Eindra Kyi's CPT Project 2019

From CSclasswiki
Revision as of 00:00, 22 September 2019 by Ekyi (talk | contribs) (How This Fits Into My General Education)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Project Description

This summer, I was part of the Self-E app team at Brown University Human-Computer Interaction Lab. The app was coded in Python, built in Android Studio, and deployed in Google Play store. It is an app which helps novices perform self-experiments. The goal is for people to be able to discover new things about themselves in a convenient yet rigorous way. This means that if someone wants to find out, for example, if caffeine improves their energy levels, they will be able to discover causal relationships through entering their experiment and following the instructions on the app. The app will calculate and interpret results for them automatically using Bayesian techniques and machine learning, while also including elements of scientific rigor in self-experiments that people who do it on pen and paper usually miss, such as phase randomization and experience sampling.

My Role

I helped design the website for the app, and iterated on the app design with real-time user feedback. I studied past data from other similar apps to find correlations in the data in Python. I presented relevant papers twice at the lab's paper club, and also presented a poster of our results at Brown University Summer Research Symposium, training my public speaking and communication skills. I led a pilot usability study with 16 participants in Providence, which includes recruiting people, interviewing them (I was the sole interviewer for 14 of them), and analyzing the data using quantitative and qualitative methods. I co-first authored a research paper on the work. I worked on analysis with my PhD supervisor initially, and then was able to independently do my 50-60% of it, splitting the work with the other co-first author (the PhD student). I did a literature review of about a little over 100 papers to get a sense of the subfield in HCI we were working within.

How This Fits Into My General Education

I became a better student researcher through this. Before this, I only had a semester of experience in HCI research as a Special Studies, which involved doing literature reviews and presenting once. While I did not carry out an actual study from finish to end, this Special Studies was foundational for building research skills such as reading papers and communicating results. My English classes and Philosophy classes at Smith helped with the writing of the paper immensely due to the critical reading skills I acquired. My senior seminar in NLU taught me how how to condense technical information in concise ways. My Algorithms and Statistics classes were useful in understanding the workings of the app, such as how the app calculates the likelihoods that a lifestyle intervention is effective for someone.