Skip to content

Latest commit

 

History

History
67 lines (46 loc) · 2.76 KB

File metadata and controls

67 lines (46 loc) · 2.76 KB

Capstone Project

Topics

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

Timeline

  • 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.

Deliverables

  • 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.

Grading rubric

  • 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%)