Occupational representativeness in Twitter

Sunghwan Mac Kim; Stephen Wan; Cécile Paris

doi:10.1145/3015022.3015036

Occupational representativeness in Twitter

Sunghwan Mac Kim, Stephen Wan, Cécile Paris

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

3 Citations (Scopus)

Abstract

This paper describes an approach to detect one particular demographic characteristic, occupation (or profession) in Twitter user profiles. In this paper, we show how effective the approach is for estimating occupational population statistics in Australian Twitter by correlating them with real-world population obtained from 2011 Australian census data. We also demonstrate that we can gain more reliable social media insights in the context of occupational representativeness in Twitter if a non-standard occupation name is mapped into a standard occupation name. To our knowledge, this is the first attempt to build a machine learning model that automatically identifies linguistically noisy or open-ended occupations in Twitter, resulting in more reliable occupational population.

Original language	English
Title of host publication	ADCS 2016
Subtitle of host publication	Proceedings of the 21st Australasian Document Computing Symposium
Editors	Sarvnaz Karimi, Mark Carman
Publisher	Association for Computing Machinery
Pages	57-64
Number of pages	8
ISBN (Electronic)	9781450348652
DOIs	https://doi.org/10.1145/3015022.3015036
Publication status	Published - 5 Dec 2016
Externally published	Yes
Event	21st Australasian Document Computing Symposium, ADCS 2016 - Caulfield, Australia Duration: 6 Dec 2016 → 7 Dec 2016

Other

Other	21st Australasian Document Computing Symposium, ADCS 2016
Country/Territory	Australia
City	Caulfield
Period	6/12/16 → 7/12/16

Keywords

Twitter
occupation
census
Conditional Random Fields

Access to Document

10.1145/3015022.3015036

Cite this

@inproceedings{c228340252404deeb1a1aadb71111470,

title = "Occupational representativeness in Twitter",

abstract = "This paper describes an approach to detect one particular demographic characteristic, occupation (or profession) in Twitter user profiles. In this paper, we show how effective the approach is for estimating occupational population statistics in Australian Twitter by correlating them with real-world population obtained from 2011 Australian census data. We also demonstrate that we can gain more reliable social media insights in the context of occupational representativeness in Twitter if a non-standard occupation name is mapped into a standard occupation name. To our knowledge, this is the first attempt to build a machine learning model that automatically identifies linguistically noisy or open-ended occupations in Twitter, resulting in more reliable occupational population.",

keywords = "Twitter, occupation, census, Conditional Random Fields",

author = "Kim, {Sunghwan Mac} and Stephen Wan and C{\'e}cile Paris",

year = "2016",

month = dec,

day = "5",

doi = "10.1145/3015022.3015036",

language = "English",

pages = "57--64",

editor = "Sarvnaz Karimi and Mark Carman",

booktitle = "ADCS 2016",

publisher = "Association for Computing Machinery",

note = "21st Australasian Document Computing Symposium, ADCS 2016 ; Conference date: 06-12-2016 Through 07-12-2016",

}

Occupational representativeness in Twitter. / Kim, Sunghwan Mac; Wan, Stephen; Paris, Cécile.

ADCS 2016: Proceedings of the 21st Australasian Document Computing Symposium. ed. / Sarvnaz Karimi; Mark Carman. Association for Computing Machinery, 2016. p. 57-64.

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

TY - GEN

T1 - Occupational representativeness in Twitter

AU - Kim, Sunghwan Mac

AU - Wan, Stephen

AU - Paris, Cécile

PY - 2016/12/5

Y1 - 2016/12/5

N2 - This paper describes an approach to detect one particular demographic characteristic, occupation (or profession) in Twitter user profiles. In this paper, we show how effective the approach is for estimating occupational population statistics in Australian Twitter by correlating them with real-world population obtained from 2011 Australian census data. We also demonstrate that we can gain more reliable social media insights in the context of occupational representativeness in Twitter if a non-standard occupation name is mapped into a standard occupation name. To our knowledge, this is the first attempt to build a machine learning model that automatically identifies linguistically noisy or open-ended occupations in Twitter, resulting in more reliable occupational population.

AB - This paper describes an approach to detect one particular demographic characteristic, occupation (or profession) in Twitter user profiles. In this paper, we show how effective the approach is for estimating occupational population statistics in Australian Twitter by correlating them with real-world population obtained from 2011 Australian census data. We also demonstrate that we can gain more reliable social media insights in the context of occupational representativeness in Twitter if a non-standard occupation name is mapped into a standard occupation name. To our knowledge, this is the first attempt to build a machine learning model that automatically identifies linguistically noisy or open-ended occupations in Twitter, resulting in more reliable occupational population.

KW - Twitter

KW - occupation

KW - census

KW - Conditional Random Fields

UR - http://www.scopus.com/inward/record.url?scp=85014723180&partnerID=8YFLogxK

U2 - 10.1145/3015022.3015036

DO - 10.1145/3015022.3015036

M3 - Conference proceeding contribution

AN - SCOPUS:85014723180

SP - 57

EP - 64

BT - ADCS 2016

A2 - Karimi, Sarvnaz

A2 - Carman, Mark

PB - Association for Computing Machinery

T2 - 21st Australasian Document Computing Symposium, ADCS 2016

Y2 - 6 December 2016 through 7 December 2016

ER -

Occupational representativeness in Twitter

Abstract

Other

Keywords

Access to Document

Other files and links

Fingerprint

Cite this