Disentangling the Structure of Tables in Scientific Literature

Nikola Milosevic; Goran Nenadic; Cassie Gregson; Robert Hernandez

doi:10.1007/978-3-319-41754-7_14

Disentangling the Structure of Tables in Scientific Literature

Nikola Milosevic, Goran Nenadic, Cassie Gregson, Robert Hernandez

AstraZeneca plc

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

781 Downloads (Pure)

Abstract

Within the scientific literature, tables are commonly used to
present factual and statistical information in a compact way, which is easy
to digest by readers. The ability to “understand” the structure of tables is
key for information extraction in many domains. However, the complexity
and variety of presentation layouts and value formats makes it difficult to
automatically extract roles and relationships of table cells. In this paper,
we present a model that structures tables in a machine readable way and
a methodology to automatically disentangle and transform tables into the
modelled data structure. The method was tested in the domain of clinical
trials: it achieved an F-score of 94.26 % for cell function identification and
94.84 % for identification of inter-cell relationships.

Original language	English
Title of host publication	Natural Language Processing and Information Systems
Subtitle of host publication	21st International Conference on Applications of Natural Language to Information Systems, NLDB 2016, Salford, UK, June 22-24, 2016, Proceedings
Place of Publication	Switzerland
Publisher	Springer Nature
Pages	162-174
Number of pages	13
Volume	9612
ISBN (Electronic)	978-3-319-41754-7
ISBN (Print)	978-3-319-41753-0
DOIs	https://doi.org/10.1007/978-3-319-41754-7_14
Publication status	Published - 17 Jun 2016
Event	21st International Conference on Applications of Natural Language to Information Systems - Media City, Salford, United Kingdom Duration: 22 Jun 2016 → 24 Jun 2016 Conference number: 21 http://www.salford.ac.uk/conferencing-at-salford/conference-management/current-conference/nldb-conference

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	9612

Conference

Conference	21st International Conference on Applications of Natural Language to Information Systems
Abbreviated title	NLDB 2016,
Country/Territory	United Kingdom
City	Salford
Period	22/06/16 → 24/06/16
Internet address	http://www.salford.ac.uk/conferencing-at-salford/conference-management/current-conference/nldb-conference

Keywords

Table mining
Text mining
Data management
Data modelling
Natural language processing

Access to Document

10.1007/978-3-319-41754-7_14

Disentangling the Structure of Tables in Scientific LiteratureAccepted author manuscript, 500 KB

Cite this

Milosevic, N., Nenadic, G., Gregson, C., & Hernandez, R. (2016). Disentangling the Structure of Tables in Scientific Literature. In Natural Language Processing and Information Systems: 21st International Conference on Applications of Natural Language to Information Systems, NLDB 2016, Salford, UK, June 22-24, 2016, Proceedings (Vol. 9612, pp. 162-174). (Lecture Notes in Computer Science; Vol. 9612). Springer Nature. https://doi.org/10.1007/978-3-319-41754-7_14

Milosevic, Nikola ; Nenadic, Goran ; Gregson, Cassie et al. / Disentangling the Structure of Tables in Scientific Literature. Natural Language Processing and Information Systems: 21st International Conference on Applications of Natural Language to Information Systems, NLDB 2016, Salford, UK, June 22-24, 2016, Proceedings. Vol. 9612 Switzerland : Springer Nature, 2016. pp. 162-174 (Lecture Notes in Computer Science).

@inproceedings{473111c252e9493abe8c1a78c5b7ce36,

title = "Disentangling the Structure of Tables in Scientific Literature",

abstract = "Within the scientific literature, tables are commonly used topresent factual and statistical information in a compact way, which is easyto digest by readers. The ability to “understand” the structure of tables iskey for information extraction in many domains. However, the complexityand variety of presentation layouts and value formats makes it difficult toautomatically extract roles and relationships of table cells. In this paper,we present a model that structures tables in a machine readable way anda methodology to automatically disentangle and transform tables into themodelled data structure. The method was tested in the domain of clinicaltrials: it achieved an F-score of 94.26 % for cell function identification and94.84 % for identification of inter-cell relationships.",

keywords = "Table mining, Text mining, Data management, Data modelling, Natural language processing",

