Purpose: to data mine science journals (including images and figures) to distill knowledge to a more simplified human understandable data-set/outline of accepted scientific facts based on a search query.
Goals: To develop Free Open Source Software (FOSS) to achieve the purpose of this project, or by using existing resources, and to collaborate and facilitate team learning.
Team: Coders, science communications or other writers, project engineers, AI specialists. All levels of skill and interest welcome.
Timeline: It would be nice to complete this in 1 semester, but that is unrealistic, so a more organic approach will be used. Possible Schedule: Fall 2018: planning and general framework, reaching out to other departments for assistance as required. Spring 2019: execute project plan, or continue developing code created previously, (run tests on cluster with large data set?) Fall 2019: refine project as required.
Notes: The basic idea for this project is an extension of work done by the Geology Dept. with https://geodeepdive.org/ to a more generalized application in an effort to advance the pace of discovery and counteract the file-drawer effect, by considering where the data leads versus opinion. We would like to collaborate as a team being faithful to the scientific method, with a results bases approach, and using Kaizen Philosophy of continuous incremental improvement. We will be developing the process of this project from scratch with team input on all aspects so it should be a fun opportunity.
Future Directions: use the mining process, to discover unsettled questions in science, and further attempt to elucidate unknown unknowns. In an attempt to reduce duplicated effort in research for efficiency.
Current Members: Andrew Leicht (Sept 2018-), Lynn Liu (Sept 2018-)
Past Members: