Department of Mathematics

Proposed Project for the MSU Industrial Math Students


Data Mining Using Public Data Along With Internal Facial Data (FACES) From Numerous Participants Located All Over the World


This project will be looking at internal R&D facial data (FACES) from numerous participants located all over the world. The exact structure of the data along with variable descriptions will be given to the students at the kickoff meeting in December. Using this internal data the student will be asked to use publicly available datasets in conjunction with FACES to derive insights that are non-obvious, or that haven’t been able to be shown before. For example, there is a correlation of the pollution in an area to the number of wrinkles people in that area have.

During the first part of the project the students will be given some ideas of available public data and asked to evaluate their usefulness, not only for this current project, but future Amway projects. Evaluating other datasets, not on the list, will be encouraged. At the end of the first portion of the project, the students will choose a subset of databases to move forward with. The next section of the project is the data mining portion. Using the selected public data along with FACES, students will look for correlations, patterns and generate hypotheses using any methods that they see fit. The goal is being that we want better information to tell a story about how your daily life affects aspects of your physical experience. How can you show an individual the outcomes on their appearance from the decisions they make throughout their lives?

Documents that will be given to students:
  • FACES data with detailed variable explanation
  • List of potentila public databases to look at
  • Document with outcome of R&D brainstorm, discussing questions that they want answered about the data.
Project Goals:
  • Using open source software to perform data analysis
  • Show Amway the benefit of using public datasets along with internal data
  • Evaluation of public databases for use by Amway in the future
  • Interesting results from FACES data
  • Recommendations for additional data to be added to FACES data collection