I scraped over 100,000 full-text articles from the Pub Med API to estimate how common code sharing is across different journals.
Insight Data Science is a popular fellowship for PhDs going into data analytics. I wanted to get a better sense of where fellows came from and ended up, so I scraped some data from the Insight website and analyzed it.
There are many data science fellowships available after graduate school, so I organized details about the ones I have found as a potentially useful resource for myself and others.
Personalized scavenger hunts are fun, they make good gifts, and they can actually be made pretty quickly. Here’s some advice on how to make one, as well as a full example.
After debating with a friend about how much/little we need to save in order to retire comfortably, I made an interactive visualization that shows how much money you will earn off interest as a function of your savings over time.
Using data collected by the US Bureau of Transportation Statistics, we analyzed the relationships between basic properties of a flight (e.g. time of day, airline) and how much they were delayed. We also trained a classifier to predict if a flight would be delayed.
We believe that the waveform shape of brain rhythms should be analyzed to extract more biological information from neural recordings.
The Open Science Grid is a free supercomputing resource for academics. This step-by-step tutorial will allow any researcher to begin running their Python-based analysis using high-throughput computing for free.
I analyzed the geographic distribution of poster viewership for posters presented at the SfN 2016 annual meeting. Posters from some states (Minnesota) and countries (Netherlands) are more popular than others. But not significantly.
At the annual neuroscience conference, I collected data to quantify the popularity of thousands of presented posters. As a first analysis, I related poster popularities to 8 of the major themes in neuroscience.
Twenty-eight people applied burritology to asses their experiences eating burritos at the famous Lucha Libre Taco Shop in San Diego.
The United States is dominating in the Olympic medal count, but maybe that’s because of the disproportionate number of medals in swimming. What would the results look like if the number of medals was even for all sports?
Interactive visualization to set weights to each medal category to visualize performance across the globe. Playing around with data visualization in Tableau Public using the Rio Summer 2016 Olympic medals dataset.
Rather than extreme zooming on small figure panels, using simple image processing, we can extract an estimate of signals plotted in papers.
A group of San Diegans quantified over 100 burrito experiences by decomposing their meal into 10 dimensions. This post describes the data and has some preliminary analysis.
Phase-amplitude coupling is a common analysis on neural oscillations. But in order to obtain meaningful results, we need to first preprocess the signal.
Rhythmic signal analysis can be improved with a transform our of the time domain. While Fourier techniques are traditionally applied, EMD offers an alternative approach to frequency analysis.
My take on a recent neuroscience study in which researchers could stimulate the brain to evoke a forgotten memory
A different perspective on how much having a perfect undergrad GPA really matters.
My first attempt at reviewing carne asada burritos across the city.
Inspired by a recent Numberphile episode, I explore why exactly Farey sequences will produce the Ford Circles fractal.
A search for the optimal strategy to win in a competitive number guessing game
A response to a recent Numberphile video which proposed a strategy for guessing random numbers. This new strategy is optimal.
My naive thoughts on the popular phenomena in neuroscience, memory replay