The Distribution of English Isograms in Google Ngrams and the British National Corpus

Research output: Contribution to journalArticlepeer-review

Standard Standard

The Distribution of English Isograms in Google Ngrams and the British National Corpus. / Breit, Florian.
In: Opticon1826, 2017.

Research output: Contribution to journalArticlepeer-review

HarvardHarvard

APA

CBE

MLA

VancouverVancouver

Author

RIS

TY - JOUR

T1 - The Distribution of English Isograms in Google Ngrams and the British National Corpus

AU - Breit, Florian

N1 - Opticon1826 ceased publishing in 2016 keep the entry as unpublished

PY - 2017

Y1 - 2017

N2 - The study of isograms—words in which each letter occurs the same number of times—has thus far largely been limited to manual search for examples in sources such as dictionaries, and accounts have principally limited themselves to simply listing the known isograms of various categories. This paper presents the results of a corpus study of English isograms from Google Ngrams (ca. 1 trillion words, ~13 million types) and the British National Corpus (ca. 100 million words, ~6 million types). The paper discusses methodological issues relating to the automated mining of isograms, explores the distribution of isograms in relation to word-length and frequency, and presents several new isograms, which have so far gone unnoticed in the literature. Moreover the paper describes the resultant dataset of English isograms and the tools used to create it, which are made freely available and can be used to further study the distribution of isogramy in English and other languages.

AB - The study of isograms—words in which each letter occurs the same number of times—has thus far largely been limited to manual search for examples in sources such as dictionaries, and accounts have principally limited themselves to simply listing the known isograms of various categories. This paper presents the results of a corpus study of English isograms from Google Ngrams (ca. 1 trillion words, ~13 million types) and the British National Corpus (ca. 100 million words, ~6 million types). The paper discusses methodological issues relating to the automated mining of isograms, explores the distribution of isograms in relation to word-length and frequency, and presents several new isograms, which have so far gone unnoticed in the literature. Moreover the paper describes the resultant dataset of English isograms and the tools used to create it, which are made freely available and can be used to further study the distribution of isogramy in English and other languages.

KW - English

KW - glottometrics

KW - isograms

KW - logology

KW - orthography

M3 - Article

JO - Opticon1826

JF - Opticon1826

SN - 2049-8128

ER -