Hi Everyone,
Are you interested in digitization workflows for natural history museum collections that involve OCR, machine learning and natural language processing? iDigBio's first Hackathon starts this morning, Wednesday February 13th and begins with 10 minute lightning talks by participants sharing their results so far to parse OCR output. Participants will spend 2 days together sharing and tweaking algorithms aimed at making parsing OCR output better. Other big topics of the hackathon are image processing/segmentation/analysis and also the continued need for user-interface development for interacting with the OCR output (front-end human-in-the-loop and api interfaces).
Tune in: http://idigbio.adobeconnect.com/augmentocr 9AM - to about noon CST. Read more at: http://tinyurl.com/aocrhackathonwiki Works better with Firefox (b/c of Flash). Headsets are a good idea for best sound. From all of us at the Hackathon, See you there. Deb