Survey of India Topographical Survey Reports Complete Table-Only Extractions

This dataset contains a comparison of output data from The Coleridge Fellowship 2025-26. Python code was compiled to extract entities relating to people and places from xml files exported from the Transkribus Optical Character Recognition platform. The project as a whole covers the years 1864 to 1873. For one year, 1865, a subset of data was also extracted manually and using several online AI services. This dataset brings together the different outputs of people's names and of places from the Python, manual and AI extractions to facilitate comparison of the output data. It is an Excel file.