Authorship Attribution: Comparison of Single-layer and
Double-layer Machine Learning

D 2012

Authorship Attribution: Comparison of Single-layer and Double-layer Machine Learning

RYGL, Jan a Aleš HORÁK

Základní údaje

Originální název

Authorship Attribution: Comparison of Single-layer and Double-layer Machine Learning

Autoři

RYGL, Jan a Aleš HORÁK

Vydání

7499. vyd. Brno, Text, Speech and Dialogue - 15th International Conference, od s. 282-289, 8 s. 2012

Nakladatel

Springer

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

60200 6.2 Languages and Literature

Stát vydavatele

Česká republika

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

tištěná verze "print"

Odkazy

URL

Impakt faktor

Impact factor: 0.402 v roce 2005

Označené pro přenos do RIV

Ano

Kód RIV

RIV/00216224:14330/12:00060281

Organizační jednotka

Fakulta informatiky

ISBN

978-3-642-32789-6

ISSN

DOI

https://doi.org/10.1007/978-3-642-32790-2_34

UT WoS

000337298700034

Klíčová slova česky

dvojvrstvé strojové učení; určování autorství; podobnost dokumentů

Klíčová slova anglicky

double layered machine learning; authorship attribution; similarity of documents

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 4. 7. 2014 16:33, RNDr. Jan Rygl

Anotace

V originále

In the traditional authorship attribution task, forensic linguistic specialists analyse and compare documents to determine who was their (real) author. In the current days, the number of anonymous docu- ments is growing ceaselessly because of Internet expansion. That is why the manual part of the authorship attribution process needs to be replaced with automatic methods. Specialized algorithms (SA) like delta-score and word length statistic were developed to quantify the similarity between documents, but currently prevailing techniques build upon the machine learning (ML) approach. In this paper, two machine learning approaches are compared: Single-layer ML, where the results of SA (similarities of documents) are used as input attributes for the machine learning, and Double-layer ML with the numerical information characterizing the author being extracted from documents and divided into several groups. For each group the machine learning classifier is trained and the outputs of these classifiers are used as input attributes for ML in the second step. Generating attributes for the machine learning in the first step of double- layer ML, which is based on SA, is described in detail here. Documents from Czech blog servers are utilized for empirical evaluation of both approaches.

Návaznosti

VF20102014003, projekt VaV

Název: Analýza přirozeného jazyka v prostředí internetu (Akronym: APJI)

Investor: Ministerstvo vnitra ČR, Analýza přirozeného jazyka v prostředí internetu

Citovat

RYGL, Jan a Aleš HORÁK. Authorship Attribution: Comparison of Single-layer and Double-layer Machine Learning. In Petr Sojka, Ales Horák, Ivan Kopecek, Karel Pala. Text, Speech and Dialogue - 15th International Conference. 7499. vyd. Brno: Springer, 2012, s. 282-289. ISBN 978-3-642-32789-6. Dostupné z: https://doi.org/10.1007/978-3-642-32790-2_34.

@inproceedings{985959,
   author = {Rygl, Jan and Horák, Aleš},
   address = {Brno},
   booktitle = {Text, Speech and Dialogue - 15th International Conference},
   doi = {https://doi.org/10.1007/978-3-642-32790-2_34},
   edition = {7499},
   editor = {Petr Sojka, Ales Horák, Ivan Kopecek, Karel Pala},
   keywords = {double layered machine learning; authorship attribution; similarity of documents},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Brno},
   isbn = {978-3-642-32789-6},
   pages = {282-289},
   publisher = {Springer},
   title = {Authorship Attribution: Comparison of Single-layer and Double-layer Machine Learning},
   url = {http://www.springerlink.com/content/v9021r456183/#section=1107995&page=1},
   year = {2012}
}

TY  - CONF
ID  - 985959
AU  - Rygl, Jan - Horák, Aleš
PY  - 2012
TI  - Authorship Attribution: Comparison of Single-layer and Double-layer Machine Learning
PB  - Springer
CY  - Brno
SN  - 9783642327896
KW  - double layered machine learning
KW  - authorship attribution
KW  - similarity of documents
UR  - http://www.springerlink.com/content/v9021r456183/#section=1107995&page=1
N2  - In the traditional authorship attribution task, forensic linguistic specialists analyse and compare documents to determine who was their (real) author. In the current days, the number of anonymous docu- ments is growing ceaselessly because of Internet expansion. That is why the manual part of the authorship attribution process needs to be replaced with automatic methods. Specialized algorithms (SA) like delta-score and word length statistic were developed to quantify the similarity between documents, but currently prevailing techniques build upon the machine learning (ML) approach. In this paper, two machine learning approaches are compared: Single-layer ML, where the results of SA (similarities of documents) are used as input attributes for the machine learning, and Double-layer ML with the numerical information characterizing the author being extracted from documents and divided into several groups. For each group the machine learning classifier is trained and the outputs of these classifiers are used as input attributes for ML in the second step. Generating attributes for the machine learning in the first step of double- layer ML, which is based on SA, is described in detail here. Documents from Czech blog servers are utilized for empirical evaluation of both approaches.
ER  -

RYGL, Jan a Aleš HORÁK. Authorship Attribution: Comparison of Single-layer and Double-layer Machine Learning. In Petr Sojka, Ales Horák, Ivan Kopecek, Karel Pala. \textit{Text, Speech and Dialogue - 15th International Conference}. 7499. vyd. Brno: Springer, 2012, s.~282-289. ISBN~978-3-642-32789-6. Dostupné z: https://doi.org/10.1007/978-3-642-32790-2\_{}34.

Přehled o publikaci