| Name: | Nikki Cheng |
|---|---|
| Email: | cheng26y@mtholyoke.edu |
| Home Institution: | Mount Holyoke College |
| Project: | Data Analysis of COVID-19 patterns |
| Mentor: | Dr. Lazaros Gallos |
About My Project
Nikki is currently an undergrad student at Mount Holyoke College. She is majoring in Statistics and minoring in Computer Science and Chinese. Some areas of interest include machine learning and data analysis. Her favorite pastime includes photography, video editing, and PC building.
Research Log
Week 1: May 24 - May 28
For the first week, we had orientation on Monday. I had meetings with my mentor to go over what we'll be working on this summer. Throughout the week, I read papers on spatial correlation and other papers provided by my mentor that will help me prepare for the project. Since we have a mini-presentation on Tuesday, I have also been working on getting it completed and proofread. So far, this week went by pretty smoothly, and I'm looking forward to the next week.
Week 2: May 31 - June 4
This week went by fast. We had presentations on Tuesday, and I thought it went well. Everyone's project was exciting too. For the rest of this week, I plan to clean up the data and get started on python. I have a data set I want to use, and hopefully, by next week, I have the code up and running.
Week 3: June 7 - June 11
Aside from attending the tripods data science boot camp, I mainly worked on finding the missing fips code in the dataset. Unfortunately, too many missing fips code in the first dataset, so I have to use a different set. At the same time, I am working on a code to find how many cases are weekly then divided by their population in each county a.k.a Z(t).
Week 4: June 14 - June 18
I attended a seminar on Tuesday from Amy Ogan and on Thursday an Ethics in Research Workshop. Both were very interesting. Amy touched on educational technology and brought to light the problem with equality and opportunities that some students don't have access to. The Ethics Workshop was a good opportunity to remind students how to conduct research properly and showcase a history of unethical research. As for my project, I'm continuing to work on the code mentioned last week.
Week 5: June 21 - June 25
This week's seminar was on Tuesday, presented by Dr. Martin Tancer. He talked about the necklace splitting problem where a group of k thieves spited an open-ended necklace with t different types of gems. We want to split cut the gem as little as possible but every thief receive the same amount of gem a.k.a (k-1)t. Dr. Tancer showed us a combinatorial and a geometric solution to this problem. In regard to my project, I continue cleaning and detrending data this week.
Week 6: June 28 - July 2
On Tuesday we had a seminar from Dr. Mykhaylo Tyomkyn from Charles University, Prague. He presented on weak saturation on graphs and hypergraphs. He showed an example of weak saturation process by comparing it to the pandemic. If two edges of a triangle are infected then the third edge will also become infected. It will continue until no further infection is possible. One of the question Dr. Tyomkyn want to answer was what is the smallest number of initial infected edges can you have in order to infect all the edges of a graph. This week I'm working on putting the data through spatial correlation formula a.k.a C(r) equation. I will be looking for the correlation of each county at different size bin ranging from 100 km to 1000 km. In order to calculated the correlation I will be using the Z(t) values found from the previous week.
Week 7: July 5 - July 9
On Wednesday I attended AI day. There were presentations and seminar from many presenters giving various information about how AI can help people. I thought the tour through Rutgers' Robotics Lab was neat addition to the seminars. I wish I had to opportunity to tour the lab in person. Afterward there were games set up in the garden for students to get to know each other and have fun. It was a nice touch to end AI day. In regard to my project, this week I focused on putting the C(r) into code form on python, it was a bit challenging but doable.
Week 8: June 28 - July 2
This is the second to the last week of the REU program. I had many meeting with my mentors and fixed some errors on my code. On Tuesday we had a graduate student panel. We got to hear the experience of current graduate students and faculties gave us helpful pointers on how to apply for graduate school. I learned the most important thing to the admission group are recommendation letters. Understanding that not many students have research experience, recommendation letters are important to show what the student's work ethic is like and whether they will be the right fit for the program. Overall, I felt overwhelmed with the amount of information, but at the same time I'm less anxious about apply to a graduate program. About the progress of my project, I was able to get the C(r) equation working with the help of my mentor. As I was running the program with the active cases data to cross check with my mentor, I notice the values of C(r) were extremely small. That was not exactly the outcome I was hoping for, and I deduced that it had to be the Z(t) values that I calculated a few weeks back. Therefore this week was devoted to fixing bugs and checking to make sure everything works.
Week 9: July 19 - July 23
During my last week with the program I worked on putting the finishing touches on my project, and putting together my paper and presentation. Overall I had a really good experience and had a lot of support from my mentor. I have learned a lot and gotten a glimpse of what it is like to study and do research as a graduate student. I'm nervous to be presenting on Friday, but very excited to show what I have been working on this summer.
Acknowledgement
This work was carried out while the author Nikki Cheng (NC) was a participant in the 2021 DIMACS REU program at Rutgers University, supported by NSF grant CCF-1852215. Thank you to the DIMACS REU program for providing this opportunity and Dr. Lazaros Gallos for being a wonderful mentor.