Abstract
Previous attempts to predict elections using social media data have attempted to emulate traditional polling by predicting the share of votes received by major parties. However, in parliamentary elections, such as those held in Canada, the party who wins the most seats in parliament forms government (which may not be the party with the most votes nationally). In this paper, a method is presented which predicts seat counts using supervised learning with Twitter, polling, and historical election data. The model was trained on the 2015 Canadian federal election and was able to accurately predict the outcome of the 2019 Canadian federal election (a Liberal minority government, despite the Conservative Party winning the plurality of votes nationally). The model was designed before the 2019 election, and predictions were made public before election day. It is demonstrated that Twitter data about local candidates is more predictive than incumbency.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
- 2.
In the UK, the pollster YouGov has used multi-level regression and post-stratification with census and historical data to construct district level estimates from polling data with some success in the 2015 UK election [12].
- 3.
- 4.
It would also be possible to fit a model for each party separately, but that is not done in this case to avoid over-fitting.
- 5.
This party only has candidates in the province of Québéc.
- 6.
References
Attarwala, A, Dimitrov, S., Obeidi, A.: How efficient is Twitter: predicting 2012 U.S. presidential elections using support vector machine via Twitter and comparing against IOWA electronic markets. In: 2017 Intelligent Systems Conference (IntelliSys). IEEE (2017)
Beauchamp, N.: Predicting and interpolating state-level polls using Twitter textual data. Am. J. Political Sci. 61(2), 490–503 (2017)
Bilal, M, et al.: Predicting elections: social media data and techniques. In: 2019 International Conference on Engineering and Emerging Technologies (ICEET). IEEE (2019)
Burnap, P., et al.: 140 characters to victory?: Using Twitter to predict the UK 2015 general election. Electoral Stud. 41, 230–233 (2016)
Fournier, P.J.: Electoral projection for 43rd Canadian general election. 338Canada.com. https://338canada.com/#2019. Accessed 15 Jan 2020
Gayo-Avello, D., Metaxas, P.T., Mustafaraj, E.: Limits of electoral predictions using Twitter. In: Proceedings of the Fifth International AAAI Conference on Weblogs and Social Media. Association for the Advancement of Artificial Intelligence (2011)
Gelman, A., King, G.: Estimating incumbency advantage without bias. Am. J. Political Sci. 34(4), 1142–1164 (1990)
Gilbert, P.D.: Brief User’s Guide: Dynamic Systems Estimation (2006). http://cran.r-project.org/web/packages/dse/vignettes/Guide.pdf
Grenier, É.: CBC Canada Poll Tracker. https://newsinteractives.cbc.ca/elections/poll-tracker/canada/. Accessed 15 Jan 2020
Han, B., Cook, P., Baldwin, T.: Text-based Twitter user geolocation prediction. J. Artif. Intell. Res. 49, 451–500 (2014)
Jackman, S.: The predictive power of uniform swing. PS Political Sci. Polit. 47(2), 317–321 (2014). https://doi.org/10.1017/S1049096514000109
Lauderdale, B.E., Bailey, D., Blumenau, J., Rivers, D.: Model-based pre-election polling for national and sub-national outcomes in the US and UK. Int. J. Forecast. 36(2), 399–413 (2019)
Li, G.: Sampling graphical networks via conditional independence coupling of Markov chains. In: Khoury, R., Drummond, C. (eds.) AI 2016. LNCS (LNAI), vol. 9673, pp. 298–303. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-34111-8_36
Metaxas, P.T., Mustafaraj, E., Gayo-Avello, D.: How (not) to predict elections. In: Privacy, Security, Risk and Trust (PASSAT) and 2011 IEEE Third Inernational Conference on Social Computing (SocialCom), pp. 165–171. IEEE (2011)
Singh, P., Sawhney, R.S., Kahlon, K.S.: Predicting the outcome of Spanish general elections 2016 using Twitter as a tool. In: Singh, D., Raman, B., Luhach, A.K., Lingras, P. (eds.) Advanced Informatics for Computing Research. CCIS, vol. 712, pp. 73–83. Springer, Singapore (2017). https://doi.org/10.1007/978-981-10-5780-9_7
Tumasjan, A., Sprenger, T.O., Sandner, P.G., Welpe, I.M.: Predicting elections with Twitter: what 140 characters reveal about political sentiment. In: ICWSM, vol. 10, pp. 178–185 (2010)
Wang, L., Gan, J. Q.: Prediction of the 2017 French election based on Twitter data analysis. In: 2017 9th Computer Science and Electronic Engineering (CEEC). IEEE (2017)
White, K.: Forecasting Canadian elections using Twitter. In: Khoury, R., Drummond, C. (eds.) AI 2016. LNCS (LNAI), vol. 9673, pp. 186–191. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-34111-8_24
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Davidson, S.M., White, K. (2020). Forecasting Seat Counts in the 2019 Canadian Federal Election Using Twitter. In: Goutte, C., Zhu, X. (eds) Advances in Artificial Intelligence. Canadian AI 2020. Lecture Notes in Computer Science(), vol 12109. Springer, Cham. https://doi.org/10.1007/978-3-030-47358-7_14
Download citation
DOI: https://doi.org/10.1007/978-3-030-47358-7_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-47357-0
Online ISBN: 978-3-030-47358-7
eBook Packages: Computer ScienceComputer Science (R0)