlin Heidelberg, 2009.
9. Tsiptsis, K., and Chorianopoulos, A., Data Mining Techniques in CRM:
Inside Customer Segmentation, Wiley, 2010.
10. Han, J. and Kamber, M., Data Mining: Concepts and Techniques, 2nd
Edition, Morgan Kaufmann, 2005.
11. Shapiro, A.F., and Jain, L.C., Intelligent and Other Computational
Techniques in Insurance: Theory and Applications, World Scientific,
2003.