Back in May, my colleague and I had some of our research from our Spring 2017 machine learning class accepted for the poster session at JupyterCon, an O’Reilly and NumFOCUS Foundation Conference coming to New York City next week. Today I want to describe the process we had for preparing for this event.
One of us is in Washington, DC while the other teammate is in San Diego, CA. I think we maximized this time distance by splitting up the work so that the East Coast person completed work earlier in the day local time and the West Coast person completed work assignments later in the day local time. We used the myriad of collaboration tools to add some touches to the project even though we have never met each other face-to-face.
Our research focus was applied big data for social good and we looked at using economic aid data and machine learning to predict future aid amounts. To our knowledge, machine learning has not been used yet in this context. During our spring 2017 class, we wanted to use machine learning to predict the aid amount to foreign countries to the dollar amount which did not work. We discovered that the data amounts were too varied from 2014 to 2016. So our next hypothesis was to predict aid amount after the discrete data was grouped into categories using data science binning techniques. This analysis worked a lot better with our algorithms. (So as not to spoil the poster, I’ll be sharing the final version a bit later after the conference for those that are interested.)
After we finished the research, we used the remaining time between the conclusion of our summer courses and the conference to design and print the poster. Because we knew there might be over 30 posters at this session, we wanted the design to be simple and eye catching with less words. We thought of our poster as an appetizer rather than the main course of the research we had done. We’re very excited for this opportunity to discuss our research with many data scientists and Python practitioners.
Stay tuned for an upcoming blog where I share the poster with you!