|Project:||Data Security in Life Sciences|
|Home Institution:||Barnard College of Columbia University|
My project aims to design software that allows scientists doing life science research to share data with collaborators securely and efficiently. Such software would allow researchers to collaborate with more control over how their data is shared and the documentation used for it.
The goal of the summer research project is to understand the existing body of work on this topic, brainstorm solutions for a user interface to address needs based on this research, and wireframe a final system.
This week, I read many journal articles to gain a better understanding of the research focus for this summer. The articles generally fell into one of two categories (though there were a few outliers): ethnographic studies of current data management practices and analyses of Attribute-Based Access Control systems and their fucntionality.
This week, I organized my findings from Week 1 to gain insights into opportunities for new research and questions for later interviews. I also presented an introduction to my project and some of this related work to the DIMACS REU participants (a version of the slides can be found here). Lastly, I found further work that relates to this area of research to gain a broader understanding of the prior work on this topic.
This week, I continued to research more about data sharing and security to find inspiration for future work and interview questions for data scientists. I looked, in particular, at current security mechanisms such as cryptography and data anonymization to familiarize myself with the system data scientists are currently using. Lastly, I attended a couple lectures of the Data Sceince Bootcamp hosted by DIMACS to gain a better understanding of common data science tools.
This week, I started to work on my final paper. Morgan and I wrote an Abstract and a Related Works section for our research. We summarized many of the papers that we read over the prior weeks, organized them into sections, and explained how they relate to what we will be working on. We also analyzed the gaps in the prior research and how our project aims to fill those gaps.
This week, Morgan and I began the brainstorming process for our final design. We began by creating four personas of different scientists who may use our system. Each of these personas had different jobs, needs, and technical skills which were informed by research we had done in prior weeks. Using these personas and the research we had done, we brainstormed many different features that could help solve our research question. We then refined our choices and outlined a final solution that considered speed, simplicity, and control. Lastly, we wrote scenarios and created storyboards that showed how our personas and users would benefit from the system that we created.
This week, I began to create a visual representation of the system and features we came up with last week. Using the software Balsamiq, I created wireframes for the different pages a user would go through in our system that showed the different functionality and options available to the user. Then, I mapped the navigation between these wireframes, showing how a user would switch between pages and use the system. I also continued to work on our abstract and related work, incorporating input from peers.
This week, I finished and presented the wireframes and navigation for our proposed system to my mentor. After presenting, I continued to work on our final paper, specifically the Limitations & Future Work section.
This week, Morgan and I finished a first draft of our final paper, which was >6000 words long. I focused on the System section, which describes our work over the past two months, and the conclusion. After finishing the draft, I also edited another team's paper and created our final presentation for DIMACS.
For the final week for DIMACS, Morgan and I prepared for our final presentation of our findings over the course of the summer. We began by creating a presentation that synthesized the topics in our paper, including the Literature Review, Method, and final Wireframes. Before presenting to the larger DIMACS program, we presented to other students in the Barnard Computer Science - Summer Research program. In that presentation, we also answered questions about our research and gave feedback to other students on their presentations. We then attended presentations from the other REU students and learned about their research topics. The week ended with giving our final presentation to all participants.
To finish off our summer research, Morgan and I created a plan for a future system evaluation for our project. We conducted research into what system evaluations typically include, and then designed an experiment from there. We chose questions for participants to answer and came up with scenarios that will show how users would interact with our system. We also came up with a grading schema, on how we would evaluate responses to the scenarios. Lastly, we made decisions as to how many participants we would need, whether the individuals running the experiment would intervene, and what our independent and dependent variables are, all grounded in the research we did.