author = "Nikola Milosevic and Goran Nenadic and Cassie Gregson and Robert Hernandez",

year = "2016",

month = jun,

day = "17",

doi = "10.1007/978-3-319-41754-7_14",

language = "English",

isbn = "978-3-319-41753-0",

volume = "9612",

series = "Lecture Notes in Computer Science",

publisher = "Springer Nature",

pages = "162--174",

booktitle = "Natural Language Processing and Information Systems",

address = "United States",

note = "21st International Conference on Applications of Natural Language to Information Systems , NLDB 2016, ; Conference date: 22-06-2016 Through 24-06-2016",

url = "http://www.salford.ac.uk/conferencing-at-salford/conference-management/current-conference/nldb-conference",

}

Milosevic, N , Nenadic, G, Gregson, C & Hernandez, R 2016, Disentangling the Structure of Tables in Scientific Literature. in Natural Language Processing and Information Systems: 21st International Conference on Applications of Natural Language to Information Systems, NLDB 2016, Salford, UK, June 22-24, 2016, Proceedings. vol. 9612, Lecture Notes in Computer Science, vol. 9612, Springer Nature, Switzerland, pp. 162-174, 21st International Conference on Applications of Natural Language to Information Systems , Salford, United Kingdom, 22/06/16. https://doi.org/10.1007/978-3-319-41754-7_14

Disentangling the Structure of Tables in Scientific Literature. / Milosevic, Nikola ; Nenadic, Goran; Gregson, Cassie et al.
Natural Language Processing and Information Systems: 21st International Conference on Applications of Natural Language to Information Systems, NLDB 2016, Salford, UK, June 22-24, 2016, Proceedings. Vol. 9612 Switzerland: Springer Nature, 2016. p. 162-174 (Lecture Notes in Computer Science; Vol. 9612).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Disentangling the Structure of Tables in Scientific Literature

AU - Milosevic, Nikola

AU - Nenadic, Goran

AU - Gregson, Cassie

AU - Hernandez, Robert

N1 - Conference code: 21

PY - 2016/6/17

Y1 - 2016/6/17

N2 - Within the scientific literature, tables are commonly used topresent factual and statistical information in a compact way, which is easyto digest by readers. The ability to “understand” the structure of tables iskey for information extraction in many domains. However, the complexityand variety of presentation layouts and value formats makes it difficult toautomatically extract roles and relationships of table cells. In this paper,we present a model that structures tables in a machine readable way anda methodology to automatically disentangle and transform tables into themodelled data structure. The method was tested in the domain of clinicaltrials: it achieved an F-score of 94.26 % for cell function identification and94.84 % for identification of inter-cell relationships.

AB - Within the scientific literature, tables are commonly used topresent factual and statistical information in a compact way, which is easyto digest by readers. The ability to “understand” the structure of tables iskey for information extraction in many domains. However, the complexityand variety of presentation layouts and value formats makes it difficult toautomatically extract roles and relationships of table cells. In this paper,we present a model that structures tables in a machine readable way anda methodology to automatically disentangle and transform tables into themodelled data structure. The method was tested in the domain of clinicaltrials: it achieved an F-score of 94.26 % for cell function identification and94.84 % for identification of inter-cell relationships.

KW - Table mining

KW - Text mining

KW - Data management

KW - Data modelling

KW - Natural language processing

U2 - 10.1007/978-3-319-41754-7_14

DO - 10.1007/978-3-319-41754-7_14

M3 - Conference contribution

SN - 978-3-319-41753-0

VL - 9612

T3 - Lecture Notes in Computer Science

SP - 162

EP - 174

BT - Natural Language Processing and Information Systems

PB - Springer Nature

CY - Switzerland

T2 - 21st International Conference on Applications of Natural Language to Information Systems

Y2 - 22 June 2016 through 24 June 2016

ER -

Milosevic N , Nenadic G, Gregson C, Hernandez R. Disentangling the Structure of Tables in Scientific Literature. In Natural Language Processing and Information Systems: 21st International Conference on Applications of Natural Language to Information Systems, NLDB 2016, Salford, UK, June 22-24, 2016, Proceedings. Vol. 9612. Switzerland: Springer Nature. 2016. p. 162-174. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-41754-7_14

Disentangling the Structure of Tables in Scientific Literature

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this