Field Journal Fix-Up

The Academy of Natural Sciences archives contains manuscripts, correspondence, films, photographs, art, and more. The archives also contain field journals, which are notebooks kept by naturalists as they travel on expeditions to collect specimens and conduct research. The journals hold many types of data including biological species names, weather reports, and environmental descriptions. They also provide glimpses into the lives of naturalists such as their daily routines and internal thoughts.

Because field journals are handwritten, traditional attempts at Optical Character Recognition (OCR) have failed to transcribe them. The Academy is using Amazon Textract, a machine-learning software, for a first attempt at transcription. This AI technology is never 100% accurate, so we are seeking help to correct the transcripts and unlock an incredible source of biodiversity data and first-hand stories of science! Once digitized and transcribed, the journals are widely available and text-searchable.