Predicting new venture survival:  A Twitter-based machine learning approach to measuring online legitimacy

Torben  Antretter; Ivo  Blohm; Dietmar Grichnik; Joakim Wincent

doi:10.1016/j.jbvi.2018.e00109

Predicting new venture survival: A Twitter-based machine learning approach to measuring online legitimacy

Torben Antretter, Ivo Blohm, Dietmar Grichnik, Joakim Wincent

Entrepreneurship, Management and Organisation, Helsinki

Research output: Contribution to journal › Article › Scientific › peer-review

55 Citations (Scopus)

Abstract

Research indicates that interactions on social media can reveal remarkably valid predictions about future events. In this study, we show that online legitimacy as a measure of social appreciation based on Twitter content can be used to accurately predict new venture survival. Specifically, we analyze more than 187,000 tweets from 253 new ventures’ Twitter accounts using context-specific machine learning approaches. Our findings suggest that we can correctly discriminate failed ventures from surviving ventures in up to 76% of cases. With this study, we contribute to the ongoing discussion on the importance of building legitimacy online and provide an account of how to use machine learning methodologies in entrepreneurship research.

Original language	English
Article number	e00109
Peer-reviewed scientific journal	Journal of Business Venturing Insights
Volume	11
Issue number	June
Pages (from-to)	1-8
Number of pages	8
ISSN	2352-6734
DOIs	https://doi.org/10.1016/j.jbvi.2018.e00109
Publication status	Published - 01.06.2019
MoE publication type	A1 Journal article - refereed

Keywords

512 Business and Management

Access to Document

10.1016/j.jbvi.2018.e00109Licence: CC BY

1-s2.0-S2352673418301197-mainFinal published version, 352 KBLicence: CC BY

OpenUrl availability

Full text

Cite this

@article{fcfa097e389743f69b9faf50861e2b34,

title = "Predicting new venture survival: A Twitter-based machine learning approach to measuring online legitimacy",

abstract = "Research indicates that interactions on social media can reveal remarkably valid predictions about future events. In this study, we show that online legitimacy as a measure of social appreciation based on Twitter content can be used to accurately predict new venture survival. Specifically, we analyze more than 187,000 tweets from 253 new ventures{\textquoteright} Twitter accounts using context-specific machine learning approaches. Our findings suggest that we can correctly discriminate failed ventures from surviving ventures in up to 76% of cases. With this study, we contribute to the ongoing discussion on the importance of building legitimacy online and provide an account of how to use machine learning methodologies in entrepreneurship research.",

keywords = "512 Business and Management",

author = "Torben Antretter and Ivo Blohm and Dietmar Grichnik and Joakim Wincent",

year = "2019",

month = jun,

day = "1",

doi = "10.1016/j.jbvi.2018.e00109",

language = "English",

volume = "11",

pages = "1--8",

journal = "Journal of Business Venturing Insights",

issn = "2352-6734",

publisher = "Elsevier Inc",

number = "June",

}

TY - JOUR

T1 - Predicting new venture survival

T2 - A Twitter-based machine learning approach to measuring online legitimacy

AU - Antretter, Torben

AU - Blohm, Ivo

AU - Grichnik, Dietmar

AU - Wincent, Joakim

PY - 2019/6/1

Y1 - 2019/6/1

N2 - Research indicates that interactions on social media can reveal remarkably valid predictions about future events. In this study, we show that online legitimacy as a measure of social appreciation based on Twitter content can be used to accurately predict new venture survival. Specifically, we analyze more than 187,000 tweets from 253 new ventures’ Twitter accounts using context-specific machine learning approaches. Our findings suggest that we can correctly discriminate failed ventures from surviving ventures in up to 76% of cases. With this study, we contribute to the ongoing discussion on the importance of building legitimacy online and provide an account of how to use machine learning methodologies in entrepreneurship research.

AB - Research indicates that interactions on social media can reveal remarkably valid predictions about future events. In this study, we show that online legitimacy as a measure of social appreciation based on Twitter content can be used to accurately predict new venture survival. Specifically, we analyze more than 187,000 tweets from 253 new ventures’ Twitter accounts using context-specific machine learning approaches. Our findings suggest that we can correctly discriminate failed ventures from surviving ventures in up to 76% of cases. With this study, we contribute to the ongoing discussion on the importance of building legitimacy online and provide an account of how to use machine learning methodologies in entrepreneurship research.

KW - 512 Business and Management

UR - http://www.scopus.com/inward/record.url?scp=85059509620&partnerID=8YFLogxK

UR - http://www.mendeley.com/research/predicting-new-venture-survival-twitterbased-machine-learning-approach-measuring-online-legitimacy

U2 - 10.1016/j.jbvi.2018.e00109

DO - 10.1016/j.jbvi.2018.e00109

M3 - Article

SN - 2352-6734

VL - 11

SP - 1

EP - 8

JO - Journal of Business Venturing Insights

JF - Journal of Business Venturing Insights

IS - June

M1 - e00109

ER -