Diploma Thesis: Unsupervised Post-Correction of OCR Errors

The trend to digitize (historic) paper-based archives has emerged in the last years. The advantages of digital archives are easy access, searchability and machine readability. These advantages can only be ensured if few or no OCR errors are present. These errors are the result of misrecognized characters during the OCR process. Large archives make it unreasonable to correct errors manually. Therefore, an unsupervised, fully-automatic approach for correcting OCR errors is proposed. The approach combines several methods for retrieving the best correction proposal for a misspelled word: A general spelling correction (Anagram Hash), a new OCR adapted method based on the shape of characters (OCR-Key) and context information (bigrams). A manual evaluation of the approach has been performed on The Times Archive of London, a collection of English newspaper articles spanning from 1785 to 1985. Error reduction rates up to 75% and F-Scores up to 88% could be achieved.