Here is a list of topics that would be appropriate for the capstone project:
- Creating an analysis report with R Markdown
- Applying the Support Vector Machine (SVM) algorithm to a machine learning problem
- Making a map visualization with
ggplot - Creating an interactive visualization with shiny
- Performing deep learning with
keras - Applying principal component analysis to reduce dimensionality of a wide dataset
- Demonstrating how to deal with Big data in R
- Performing network analysis and visualizing a network
- Performing machine learning on videos, images, or audio files
The above is a preliminary list. Students are welcome to propose a topic not included in the list above.
Here is a list of the confirmed project topics:
| Group Number |
Project Description | Assigned to |
|---|---|---|
| 1 | Classifying retinal image scans according to retinal damage due to diabetes | DE, KDS |
| 2 | Applying cluster analysis to detect fake Amazon reviews | SM |
| 3 | Classifying different styles of beer | IR, XR |
| 4 | Using shiny to visualize data from yelp reviews | CL, BW |
| 5 | Classifying hand written digits with SVM | SX, YZ |
| 6 | Visualizing US population data | TM, SK |
| 7 | Building a movie recommender system | KM, YZ |
| 8 | Investigating the relationship between meat consumption and mortality | SC, KJ |
| 9 | Detecting particle trajectories in a high energy physics experiment | DF, CP |
| 10 | Classifying images using convolutional neural network | BM, XC |
-
Week 2: Find a team member to work with.
-
Week 3: Identify a motivating dataset to work with.
-
Week 4: Discuss idea with instructor and get approval to proceed.
-
week 5: Progress meeting with instructor.
-
week 6: Progress meeting with instructor.
-
week 7: Progress meeting with instructor.
-
week 8: Draft product ready (R script, Rmd report, shiny website, etc.).
-
week 9: Presentation ready.
-
week 10: Present your work to the class.
- You will present (as a team) your project to the class in the form of a .ppt document. The presentation may also include a live demo in R if this's useful. You have a total of 12 minutes to present your work.
- 24 hours before the date of the presentation, the .ppt file, .R scripts, and datasets used shall be made available to the course Instructor via a shared folder.
- Meeting with Instructor to discuss the progress of the project (10%)
- Delivering the files in final form by the due date (10%)
- How well you explain the main theoretical concept(s) (20%)
- How well you explain the steps involved in the analysis/modeling (20%)
- Soundness of your approach to the problem (e.g. did you miss some obvious steps) (20%)
- Performance on answering judges’ questions (Individual basis) (20%)