Handling Heavily Abbreviated Manuscripts: HTR Engines vs Text Normalisation Approaches. Camps, J., Vidal-Gorène, C., & Vernet, M. In Barney Smith, E. H. & Pal, U., editors, Document Analysis and Recognition – ICDAR 2021 Workshops, of Lecture Notes in Computer Science, pages 306–316, Cham, 2021. Springer International Publishing. doi abstract bibtex Although abbreviations are fairly common in handwritten sources, particularly in medieval and modern Western manuscripts, previous research dealing with computational approaches to their expansion is scarce. Yet abbreviations present particular challenges to computational approaches such as handwritten text recognition and natural language processing tasks. Often, pre-processing ultimately aims to lead from a digitised image of the source to a normalised text, which includes expansion of the abbreviations. We explore different setups to obtain such a normalised text, either directly, by training HTR engines on normalised (i.e., expanded, disabbreviated) text, or by decomposing the process into discrete steps, each making use of specialist models for recognition, word segmentation and normalisation. The case studies considered here are drawn from the medieval Latin tradition.
@inproceedings{campsHandlingHeavilyAbbreviated2021a,
address = {Cham},
series = {Lecture {Notes} in {Computer} {Science}},
title = {Handling {Heavily} {Abbreviated} {Manuscripts}: {HTR} {Engines} vs {Text} {Normalisation} {Approaches}},
isbn = {978-3-030-86159-9},
shorttitle = {Handling {Heavily} {Abbreviated} {Manuscripts}},
doi = {10.1007/978-3-030-86159-9_21},
abstract = {Although abbreviations are fairly common in handwritten sources, particularly in medieval and modern Western manuscripts, previous research dealing with computational approaches to their expansion is scarce. Yet abbreviations present particular challenges to computational approaches such as handwritten text recognition and natural language processing tasks. Often, pre-processing ultimately aims to lead from a digitised image of the source to a normalised text, which includes expansion of the abbreviations. We explore different setups to obtain such a normalised text, either directly, by training HTR engines on normalised (i.e., expanded, disabbreviated) text, or by decomposing the process into discrete steps, each making use of specialist models for recognition, word segmentation and normalisation. The case studies considered here are drawn from the medieval Latin tradition.},
language = {en},
booktitle = {Document {Analysis} and {Recognition} – {ICDAR} 2021 {Workshops}},
publisher = {Springer International Publishing},
author = {Camps, Jean-Baptiste and Vidal-Gorène, Chahan and Vernet, Marguerite},
editor = {Barney Smith, Elisa H. and Pal, Umapada},
year = {2021},
keywords = {Abbreviations, Handwritten text recognition, Medieval western manuscripts},
pages = {306--316},
}
Downloads: 0
{"_id":"AZNPywoHRnrjk6NsX","bibbaseid":"camps-vidalgorne-vernet-handlingheavilyabbreviatedmanuscriptshtrenginesvstextnormalisationapproaches-2021","author_short":["Camps, J.","Vidal-Gorène, C.","Vernet, M."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"Cham","series":"Lecture Notes in Computer Science","title":"Handling Heavily Abbreviated Manuscripts: HTR Engines vs Text Normalisation Approaches","isbn":"978-3-030-86159-9","shorttitle":"Handling Heavily Abbreviated Manuscripts","doi":"10.1007/978-3-030-86159-9_21","abstract":"Although abbreviations are fairly common in handwritten sources, particularly in medieval and modern Western manuscripts, previous research dealing with computational approaches to their expansion is scarce. Yet abbreviations present particular challenges to computational approaches such as handwritten text recognition and natural language processing tasks. Often, pre-processing ultimately aims to lead from a digitised image of the source to a normalised text, which includes expansion of the abbreviations. We explore different setups to obtain such a normalised text, either directly, by training HTR engines on normalised (i.e., expanded, disabbreviated) text, or by decomposing the process into discrete steps, each making use of specialist models for recognition, word segmentation and normalisation. The case studies considered here are drawn from the medieval Latin tradition.","language":"en","booktitle":"Document Analysis and Recognition – ICDAR 2021 Workshops","publisher":"Springer International Publishing","author":[{"propositions":[],"lastnames":["Camps"],"firstnames":["Jean-Baptiste"],"suffixes":[]},{"propositions":[],"lastnames":["Vidal-Gorène"],"firstnames":["Chahan"],"suffixes":[]},{"propositions":[],"lastnames":["Vernet"],"firstnames":["Marguerite"],"suffixes":[]}],"editor":[{"propositions":[],"lastnames":["Barney","Smith"],"firstnames":["Elisa","H."],"suffixes":[]},{"propositions":[],"lastnames":["Pal"],"firstnames":["Umapada"],"suffixes":[]}],"year":"2021","keywords":"Abbreviations, Handwritten text recognition, Medieval western manuscripts","pages":"306–316","bibtex":"@inproceedings{campsHandlingHeavilyAbbreviated2021a,\n\taddress = {Cham},\n\tseries = {Lecture {Notes} in {Computer} {Science}},\n\ttitle = {Handling {Heavily} {Abbreviated} {Manuscripts}: {HTR} {Engines} vs {Text} {Normalisation} {Approaches}},\n\tisbn = {978-3-030-86159-9},\n\tshorttitle = {Handling {Heavily} {Abbreviated} {Manuscripts}},\n\tdoi = {10.1007/978-3-030-86159-9_21},\n\tabstract = {Although abbreviations are fairly common in handwritten sources, particularly in medieval and modern Western manuscripts, previous research dealing with computational approaches to their expansion is scarce. Yet abbreviations present particular challenges to computational approaches such as handwritten text recognition and natural language processing tasks. Often, pre-processing ultimately aims to lead from a digitised image of the source to a normalised text, which includes expansion of the abbreviations. We explore different setups to obtain such a normalised text, either directly, by training HTR engines on normalised (i.e., expanded, disabbreviated) text, or by decomposing the process into discrete steps, each making use of specialist models for recognition, word segmentation and normalisation. The case studies considered here are drawn from the medieval Latin tradition.},\n\tlanguage = {en},\n\tbooktitle = {Document {Analysis} and {Recognition} – {ICDAR} 2021 {Workshops}},\n\tpublisher = {Springer International Publishing},\n\tauthor = {Camps, Jean-Baptiste and Vidal-Gorène, Chahan and Vernet, Marguerite},\n\teditor = {Barney Smith, Elisa H. and Pal, Umapada},\n\tyear = {2021},\n\tkeywords = {Abbreviations, Handwritten text recognition, Medieval western manuscripts},\n\tpages = {306--316},\n}\n\n","author_short":["Camps, J.","Vidal-Gorène, C.","Vernet, M."],"editor_short":["Barney Smith, E. H.","Pal, U."],"key":"campsHandlingHeavilyAbbreviated2021a","id":"campsHandlingHeavilyAbbreviated2021a","bibbaseid":"camps-vidalgorne-vernet-handlingheavilyabbreviatedmanuscriptshtrenginesvstextnormalisationapproaches-2021","role":"author","urls":{},"keyword":["Abbreviations","Handwritten text recognition","Medieval western manuscripts"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://api.zotero.org/groups/2386895/collections/7PPRTB2H/items?format=bibtex&limit=100","dataSources":["u8q5uny4m5jJL9RcX"],"keywords":["abbreviations","handwritten text recognition","medieval western manuscripts"],"search_terms":["handling","heavily","abbreviated","manuscripts","htr","engines","text","normalisation","approaches","camps","vidal-gorène","vernet"],"title":"Handling Heavily Abbreviated Manuscripts: HTR Engines vs Text Normalisation Approaches","year":2021}