ABOUT ME

Research Obsessed

My middle school ‘Science Talent’ classes imbued me with a strong passion for research, data collection, and clinical writing that has not waned to this day. As an undergraduate and graduate student, I gained practical skills in research design, questionnaire creation and distribution, and statistical analysis. I spent several years volunteering, and subsequently working for doctors at New York Presbyterian Weill-Cornell assisting them with their research projects.

Despite my love for the work I was doing, I knew medicine was not the path for me. A course in linguistics sparked an interest in what is now my current career―Speech-Language Pathology. While there are many aspects of my work that I enjoy greatly, such as directly being able to help others and facing new challenges daily, I feel as though my true passion for data and research remain unfulfilled.

To scratch this itch, I decided to pivot my career to data analytics. My goal is to combine all of my knowledge and skills and direct them towards solving problems that result in work which is meaningful and beneficial to others.

​“Information is a beacon, a cudgel, an olive branch, a deterrent―all depending on who wields it and how.”
​- Steven D. Levitt

PORTFOLIO

Project: Dunder Mifflin Salesperson Analysis

I analyzed a variety of sales metrics utilizing Pivot Tables to determine which employee is to be promoted to the Sales Lead position (spoiler: it was Dwight).

Project: Amsterdam Airbnb Analysis

Using "Inside Airbnb" data for the city of Amsterdam, I utilized SQL to join multiple tables, calculate bookings, and identify which properties have high and low availability. With this dataset, I created a presentation to recommend areas in which to invest.

Fig. 1 :
The average monthly revenue is €1,160 with an average of 3.3 monthly bookings across all neighbourhoods. The boroughs with the most profitable neighbourhoods are Centrum, Zuid, and West.  Additionally, Oost and Noord have some potential areas to consider.

Fig. 2:
The data shows that the maximum average monthly income is reached in a 6 bedroom property. It was found that 88% of listings accommodate between 2 and 4 people.

Of the 9,771 listings, only 234 (or 2.4%) accommodate more than 6 individuals.

Fig. 3:
The ratings of the above-average scoring Neighbourhoods was analyzed, narrowing down the field of potential property locations to a top six.

Additionally,  an analysis was conducted to determine which neighbourhoods were in high demand for hosts. As seen in the accompanying graph, 4 of the 6 previously identified areas have shown the greatest market demand from 2009 to 2015.

Fig. 4:
The data suggests that the greatest profit could be made by purchasing a house in Zuid and renting out the entire home.

On average, entire homes rented in Zuid have an income of €2,137 and are used 7.6 days per month.

The second highest grossing neighbourhood is Westerpark where an entire home averages €1,404  (a significant drop in potential revenue)

Fig. 5:
A deeper analysis of Zuid data shows 9 distinct zip codes, with a total of 57 house listings whose average income was determined to be €2,153.

A home purchased in this area, rented out in its entirety would produce maximal profits.

Project: Converting Non-Subscriber Citi Bikers

Using Citi Bike ridership data for June 2015, my task was to identify five stations that had the greatest potential for marketing to non-subscribers in the hopes of converting them to full subscribers. I utilized SQL to explore the data, remove invalid data (e.g., rides lasting fewer than five minutes). Once entered into Excel, I was able to determine the best days, stations, and times to reach non-subscribers.

Fig. 1:
All stations were analyzed to determine which had the highest total riders, and then further parsed to find those which the highest ratio of non-subscribers.

It was found that the Central Park S and 6th Avenue station has a significantly higher ratio of non-subscribers to subscribers (71%)

Fig. 2:
Central Park S and 6th also has the greatest number of riders; particularly on Saturday and Sunday. For this reason, in order to reach the greatest number of non-subscribers, this station would be ideal to target both weekend days.

Time of day was then determined by greatest frequency of both subscribers and non-subscribers hourly.

borkina.anna@gmail.com
linkedin
resume