Dating texts by multi-class classification with sliding time intervals

Gregory Toner; Xiwu Han

doi:10.1109/CISP-BMEI.2017.8302102

Dating texts by multi-class classification with sliding time intervals

Gregory Toner
, Xiwu Han

School of Arts, English and Languages

Research output: Contribution to conference › Paper › peer-review

3 Citations (Scopus)

Abstract

We propose a practical method to date texts by classification with sliding time intervals (STI). This further explores the advantage of multi-class text classification, while drawing upon temporal characteristics in the training corpus. Extensive experiments were made on English and medieval Irish texts. Results showed that our STI dating method significantly outperformed classifiers with fixed time intervals (FTI). The Naïve Bayes Multinomial (NBM) with STI achieved the state-of-the-art dating precision on DTE Subtask 2 though only involving features of n-gram characters and words. Experiments on dating long documents and further analysis also indicated some promising points for further text dating research and other humanities fields.

Original language	English
Pages	1
Number of pages	6
DOIs	https://doi.org/10.1109/CISP-BMEI.2017.8302102
Publication status	Published - 27 Feb 2018
Event	International Congress on Image and Signal Processing, BioMedical Engineering and Informatics - Shanghai, China Duration: 14 Oct 2017 → 16 Oct 2017

Conference

Conference	International Congress on Image and Signal Processing, BioMedical Engineering and Informatics
Abbreviated title	CISP-BMEI 2017
Country/Territory	China
City	Shanghai
Period	14/10/2017 → 16/10/2017

Keywords

Bayes methods
Naïve Bayes Multinomial
sliding time intervals
medieval Irish
text dating
machine learning
annals

Access to Document

10.1109/CISP-BMEI.2017.8302102

Cite this

@conference{02d4fbbb5ea04dc388c457b70235e3e5,

title = "Dating texts by multi-class classification with sliding time intervals",

abstract = "We propose a practical method to date texts by classification with sliding time intervals (STI). This further explores the advantage of multi-class text classification, while drawing upon temporal characteristics in the training corpus. Extensive experiments were made on English and medieval Irish texts. Results showed that our STI dating method significantly outperformed classifiers with fixed time intervals (FTI). The Na{\"i}ve Bayes Multinomial (NBM) with STI achieved the state-of-the-art dating precision on DTE Subtask 2 though only involving features of n-gram characters and words. Experiments on dating long documents and further analysis also indicated some promising points for further text dating research and other humanities fields.",

keywords = "Bayes methods, Na{\"i}ve Bayes Multinomial, sliding time intervals, medieval Irish, text dating, machine learning, annals",

author = "Gregory Toner and Xiwu Han",

year = "2018",

month = feb,

day = "27",

doi = "10.1109/CISP-BMEI.2017.8302102",

language = "English",

pages = "1",

note = "International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017 ; Conference date: 14-10-2017 Through 16-10-2017",

}

TY - CONF

T1 - Dating texts by multi-class classification with sliding time intervals

AU - Toner, Gregory

AU - Han, Xiwu

PY - 2018/2/27

Y1 - 2018/2/27

N2 - We propose a practical method to date texts by classification with sliding time intervals (STI). This further explores the advantage of multi-class text classification, while drawing upon temporal characteristics in the training corpus. Extensive experiments were made on English and medieval Irish texts. Results showed that our STI dating method significantly outperformed classifiers with fixed time intervals (FTI). The Naïve Bayes Multinomial (NBM) with STI achieved the state-of-the-art dating precision on DTE Subtask 2 though only involving features of n-gram characters and words. Experiments on dating long documents and further analysis also indicated some promising points for further text dating research and other humanities fields.

AB - We propose a practical method to date texts by classification with sliding time intervals (STI). This further explores the advantage of multi-class text classification, while drawing upon temporal characteristics in the training corpus. Extensive experiments were made on English and medieval Irish texts. Results showed that our STI dating method significantly outperformed classifiers with fixed time intervals (FTI). The Naïve Bayes Multinomial (NBM) with STI achieved the state-of-the-art dating precision on DTE Subtask 2 though only involving features of n-gram characters and words. Experiments on dating long documents and further analysis also indicated some promising points for further text dating research and other humanities fields.

KW - Bayes methods

KW - Naïve Bayes Multinomial

KW - sliding time intervals

KW - medieval Irish

KW - text dating

KW - machine learning

KW - annals

UR - http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8302102&isnumber=8301897

U2 - 10.1109/CISP-BMEI.2017.8302102

DO - 10.1109/CISP-BMEI.2017.8302102

M3 - Paper

SP - 1

T2 - International Congress on Image and Signal Processing, BioMedical Engineering and Informatics

Y2 - 14 October 2017 through 16 October 2017

ER -

Dating texts by multi-class classification with sliding time intervals

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this