MONK in Practice: Indexing Heterogeneous Handwritten Collections

University of Twente Research Information

Abstract

This short paper describes how MONK, a machine-learning driven handwriting recognition system, can be used to rapidly index a heterogeneous handwritten collection with the help of volunteers. We discuss the setup and results of an event which saw volunteers come together to enrich a subset of the digitized Prize paper collection, a collection of historical handwritten documents of the High Court of Admiralty (1652-1815).

Authors 3

1. Anna Caceres Aachen

   Leiden University

   Affiliation as printed

   Leiden University

2. Andreas Weber

   University of Twente

   Affiliation as printed

   University of Twente

   Science, Technology & Policy Studies

3. Lambert Schomaker

   University of Groningen

   Affiliation as printed

   University of Groningen