Predictive Modeling II Course Materials

Final Project

 The final project is the combination of all the data mining methods and techniques discussed during the entire course of the semester. The project work will involve working with real data.

Important Dates

During mid-term, groups will be formed. Groups will come up with a topic for their final projects, and submit a one-page draft that describes the data, business problem and expected outcomes.

Nail down a topic: Nov 1

 Final Presentation: Nov 13, 15

Submit Written Report: Nov 15


1)      Preferably, each group member should present for about 3 minutes.

2)      Please be sure to show up to conduct the peer evaluations (for your group members and other groups), or you will lost your points for the whole presentation part.


Final Project Grading

20% --> The presentation was delivered in a professional manner (e.g., presenters exhibited enthusiasm for the topic, did not read their presentation, handled questions concisely and knowledgeably, the presentation appeared as an integrated whole)

80% --> According to the presentation and the final report, the analysis was conducted in a professional manner by providing sound prediction results and reasonable model configurations. Models covered in our class were appropriately employed. Group members have made great effort in tuning the model performance. The final report followed the standard writing format for each section (see below).

Final Submissions

The final project report should provide a clear and thorough explanation and documentation discussing all decisions about data mining analyses. Reports (Presentation Slides and a Word document) must be prepared in a professional manner and appearance. Students will present (PowerPoint) the findings of the project in-class. The report in Word MUST contain the following sections. Each part should be concisely and clearly presented.

 The length of the report is expected to be around 10 pages (double-spaced, not including Appendix).

a.         Title Page – Name of the students and title (2 points)

b.         Executive Summary – A brief description of the business problem and conclusions from analysis. 1-2 paragraphs  (5 points)

c.         Table of Contents (3 points)

d.         Literature Review/Background – Literature and background of the study  (10 points)

e.         Introduction/Business Problem – Complete description of the business problem, data collection approaches and variables. (10 points)

f.          Methodology – Type of analysis conducted and rationale for using the technique. Thoroughly document each analysis step. (30 points)

g.         Results – Discuss your findings and compare model performance. (20 points)

h.         Conclusions/Discussion – Conclusions and limitations of the analysis. (10 points)

i.          Appendix – Output from software and relevant screen shots. Graphs and tables are clearly labled and easy to read.(10 points)


Groups of Fall 2017

On-campus groups:

1. PhD group 1: Ross Gruetzemacher

2. PhD group 2: Joonghee Lee, James Locke

3. group 3:  Predict future bitcoins values (Surya Kiran Malladi, Austin McCombs, Tori Sims, Huang Tang)

4. group 4:  US Visa Applications Prediction (Satya Srujan Godavarthi, Akhil Radhakrishnan)


Remote groups:

1. Gender wage gap prediction (Shannon Cassidy)

 2. Fraudulent fare cards detection (Jay Claiborne)

Groups of Fall 2016

Students will be working in groups of 2 to 4 people under the guidance of the instructor. Groups are expected to work on diffent projects.

Group1: Tutoring Prediction  (Shelja Anand, Apoorva Sadhanala, Siddhartha Yamalakonda)

Group2:  Mental health of Auburn University students  (Nicholas Chapek, Carson May, Cheyenne Moon

Group3:  Auburn Basketball (Jzyk Ennis, Michael Messina, Daniel Morrison)

Group4: When coach got fired (Brandon Daley, Jackie Evans, Kathryn Kaish, Anna Peeples)

Group5:  Auburn capital campaign (Aaron Mcdonald, Hopson Nance, Grantham Thomas, Brandon Traylor)

Group6: US News University Ranking (Micaela Creighton, Jena DeLaney, Alexandra Ess)

Group7: Auburn Football performance (Siqi Yang, Ke Zhang, Bing Zhao

Group8: Freshmen Enrollment (Travis Gajkowski)


Remote class:

Group1: School absences of Auburn City Schools (Amanda Little,  Eric Ezzell)

Group2: Local housing price (James Ratliff)

Group3: NCAA football player Injury (Jamie Lunn )

Group4: Weather prediction (Ravi Sagar)









The theme for the final project this year is "Making Auburn a Better Place to Live". All topics should be related to Auburn. I listed some topics here to give you a general idea about the type of topics that you might work on, but you are not required to choose one from this list. You are encouraged to come up with your own topics.


1) Investigate the usage of Auburn Bike Share System;

2) Sentiment Analysis of Auburn Football Fan page;

3) Auburn Real Estate Trend;

4) Student Enrollment Management

5) An innovative business idea: Parking Service, Smart City ...