Summer Research, 2017
Student: Lisa Oshita
Faculty advisor: Shannon Pileggi
Industry advisor: Anthony Pileggi
Objective
The objective of this summer research is to utilize survival analysis techniques to model time until a question is answered from iFixit's Q&A forum. (https://www.ifixit.com/Answers)
Specific Aims
- Utilize GitHub to collaborate on project materials and updates.
- Write all R code according to Hadley Wickam's Style Guide (http://adv-r.had.co.nz/Style.html). All R code should be written in a reproducible manner, such that code will execute when applied to a new data set.
- Thoroughly explore the iFixit Answer's forum in conjunction with the data to understand what the data represents.
- Perform a literature review to determine if other researchers have identified characteristics associated with either likelihood or timeliness of a question being answered.
- Identify potential parametric distributions suitable for the time to event data. Compare pros and cons of parametric versus nonparametric models.
- Create new variables necessary to model time until a question is answered. For example, this may involve classifications of product type or parsing text strings to identify if a question mark is present.
- Research the incorporation and interpretation of ptential time varying covariates. Present an argument for the best treatment of time varying covariates, if appropriate.
- Create a model to predict time until question is answered.
- Turn this model into a function that can be applied to any new data set
- Allow users to calculate or sort through predictions for specific questions
- Write a user manual on how to utlize the function and interpret the results
- Take some data camp courses to learn new R skills (listed in order of priority).