DSfS Community Night: Cybersecurity in Energy, Solar Leads
Details
Recommended $5 contribution (http://paypal.me/DSfSustainability)
Featuring interesting talks from our own members: David Stier, Daniel Roesler
First speaker: David Stier (http://linkedin.com/in/davidjstier)
Who wants solar? How installers in Boston could save $90,000 using machine learning on public datasets
Lowering the soft costs of installing residential solar has become a key issue for increasing the number of residential solar installations. This presentation provides a demonstration of how we can source sales leads at a lower cost by applying machine learning to datasets now available by many local governments (e.g. https://data.sfgov.org/ (https://data.sfgov.org/)).
NREL has a number of programs to understand who will install solar and some modeling has been done to predict adoption rates. My work is the first project to provide a household level prediction score for more than 20,000 homes on the likelihood that a particular homeowner will install solar. I've shared my results with the director of NREL's SEEDS program and we are in discussions about co-authoring a report. The goals of this presentation are:
-
Provide a framework for assessing the information gain from feature engineering.
-
Offer an example of how framing a complex issue can yield impactful business results.
-
Increase awareness of currently available public datasets that are extensive and rich in detail
Second speaker: Daniel Roesler (http://www.intersolar.us/en/conference/speaker/daniel-roesler-3810.html)
A Cybersecurity and Privacy Checklist for Data Scientists in Energy
Daniel Roesler is the Co-founder and CEO at UtilityAPI. He has a degree in chemical engineering from the University of Texas at Austin and nine years of experience in the energy and software sectors.