We are a home for Earth science data and computing professionals. Our sessions bring together the community for hands-on, interdisciplinary deep dives as we explore "Innovation to Impact" this year. Learn more about ESIP: esipfed.org
Description With the advent and popularity of ChatGPT and other chatbot technologies, we see an opportunity to improve the quality of search and discovery within earth science data archives. Multiple machine learning technologies could be brought to bear on the problem of producing high quality results from earth science queries. Those technologies include, but are not limited to: Chatbots Knowledge Graphs Large language models Vector embeddings
These technologies could be applied to both the search of earth science archives, where the user has an idea of what they are searching for, and the discovery of data, where the user may discover new knowledge from existing data.
We invite agencies, organizations and individuals to present their experiences with the above technologies applied to earth science search and discovery and what they plan to do in the future so that we can learn and provide better solutions to our community.
Agenda
Talk: Use of natural language to discovery science data Presenter: Ryan Abbott
Talk: Measuring and adjusting the quality of responses to AI searches Presenter: Deep Mistry
Talk: UX considerations for mixing AI and user interfaces Presenter: Trevor Lang
Talk: Impact the quality of underlying data has on our ability to discover it Presenter: Ryan Abbott
Talk: Discovering signal in the noise - making sense of the torrent of metrics and data Presenters: Jonathan Blake and Han Mai
Value to Session Participants: NASA ESDIS are about to make it easier to find your data. Hopefully.
Recommended Ways to Prepare for this Session: Engineer your prompts!
