Big Social Data Analytics in Football

Predicting Spectators and TV Ratings from Facebook Data

Nicolai H. Egebjerg, Niklas Hedegaard, Gerda Kuum, Raghava Rao Mukkamala, Ravi Vatrapu

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningpeer review

503 Downloads (Pure)

Resumé

This paper explores the predictive power of bigsocial data in regards to football fans’ off-line and on-linebehaviours. We address the research question of to what extentcan big social data from Facebook predict the numberof spectators and TV ratings in the case of Danish NationalFootball Association (DBU). The predictive model was built fromFacebook, match attendance, and TV ratings data sets from 2014-2016. The best fit was a linear regression model with GLM coding.Ultimately, the model did best when predicting the number ofspectators based on the Facebook activity during a match as wellas the activity from the last two weeks leading up to the match.Furthermore, the data reveals that photos generates the mostactivity on the national team’s page and with videos runningat higher production costs there might be some unexploitedpotential for DBU to improve its social media marketing strategy.Although data limitations are present, this research concludesthat predictive models based on big social data can indeed offerimportant insights for companies to understand their customerbase and how to improve marketing strategies.Index Terms—Big data, Big social media data, Danish NationalTeam, DBU, Facebook data, Football fans, Spectators, TV rating
OriginalsprogEngelsk
TitelProceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017
RedaktørerGeorge Karypis, Jia Zhang
Udgivelses stedLos Alamitos, CA
ForlagIEEE
Publikationsdato2017
Sider81-88
Artikelnummer8029312
ISBN (Trykt)9781538619964
ISBN (Elektronisk)9781538619957, 9781538619971
DOI
StatusUdgivet - 2017
Begivenhed6th IEEE International Congress on Big Data. BigData Congress 2017 - Hilton Hawaiian Village Waikiki Beach Resort, Honolulu, USA
Varighed: 25 jun. 201730 jun. 2017
Konferencens nummer: 6
http://www.bigdatacongress.org/2017/

Konference

Konference6th IEEE International Congress on Big Data. BigData Congress 2017
Nummer6
LokationHilton Hawaiian Village Waikiki Beach Resort
LandUSA
ByHonolulu
Periode25/06/201730/06/2017
Internetadresse

Emneord

  • Big data
  • Big social media data
  • Danish National Team
  • DBU
  • Facebook data
  • Football fans
  • Spectators
  • TV ratings

Citer dette

Egebjerg, N. H., Hedegaard, N., Kuum, G., Mukkamala, R. R., & Vatrapu, R. (2017). Big Social Data Analytics in Football: Predicting Spectators and TV Ratings from Facebook Data. I G. Karypis, & J. Zhang (red.), Proceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017 (s. 81-88). [8029312] Los Alamitos, CA: IEEE. https://doi.org/10.1109/BigDataCongress.2017.20
Egebjerg, Nicolai H. ; Hedegaard, Niklas ; Kuum, Gerda ; Mukkamala, Raghava Rao ; Vatrapu, Ravi. / Big Social Data Analytics in Football : Predicting Spectators and TV Ratings from Facebook Data. Proceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017. red. / George Karypis ; Jia Zhang. Los Alamitos, CA : IEEE, 2017. s. 81-88
@inproceedings{e5ca6fd03eb24273af03006decd3a27a,
title = "Big Social Data Analytics in Football: Predicting Spectators and TV Ratings from Facebook Data",
abstract = "This paper explores the predictive power of bigsocial data in regards to football fans’ off-line and on-linebehaviours. We address the research question of to what extentcan big social data from Facebook predict the numberof spectators and TV ratings in the case of Danish NationalFootball Association (DBU). The predictive model was built fromFacebook, match attendance, and TV ratings data sets from 2014-2016. The best fit was a linear regression model with GLM coding.Ultimately, the model did best when predicting the number ofspectators based on the Facebook activity during a match as wellas the activity from the last two weeks leading up to the match.Furthermore, the data reveals that photos generates the mostactivity on the national team’s page and with videos runningat higher production costs there might be some unexploitedpotential for DBU to improve its social media marketing strategy.Although data limitations are present, this research concludesthat predictive models based on big social data can indeed offerimportant insights for companies to understand their customerbase and how to improve marketing strategies.Index Terms—Big data, Big social media data, Danish NationalTeam, DBU, Facebook data, Football fans, Spectators, TV rating",
keywords = "Big data, Big social media data, Danish National Team, DBU, Facebook data, Football fans, Spectators, TV ratings, Big data, Big social media data, Danish National Team, DBU, Facebook data, Football fans, Spectators, TV ratings",
author = "Egebjerg, {Nicolai H.} and Niklas Hedegaard and Gerda Kuum and Mukkamala, {Raghava Rao} and Ravi Vatrapu",
year = "2017",
doi = "10.1109/BigDataCongress.2017.20",
language = "English",
isbn = "9781538619964",
pages = "81--88",
editor = "Karypis, {George } and Zhang, {Jia }",
booktitle = "Proceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017",
publisher = "IEEE",
address = "United States",

