This week, our team had the privilege of attending the Machine Learning in Science and Engineering Conference. The conference focused on showcasing machine learning research in interdisciplinary fields – our group mostly attended the talks on public policy and computer engineering. Some of the details were difficult to understand for undergraduates, but we got a sense of the depths of research going on in this field. We also attended the Women in Data Science workshop and got to hear from some inspiring women working with data nationwide.
(from left: David Reynolds, Olivia Fiol, Mirabel Reid, Billy Jang)
We focused on a particular research question: what is the impact of housing programs in Albany geared toward energy efficiency on utilities consumption? In order to build the database that we would use to answer this question, we investigated and cleaned weather and utilities data of Albany, answering the same 8 questions from the week prior. We also tried to incorporate census data, but ran into a lot of challenges; the website was confusing, and it seemed impossible to download all the data we needed at once. Ultimately, we decided to push this data collection off until we can access more resources.
Next, we all gathered in front of the whiteboard to brainstorm how we could link the different sheets together in a relational database format. This involved drawing connections between similar fields in the utilities and housing datasheets. We decided that the primary identifier would be the address (which would then be tied to parcel, block group, tract, and XY coordinate). This whiteboard map serves as our plan of action when we create the database.
Lastly, we got access to ArcGIS, which will allow us to investigate Albany spatially. Hopefully, by next week we’ll have created some enlightening maps to share with you.