My name is Victoria Sabo and I am a mathematics and Spanish double major at Georgetown University. I am very interested in applying math to problem solving in the real world, such as using programing and data in security, population modeling, analyzing businesses, or even tracking supermarket inventory to minimize product waste.

My research interests are why I applied to the SAMSI undergraduate workshop from May 14-19. The workshop gave me the opportunity to apply mathematics and computer science to realms usually isolated from the sciences. Based on the description, I imagined being exposed to new applications of math, stats, and computing while having the opportunity to harness my mathematics knowledge to solve an actual problem that I may not have known could be solved using the skills of a mathematics major. In the end, I gained ample skills, both academic and professional, and I was able to test them out while working on my own group research project.

After dedicated postdocs presented the overviews of six projects, we were allowed to rank our top choices:

- Lightcurve Classification for Periodically Varying Stars
*(Light Curves Project)* - Distributionally Robust Stochastic Programming for Financial Applications
*(Finance)* - Finding Exoplanets Using Radial Velocity Data
*(Exoplanets)* - Automatic Genre Classification of Music Pieces
*(Music)* - Time Delay Estimation for Gravitationally Lensed Light Curves
*(Time Delay)* - Data Assimilation for Numerical Weather Prediction
*(NWP)*

I was fortunate enough to receive my first choice which was the ** Automatic Music Genre Classification** project. That meant for the entire week, I would work on a team to investigate algorithms used for supervised learning, where training data taken from a music dataset, to be used to create a system for predicting the genres of unlabeled songs.

When we first met in groups, we discussed how to read the data and began thinking of probability techniques common to machine learning that would be useful for the task. We read scholarly articles about previous approaches to the problem, then met the following day to begin coding programs based on our dataset.

– Kevin Multani, Applied Science, Department of Engineering Physics, University of British Columbia – Vancouver, Canada“I was pleasantly surprised at the diversity of the attendees at the workshop. The backgrounds of the students ranged from civil engineering, to a double major in math and piano…This variety in background facilitates the sharing and cross-pollination of ideas from different fields, which I deeply appreciated.”

As the week went by, we experimented with different combinations of song features, such as *loudness*, *danceability*, and *song_hotttnesss* (no, not a typo), and various techniques. The techniques, used for coding the data, aimed at achieving the highest accuracy in song genre classification. The techniques included: k-means clustering; k nearest neighborhood; Gaussian classifiers; PCA; and t-SNE. Through this process it was very interesting to note the limitations on our research and how the attributes, such as the data set qualities or the time constraint, affected what we could accomplish. Overall, this research project introduced me to what it was like to work on a team to conduct formal research. I also enjoyed spending the week bouncing ideas off of my other group members as we worked to solve a problem found at the intersection of two distinct subjects: math and music.

Besides just the experience of working in a research group, I created lifelong memories from this workshop thanks to the incredibly intelligent people I had the pleasure of meeting. I was introduced to undergraduates from across the United States and Canada, many of whom had international backgrounds as well. Everyone possessed a unique skill set, from their university, when it came to computer programing. The diverse backgrounds of every participant contributed to the success of the research project because of the various courses taken by the undergraduate students. I loved hearing about everyone’s majors and their career goals. I found it was invaluable to be able to exchange advice with people who were as interested in the sciences as myself. During meals and breaks, we would discuss our intended graduate study goals as well as past research we had conducted thus far. I was given advice on what conferences to attend and which schools were best for certain master’s degree or Ph.D. programs. I definitely have reevaluated my future plans since conversing with and listening to such a wide range of science and math students.

One of my peers in the workshop, Kevin Multani, an undergraduate student from the University of British Columbia – Vancouver, Canada had similar points to share:

I was pleasantly surprised at the diversity of the attendees at the workshop. The backgrounds of the students ranged from civil engineering, to a double major in math and piano — there was even a student who was double majoring in Philosophy and History (of Mathematics)! This variety in background facilitates the sharing and cross-pollination of ideas from different fields, which I deeply appreciated. Most of my learning came from discussion and conversation with the students and mentors. In fact, through conversation with my mentor, David Jones, I’ve gained a solid understanding on what to expect for graduate school. Overall, the SAMSI Undergraduate Workshop was a refreshing experience, both personally and academically.

Even though the friends that I made during the week were an enriching part of this SAMSI undergraduate workshop experience, the panels and talks organized for us also made an impact on me academically. We received information on North Carolina State University’s master’s program for science in analytics, since the Institute for Applied Analytics, where the event was hosted, is located on the university’s campus. I came out of this workshop with a broader understanding of the great career opportunities in data analytics. Thanks to the talk from Michael Rappa on opportunities in data analytics and his program within the institute, my eyes were opened as to how many different applications of data analytics there are for people with those skills. For instance, I had never considered that someone with a math background was needed to calculate the appropriate amount of supermarket inventory to prevent over and under stocking? Likewise, I did not know that companies hired analysts to evaluate their businesses in order to maximize the efficiency of their hiring efforts. Due to my interests in applying math to real world problems, I am now going to focus my efforts on exploring this area as a possible career path. I am also looking forward to augmenting my computer programming skills because I recognize now, that for these types of jobs, coding and programming, in addition to a solid linear algebra and classical mathematics background, are essential skills for the type of work in which I am interested.

I entered this SAMSI workshop as a mathematics major, but I lacked the knowledge of how I could put that degree to good use applying math knowledge to real world problems. After the workshop however, I have now conducted research in an application of math to music; something I never imagined was possible!

I was also introduced to countless other opportunities available for individuals trained in math, computer science, and analyzation techniques. I feel that by taking more courses geared towards applications of math in the real world, I can better prepare myself to succeed in a career in data analytics. Additionally, I am now informed on what it takes to create a successful application to graduate school and which programs I should consider that will best prepare me for a productive and fulfilling future.

Therefore, this undergraduate research workshop not only provided me with research, public speaking, and teamwork experience, but it also educated me on what options exist for my future. Although I have much more to think about, SAMSI was a starting point in helping me determine where I would like to see myself in the coming years and helped to catalyze the best way for me to utilize my mathematics and computing knowledge to benefit others in the future.