}

Egebjerg, NH, Hedegaard, N, Kuum, G, Mukkamala, RR & Vatrapu, R 2017, Big Social Data Analytics in Football: Predicting Spectators and TV Ratings from Facebook Data. i G Karypis & J Zhang (red), Proceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017., 8029312, IEEE, Los Alamitos, CA, s. 81-88, 6th IEEE International Congress on Big Data. BigData Congress 2017, Honolulu, USA, 25/06/2017. https://doi.org/10.1109/BigDataCongress.2017.20

Big Social Data Analytics in Football : Predicting Spectators and TV Ratings from Facebook Data. / Egebjerg, Nicolai H. ; Hedegaard, Niklas ; Kuum, Gerda; Mukkamala, Raghava Rao; Vatrapu, Ravi.

Proceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017. red. / George Karypis; Jia Zhang. Los Alamitos, CA : IEEE, 2017. s. 81-88 8029312.

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningpeer review

TY - GEN

T1 - Big Social Data Analytics in Football

T2 - Predicting Spectators and TV Ratings from Facebook Data

AU - Egebjerg, Nicolai H.

AU - Hedegaard, Niklas

AU - Kuum, Gerda

AU - Mukkamala, Raghava Rao

AU - Vatrapu, Ravi

PY - 2017

Y1 - 2017

N2 - This paper explores the predictive power of bigsocial data in regards to football fans’ off-line and on-linebehaviours. We address the research question of to what extentcan big social data from Facebook predict the numberof spectators and TV ratings in the case of Danish NationalFootball Association (DBU). The predictive model was built fromFacebook, match attendance, and TV ratings data sets from 2014-2016. The best fit was a linear regression model with GLM coding.Ultimately, the model did best when predicting the number ofspectators based on the Facebook activity during a match as wellas the activity from the last two weeks leading up to the match.Furthermore, the data reveals that photos generates the mostactivity on the national team’s page and with videos runningat higher production costs there might be some unexploitedpotential for DBU to improve its social media marketing strategy.Although data limitations are present, this research concludesthat predictive models based on big social data can indeed offerimportant insights for companies to understand their customerbase and how to improve marketing strategies.Index Terms—Big data, Big social media data, Danish NationalTeam, DBU, Facebook data, Football fans, Spectators, TV rating

AB - This paper explores the predictive power of bigsocial data in regards to football fans’ off-line and on-linebehaviours. We address the research question of to what extentcan big social data from Facebook predict the numberof spectators and TV ratings in the case of Danish NationalFootball Association (DBU). The predictive model was built fromFacebook, match attendance, and TV ratings data sets from 2014-2016. The best fit was a linear regression model with GLM coding.Ultimately, the model did best when predicting the number ofspectators based on the Facebook activity during a match as wellas the activity from the last two weeks leading up to the match.Furthermore, the data reveals that photos generates the mostactivity on the national team’s page and with videos runningat higher production costs there might be some unexploitedpotential for DBU to improve its social media marketing strategy.Although data limitations are present, this research concludesthat predictive models based on big social data can indeed offerimportant insights for companies to understand their customerbase and how to improve marketing strategies.Index Terms—Big data, Big social media data, Danish NationalTeam, DBU, Facebook data, Football fans, Spectators, TV rating

KW - Big data

KW - Big social media data

KW - Danish National Team

KW - DBU

KW - Facebook data

KW - Football fans

KW - Spectators

KW - TV ratings

KW - Big data

KW - Big social media data

KW - Danish National Team

KW - DBU

KW - Facebook data

KW - Football fans

KW - Spectators

KW - TV ratings

U2 - 10.1109/BigDataCongress.2017.20

DO - 10.1109/BigDataCongress.2017.20

M3 - Article in proceedings

SN - 9781538619964

SP - 81

EP - 88

BT - Proceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017

A2 - Karypis, George

A2 - Zhang, Jia

PB - IEEE

CY - Los Alamitos, CA

ER -

Egebjerg NH, Hedegaard N, Kuum G, Mukkamala RR, Vatrapu R. Big Social Data Analytics in Football: Predicting Spectators and TV Ratings from Facebook Data. I Karypis G, Zhang J, red., Proceedings of the 6th IEEE International Congress on Big Data. BigData Congress 2017. Los Alamitos, CA: IEEE. 2017. s. 81-88. 8029312 https://doi.org/10.1109/BigDataCongress.2017.20