Course 9: Applied Data Analytics
Course Objective: The objective of this course is to help students develop competences on
statistical techniques needed for data analysis, and various data mining techniques and
algorithms used in practical problems that require processing big data for decision making
purpose.
Learning Outcomes:
The students on the completion of this course would be able to
• Apply various inferential statistical analysis techniques to describe data sets and
withdraw useful conclusions from the data sets (e.g., confidence interval, hypothesis
testing)
• Apply data visualization techniques and key data mining techniques (e.g., classification
analysis, associate rule learning, anomaly/outlier detection, clustering analysis,
regression analysis) in dealing with big data sets
• Implement the analytic algorithms for practical data sets
• Perform large scale analytic projects in various industrial sectors
• Work and communicate effectively in teamwork
Prerequisite: None
Course Outline:
Module 1: Basic Data Analysis
I. Basic Concepts
1. Descriptive Statistics
2. Statistical Inferences
3. Data Measurement
4. Measures of Central Tendency and Dispersion
5. Common Statistical Graphs
6. Determination of Outliers
II. Statistical Inferences
1. Point Estimation and Required Properties of Point Estimators
2. Interval Estimations for Mean, Proportion and Variance of Population
3. Sample Size Determination
III. Hypothesis Testing
1. Hypothesis Testing for Mean, Proportion and Variance of Population – Single
Sample Test
2. Hypothesis Testing for Mean, Proportion and Variance of Population – Two
Samples Test
3. Type I and Type II Errors – Power of the Test
4. Observed Significance Level
Module 2: Data Visualization
IV. Data Visualization
1. Introduction to Data Visualization
2. Basic Charts for Numerical Data and Categorical Data
3. Distribution Plots
4. Multivariate Charts: Combo Chart, Combination Chart, Stacked Column Chart
V. Data Dashboard
1. What is a Data Dashboard?
2. Applications and Benefits of Data Dashboard
3. Design and Construct a Data Dashboard
Module 3: Key Data Mining Techniques
VI. Regression Analysis
1. Linear Regression and Least Square Method
2. Residual Analysis
3. Multiple Regression
4. Goodness of Fit Tests
VII. Data Classification
1. k-Nearest Neighbor Algorithm for Estimation and Prediction
2. Distance Functions: Euclidian, Manhattan, Minkowski, Min-Max Normalization,
Z-Score Standardization
3. Logistics Regression
4. Bayesian Networks
5. Model Evaluation Measures for Classification Task
VIII. Data Clustering
1. Hierarchical Clustering Method
2. k-Means Clustering
3. Measuring Cluster Goodness: The Silhouette Method and The Pseudo-F Statistic
IX. Association Rules
1. Affinity Analysis
2. The a Priori Algorithm – Generating Frequent Itemsets
3. The a Priori Algorithm – Generating Association Rules
4. Measure the Usefulness of Associate Rules
X. Case Studies/Group Projects
Laboratory Sessions: None
Learning Resources:
Textbook: No designated textbook, but class notes and handouts will be provided
Reference books:
1. Larose, D.T. and Larose, C.D., Data Mining and Predictive Analytics, 2nd edition,
Wiley, 2015
2. Shmueli, G., Bruce, P.C., Yahav, I., Patel, N.R. and Lichtendahl Jr., K.C., Data Mining
for Business Analytics – Concepts, Techniques, and Application in R, Wiley, 2018
3. Ankam, V., Big Data Analytics, Packt, 2016
4. Walkowiak, S., Big Data Analytics with R, Packt, 2016
5. Grolemund, G., Hands-on Programming with R, O’Reilly, 2014
6. Wickham, H. and Grolemund, G., R for Data Science, O’Reilly, 2017
7. Wexler, S., Shaffer, J. and Cotgreave, A., The Big Book of Dashboards: Visualizing
Your Data Using Real-World Business Scenarios, Wiley, 2017
8. O’Cornor, E., Microsoft Power BI Dashboards Step by Step, Practice Files, 2019
Teaching and Learning Method:
The teaching is done via lectures by the instructor. Tutorial sessions are conducted on the use
of tools in each subject. The learning methods include group discussion, individual/group
assignment and group project/case study.
Time Distribution and Study Load:
Lectures: 30 hours
Tutorials/Group Discussions: 30 hours
Self-study: 45 hours
Group project: 40 hours
Time Allocation
Session Activities Time
allocation
I. Basic Concepts Lectures 3 hours
Other activities: 6 hours
· Tutorials on using Excel, R
· Quizzes
· Out-of-class group discussions
· Individual Assignments
II. Statistical Inferences Lectures 3 hours
Other activities: 4 hours
· Tutorials on using Excel, R
· Quizzes
· Out-of-class group discussions
· Individual Assignments
III. Hypothesis Testing Lectures 4 hours
Other activities: 3 hours
· Tutorials on using Excel, R
· Quizzes
· Out-of-class group discussions
· Case Studies (in group)
IV. Data Visualization Lectures 5 hours
Other activities: 17 hours
· Tutorials on using Power BI and R (ggplot2)
· Out-of-class group discussions
· Group Assignments
V. Data Dashboard Lectures 5 hours
Other activities: 15 hours
· Tutorials on using Power BI
· Out-of-class group discussions
· Group Assignments
VI. Regression Analysis Lectures 2 hours
Other activities: 4 hours
· Tutorials on using Minitab, R
· Out-of-class group discussions
· Case Studies (in group)
VII. Data Classification Lectures 3 hours
Other activities: 9 hours
· Tutorials on using R
· Out-of-class group discussions
· Mini Group Projects
VIII. Data Clustering Lectures 3 hours
Other activities: 11 hours
· Tutorials on using R
· Out-of-class group discussions
· Mini Group Projects
IX. Association Rules Lectures 2 hours
Other activities: 6 hours
· Tutorials on using R
· Out-of-class group discussions
· Mini Group Projects
X. Group Projects Each group is required to work on a practical dataset and 40 hours
analyze to withdraw insightful conclusions
Activities:
· Weekly group meeting
· Weekly progress report
· Midway report and presentation
· Final report and presentation
Evaluation Scheme: The final grade will be computed according to the following weight
distribution: Mid-semester examination 20%, assignments and group projects 50%, final
examination 30%. In final grading,
An “A” would be awarded if a student shows a deep understanding of the knowledge learned
through home assignments, project works, and exam results.
A “B” would be awarded if a student shows an overall understanding of all topics.
A “C” would be given if a student meets below average expectation in understanding and
application of basic knowledge.
A “D” would be given if a student does not meet expectations in both understanding and
application of the given knowledge.
Instructor:
