=Paper=
{{Paper
|id=Vol-2402/paper7
|storemode=property
|title=Multimodal Datasets of the Berlin State Library
|pdfUrl=https://ceur-ws.org/Vol-2402/paper7.pdf
|volume=Vol-2402
|authors=David Zellhöfer
|dblpUrl=https://dblp.org/rec/conf/ldk/Zellhofer19
}}
==Multimodal Datasets of the Berlin State Library==
https://ceur-ws.org/Vol-2402/paper7.pdf
Multimodal Datasets of the Berlin State Library
David Zellhöfer
Berlin State Library/Staatsbibliothek zu Berlin, Germany
https://staatsbibliothek-berlin.de/
david.zellhoefer@sbb.spk-berlin.de
Abstract
To facilitate the handling of digital library content and its accompanying metadata, four multimodal
and multilingual datasets are presented that are relying on the publicly available information systems
of the Berlin State Library. They range from pre-processed extracts of the full main catalog of the
library with ca. 9.8 million records, over various networks graphs modeling, e.g., relations between
authors and languages, to more than half a million extracted illustrations detected by the day-to-day
OCR process of ca. 22,000 historical media units such as historical books.
2012 ACM Subject Classification Applied computing → Digital libraries and archives
Keywords and phrases Digital Library, Bibliographic Metadata, Graph, Digitized Media, Dataset
Supplement Material https://github.com/elektrobohemian/StabiHacks
Funding Funded by the BMBF as part of the ‘QURATOR - Curation Technologies‘ project.
1 Introduction
Most large research libraries such as the Berlin State Library are handling the core challenge
of the digital transformation – the presentation of digitized content and retro-converted
digitized catalog records – as a day-to-day routine, even at large scale. Media are digitized at
a daily basis, extended with structural information, indexed, and treated with OCR engines,
to be presented in web-based digitized collections1 or to be distributed via typical metadata
interchange interfaces such as OAI-PMH2 .
However, new tasks for research libraries are emerging, e.g., the digital curation of the
owned collections and the provision of data for various research tasks from the wide field
of digital humanities (DH). As these use cases fall outside the traditional bibliographic use
case, i.e., the indexing and retrieval of different media, traditional bibliographic records do
not not satisfy the requirements of both researchers in DH as well as digital curators. On
the one hand, these records are often missing vital information such as named entities or
other information that can be used to enable explorative information seeking strategies. On
the other hand, these records contain very detailed information that is necessary for the
bibliographic use case while being over-complex and cryptic for DH researchers. Furthermore,
proprietary character encodings or system-specific annotations are putting an additional
burden on the usage of the data outside the scope of common library tasks. Listing 1
illustrates this phenomenon very well with the help of the library management system’s
internal Pica+ format.
Because of the sheer amount of data available in large libraries, a manual conversion or
augmentation of these records to fit the aforementioned needs would be very cost-intensive
and hardly possible if it had to be carried out by library staff. Thus, a machine-assisted
approach to transform traditional metadata records into datasets usable by digital curators
or DH researchers is needed to cope with this problem. A recent proof of concept [2] shows
1
https://digital.staatsbibliothek-berlin.de/
2
https://www.openarchives.org/OAI/openarchivesprotocol.html
© David Zellhöfer;
licensed under Creative Commons License CC-BY
LDK 2019 - Posters Track.
Editors: Thierry Declerck and John P. McCrae
XX:2 Multimodal Datasets of the Berlin State Library
Listing 1 Excerpt of a bibliographic record in Pica+ format
011 @ a1812
011 B a2004 - b2007
019 @ aXD - US
021 A aAn @oration pronounced at Dedham on the anniversary of
American independence , July 4 , 1812 hby Jabez Chickering
028 A dJabez aChickering h1753 -1812
033 A pBoston nPrinted by Joshua Belcher
101 @ a11
201 B /01 014 -03 -17 t23 :01:04.000
the feasibility of such an approach relying on methods from machine-based learning, data
analysis, and traditional data management and batch processing.
The following section presents the core characteristics of four multimodal and multilingual
datasets based on publicly available catalog and other metadata of the Berlin State Library
that have been transformed by tools presented in [2]. For the sake of reproductiveness and
transparency, all scripts are made available3 with a permissive license.
2 Characteristics of the Datasets
All of the presented datasets are inter-linkable with the help of the so-called PPN (Pica
production number). In most cases, the PPN can be seen as a unique identifier for analog or
digitized media that is used in many systems of the Berlin State Library and the libraries
of the GBV alliance4 , e.g., the central catalog. PPN can also be used to download image
content via the IIIF5 endpoint or metadata and OCR content via the OAI-PMH interface.
For some sample scenarios, refer to [7].
2.1 Extract from the Library’s Main Catalog
This dataset [3] is derived from the Pica+ serialization of the full library’s main catalog
from 2018 containing 9,850,467 records of analog, digitized, and digital-born material. The
following fields have been extracted: title, author (incl. optional GND6 ID), publisher,
place of publication, country of publication, and year of publication. To facilitate further
processing, the publications are split by language groups (ranging from ancient to modern
languages).
The records are stored in a simple tabulator-separated field-based text format. Records
are isolated by empty lines, whereas @ serves as a subfield indicator in case a GND ID or
detailed location information is available. Table 1 presents a sample records, whose complete
data can be referenced with the help of the given PPN7 . Details on the different Pica+ field
IDs and their contents are available under [3] accompanied by the creation script. A full list
of available fields (in German) is also available8 .
3
https://github.com/elektrobohemian/StabiHacks
4
https://www.gbv.de/?set_language=en
5
https://iiif.io/
6
https://www.dnb.de/EN/Standardisierung/GND/gnd_node.html
7
http://stabikat.de/DB=1/SET=1/TTL=1/PRS=PP%7F/PPN?PPN=0249445468
8
https://www.gbv.de/bibliotheken/verbundbibliotheken/02Verbund/01Erschliessung/02Richtlinien/
01KatRicht/inhalt.shtml
David Zellhöfer XX:3
Table 1 Sample record of PPN 0249445468
PPN Pica+ field ID Content
0249445468 011@ 1939
0249445468 019@ XD-US
0249445468 021A The plays of William Shakespeare in thirty-seven volumes
0249445468 028A Shakespeare, William@gnd/118613723
0249445468 033A The Limited Editions Club@New York, NY
2.2 Metadata, Title Pages, and Network Graph of the Digitized
Content of the Berlin State Library
The dataset has been downloaded via the OAI-PMH Dublin Core endpoint of the Berlin
State Library’s Digitized Collections9 and has been converted into common tabular formats
and graph representations in GML. It contains 146,000 records of digitized material older
than 1920 in the format described in Table 2.
In addition to the bibliographic metadata, representative images of the works have been
downloaded and resized to a 512 pixel maximum thumbnail JPEG image preserving the
original aspect ratio. Title pages have been derived from structural metadata created by
scan operators and librarians. If this information was not available, first pages of the media
have been downloaded. In case of multi-volume media, title pages are not available. As
a consequence, only 141,206 images title/first pages are present. Additionally, geo-spatial
coordinates have been added to each record using the OpenStreetMap web service10 . For
details, refer to [5].
2.3 Title, Author, Publisher, Place of Publication, and
Language-related Network Graphs of the Library Main Catalog
Three graphs (in GraphML, GML, and JSON) are made available in this dataset linking:
authors, publishers, and places of publication (author_publisher_location);
authors, publishers, places of publication, and titles (author_publisher_location_title);
authors, publishers, and the language of publication (languageLink).
The languages of publication graphs spans all of the languages mentioned above and
has 1,555,119 nodes and 1,659,596 edges (see Fig. 1 for an exemplary subgraph). Table 3
subsumes the core properties of each provided graph. For additional details, see [6].
2.4 Extracted Illustrations of the Berlin State Library’s Digitized
Collections
The largest dataset consists of ca. 22,142 digitized media units11 that have been OCR-
processed with the ABBYY FineReader Engine (at least version 11) and whose full-texts are
made available in ALTO XML referenced in the METS/MODS XML file of each object12 .
Based on the results of the OCR, all found illustrations have been extracted and saved in
9
https://digital.staatsbibliothek-berlin.de/oai
10
https://www.openstreetmap.org/
11
This number is subject to change as the OCR and image extraction process is ongoing.
12
The data is available over the OAI-PMH METS/MODS endpoint of the digitized collections.
L D K Po s t e r s
XX:4 Multimodal Datasets of the Berlin State Library
Table 2 Description of the tabular format of the extended metadata
Column Name Description
title The title of the medium
creator Its creator (family name, first name)
subject A collection’s name as provided by the library
type The type of medium
format A MIME type for full metadata download
identifier An additional identifier (most often the PPN)
language A 3-letter language code of the medium
date The date of creation/publication or a time span
relation A relation to a project or collection a medium has been digitized for.
coverage The location of publication or origin (ranging from cities to continents)
publisher The publisher of the medium.
rights Copyright information.
PPN The unique identifier that can be used to find more information about the
current medium in all information systems of Berlin State Library.
The following fields contain data that is based on different processing steps.
spatialClean In case of multiple entries in coverage, only the first place of origin has been
extracted. Additionally, characters such as question marks, brackets, or the like
have been removed. The entries have been normalized regarding whitespaces
and writing variants with the help of regular expressions.
dateClean As the original date may contain various format variants to indicate unclear
creation dates (e.g., time spans or question marks), this field contains a mapping
to a certain point in time.
spatialCluster The cluster ID determined with the help of the Jaro-Winkler distance on the
spatialClean string. This step is needed because the spatialClean fields still
contain a huge amount of orthographic variants and latinizations of geographic
names.
spatialClusterName A verbal cluster name (controlled manually).
latitude The latitude provided by OpenStreetMap of the spatialClusterName if the
location could be found.
longitude The longitude provided by OpenStreetMap of the spatialClusterName if the
location could be found.
century A century derived from the date.
textCluster A text cluster ID on the basis of a k-means clustering relying on the title field
with a vocabulary size of 125,000 using the tf*idf model and k=5,000.
creatorCluster A text cluster ID based on the creator field with k=20,000.
titleImage The path to the first/title page relative to the img/ subdirectory or None in
case of a multi-volume work.
David Zellhöfer XX:5
Table 3 Graph properties per language
Language Graph Type Nodes Edges Records
fry author_publisher_location 298 264 360
fry author_publisher_location_title 622 726 360
ice author_publisher_location 505 448 1,200
ice author_publisher_location_title 1,509 1,393 1,200
por author_publisher_location 5,217 5,392 8,937
por author_publisher_location_title 12,848 15,219 8,937
nor author_publisher_location 4,948 6,049 12,016
nor author_publisher_location_title 15,276 21,737 12,016
dan author_publisher_location 9,127 11,711 20,089
dan author_publisher_location_title 26,144 39,278 20,089
swe author_publisher_location 15,350 18,367 30,628
swe author_publisher_location_title 41,933 61,000 30,628
spa author_publisher_location 24,404 27,477 42,540
spa author_publisher_location_title 59,339 77,779 42,540
dut author_publisher_location 36,503 42,128 67,000
dut author_publisher_location_title 94,803 127,785 67,000
ita author_publisher_location 71,151 95,054 158,851
ita author_publisher_location_title 206,656 316,282 158,851
lat author_publisher_location 91,584 148,224 230,588
lat author_publisher_location_title 273,724 469,322 230,588
fre author_publisher_location 174,650 204,299 380,569
fre author_publisher_location_title 487,053 693,245 380,569
eng author_publisher_location 606,112 880,989 1,309,172
eng author_publisher_location_title 1,778,957 2,710,807 1,309,172
ger author_publisher_location 705,468 1,104,502 2,316,600
ger author_publisher_location_title 2,497,239 3,947,482 2,316,600
n/a languageLink 1,555,119 1,659,596 4,578,537
L D K Po s t e r s
XX:6 Multimodal Datasets of the Berlin State Library
Figure 1 Network of Authors, Publishers, and Languages of Publication (Subgraph of the
Languages: fry, ice, por, nor, dan, swe)
original size in JPEG format. In total, 531,484 illustrations have been extracted from 22,142
media units, i.e., an average of 24 extracted illustrations per unit [4].
In order to remove false positives from the corpus, e.g., stamps, hand-written signatures,
or empty pages, pre-trained classifiers are provided in form of different Python scripts[1]
based on a pre-trained VGGnet models implemented with Keras/TensorFlow.
References
1 Julia Berauer, Ralitsa Doncheva, Linh Nguyen, Luisa Rademacher, Carlos Tan, and Caglar
Özel. Chasing Unicorns and Vampires in a Library. Technical report, HTW Berlin, 2018.
URL: https://github.com/elektrobohemian/imi-unicorns.
2 David Zellhöfer. Exploring Large Digital Libraries by Multimodal Criteria. In Norbert
Fuhr, László Kovács, Thomas Risse, and Wolfgang Nejdl, editors, Research and Advanced
Technology for Digital Libraries - 20th International Conference on Theory and Practice of
Digital Libraries, TPDL 2016, Hannover, Germany, September 5-9, 2016, Proceedings, volume
9819 of Lecture Notes in Computer Science, pages 307–319. Springer, 2016.
3 David Zellhöfer. Extract from the Library’s Main Catalog, March 2019. URL: https:
//doi.org/10.5281/zenodo.2590752.
4 David Zellhöfer. Extracted Illustrations of the Berlin State Library’s Digitized Collections,
March 2019. URL: http://doi.org/10.5281/zenodo.2602431.
5 David Zellhöfer. Metadata, Title Pages, and Network Graph of the Digitized Content of the
Berlin State Library (146,000 items), March 2019. URL: https://doi.org/10.5281/zenodo.
2582482.
6 David Zellhöfer. Title, Author, Publisher, Place of Publication, and Language-related Network
Graphs of the Library Main Catalog, March 2019. URL: https://doi.org/10.5281/zenodo.
2587801.
7 David Zellhöfer. What is a PPN and Why is it Helpful?, May 2019. URL: http://doi.org/
10.5281/zenodo.2702544.