Inducing Wrappers for OCRed Lists to Efficiently Populate Ontologies

This is an exploration of data- and software-driven approaches to pulling useful data out of semi-structured text in electronic format that may or may not contain OCR errors and other noise. The methods we outline are more accurate and cost-effective in terms of human time than previously existing methods and tools. The resulting data is highly structured and useful for many kinds of down-stream applications.

Autorentext

Thomas L. Packer earned a PhD in computer science at Brigham Young University. He pursues a career in data science and applied research. He is married with five children and is an active member of the Church of Jesus Christ of Latter-day Saints.