Data mining and Exploration
Coursework

Grade 60% from total
Team assignments: 2-3 members

1. Description:

The coursework consists of two main components: 1) a critical survey, due on the 4th week, and 2) a research project with final report and presentation that are due on the 9th week. This is a team project; each team shall consist of 2-3 members.

2. Milestones:
1. Critical Survey [due on the 4th Week, weight: 20%]
2. Final Report [due on the 9th Week, weight: 30%]
3. Presentation [due on the 9th Week, weight: 10%]

Due time is at the beginning of the first lecture in the due week (i.e. Wednesday at 6pm)

The following is a summary of coursework steps and milestones.
3. Critical Survey
This year all projects should be related to real UAE data. For example:

• Analyzing stock market historical data. (Available here but need some work to prepare for analysis http://www.dfm.ae/mw/tradinghistory.aspx?c=1054)
• Analyzing health/car insurance claims and flag suspicious claims
• Analyzing real estate classifieds. (Data available with instructor)

Each team needs to survey at least 4 papers per member (e.g. team of 3 students will need to survey at least 12 papers). Use google scholar and other online resources in order to obtain papers.

Deliverables:
A survey, no more than 5000 words, single column, describing briefly and precisely

• Title of the topic (to be surveyed) and the team members (name and ID of each).
• Division of work: what did each team member do?
• Survey: summarize the different papers, highlighting:
• What is the data that were analyzed? Is the data available (e.g. can you download it?)
• how the data were represented
• what analysis was conducted on the data and what patterns were discovered
• the more integrated the survey is, the better (i.e., not simple collection of paper summaries.)
• Proposal: From what you read, what ideas you think are worth pursuing?
4. Final Report
Choose an idea from the previous report to pursue in further detail. It is important to make sure the data you need is available. Analyze your data set and report the results.

Deliverables:
a report, no more than 6000 words, single column, describing briefly:

• Title of the topic.
• Division of work: what did each team member do?
• Setting
o What tools did you use?
o What measurements did you use for evaluation?
• Results:
o At least 3 figures plotting different measure combinations. For example, a figure plotting the node degree vs. the node strength. Be creative and try to find interesting patterns.
• Analysis:
o Comments on the results. This is the heart of the analysis. Point out what relationships you observed.
5. Presentation
20 minutes presentation about the work done in the project. Each team member should participate in the presentation, to describe what s/he did. The presentation should include:

o Brief introduction
Problem
Related work
o Approach
Technique used & why chosen?
Tools used & why chosen?
o Results
Main results
Analysis & findings
o Conclusion
What is the main lesson

7. Important Notes
a) More than one team can choose the same topic.
b) All deliverables to be submitted through turnitin on the blackboard.
c) Acknowledge all assistance on assignments and all references.
d) Use blackboard for questions
8. Late submission
According to the university policy, every day late you lose (-) 2%, up to 5 working days excluding the weekend and holidays. If you are late more than the 5 days you will get zero.

Notice that. It is the student responsibility to make sure that the deliverable satisfies the requirements.

Marking Scheme

The grading will be broken down based on the following criteria:

Deliverable Criterion Max Actual
Survey Clarity & quality of writing 8%
Quantity & Quality of surveyed papers 4%
Originality of proposed ideas 4%
Individual contribution 4%
Total for Survey 20%
Report difficulty of the problem (amount of work) & originality of the solution (how it differs from related work) 6%
Overall organization, readability & academic writing style 6%
Results, Depth of analysis, Critical evaluation of own work (discussion of results & conclusion) 10%
Individual contribution 6%
Referencing (style and completeness) 2%
Total for the report 30%
Presentation Clarity of slides & presentation 4%
Innovation & quality of slides (e.g. illustrative animation) 2%
Individual understanding 4%
Total for Presentation 10%