The proceedings are scanned and then we use optical character recognition (OCR) software to make the articles searchable.This technology is not fool-proof. Especially for the earlier proceedings where the letters are sometimes very unclear, not all characters are recognized correctly. This is not a problem that can be fixed automatically.
The main purpose of providing the recognized text along with the scan, is to allow the proceedings to be searched. Often, an important word will occur in the article more than once so the article will still be found when searching for that term, even if some OCR mistakes have been made.
Because the proceedings archive consists of thousands of pages, the time it would take to proofread all the articles would be disproportional to the benefits.
The proceedings are scanned and then we use optical character recognition (OCR) software to make the articles searchable.This technology is not fool-proof. Especially for the earlier proceedings where the letters are sometimes very unclear, not all characters are recognized correctly. This is not a problem that can be fixed automatically.
The main purpose of providing the recognized text along with the scan, is to allow the proceedings to be searched. Often, an important word will occur in the article more than once so the article will still be found when searching for that term, even if some OCR mistakes have been made.
Because the proceedings archive consists of thousands of pages, the time it would take to proofread all the articles would be disproportional to the benefits.