We'll cover turning scans and digitized materials into structured data using optical character recognition (OCR) and handwritten text recognition (HTR) tools, then focus on evaluation: using the structure itself, such as agreement across a row, cross-checking a name against a tenure date, to catch errors instead of trusting model output at face value. This session frames a shift that runs through the whole series: from writing code to reviewing and evaluating what a model produces.

Participants will come away with a working OCR/HTR workflow for their own image materials and a concrete set of evaluation techniques to check predictive systems' output for accuracy and consistency before relying on it in their research.