Dissertation defence (Information Systems Science): KTM Tapio Vepsäläinen
Time
30.5.2025 12.15 – 16.15
KTM Tapio Vepsäläinen defends the dissertation in Information Systems Science titled “Forecasting Future Events with Publicly Accessible Online Data: A Study on Finnish Parliamentary Elections from 2015 to 2023” at the University of Turku on 30 May 2025 at 12.15 (University of Turku, Turku School of Economics, Osuuskauppa lecture hall, Rehtorinpellonkatu 3, Turku).
The audience can participate in the defence by remote access: https://utu.zoom.us/j/61123165154
Opponent: Professor, Dr. h.c. Jörg Becker (University of Münster, Germany)
Custos: Professor Reima Suomi (University of Turku)
Doctoral Dissertation at UTUPub: https://urn.fi/URN:ISBN:978-952-02-0189-0
***
Summary of the Doctoral Dissertation:
The presented thesis is an examination of how publicly accessible digital data can be harnessed to predict the future. Numerous researchers have proposed that publicly accessible digital data, such as social media interactions, search queries, and online purchasing behaviors, provide valuable insights into human behavior and societal trends.
Focusing on electoral forecasting, the thesis presents a series of models used to forecast the results of Finnish parliamentary elections. By assessing the accuracy and limitations of these predictive models, the research aims to bridge the gap between emerging data science techniques and Information Systems research, providing valuable insights into the broader implications of the utilization of digital data in decision-making processes.
The results demonstrate the potential of using publicly available online data to forecast election outcomes. The study focuses on three election cycles (2015, 2019, and 2023). The final model integrates diverse data sources, including social media interactions, electoral history, and candidate attributes. Progressive improvements in accuracy were observed throughout the study, and the models eventually approached the precision of traditional polling methods. Although current models exhibit robust predictive capabilities, their practical applicability compared to opinion polls is limited.
The study underscores the incremental benefits of incorporating diverse data types while addressing the challenges associated with data collection and feature ion. The results suggest that there is substantial promise for future enhancements.
The audience can participate in the defence by remote access: https://utu.zoom.us/j/61123165154
Opponent: Professor, Dr. h.c. Jörg Becker (University of Münster, Germany)
Custos: Professor Reima Suomi (University of Turku)
Doctoral Dissertation at UTUPub: https://urn.fi/URN:ISBN:978-952-02-0189-0
***
Summary of the Doctoral Dissertation:
The presented thesis is an examination of how publicly accessible digital data can be harnessed to predict the future. Numerous researchers have proposed that publicly accessible digital data, such as social media interactions, search queries, and online purchasing behaviors, provide valuable insights into human behavior and societal trends.
Focusing on electoral forecasting, the thesis presents a series of models used to forecast the results of Finnish parliamentary elections. By assessing the accuracy and limitations of these predictive models, the research aims to bridge the gap between emerging data science techniques and Information Systems research, providing valuable insights into the broader implications of the utilization of digital data in decision-making processes.
The results demonstrate the potential of using publicly available online data to forecast election outcomes. The study focuses on three election cycles (2015, 2019, and 2023). The final model integrates diverse data sources, including social media interactions, electoral history, and candidate attributes. Progressive improvements in accuracy were observed throughout the study, and the models eventually approached the precision of traditional polling methods. Although current models exhibit robust predictive capabilities, their practical applicability compared to opinion polls is limited.
The study underscores the incremental benefits of incorporating diverse data types while addressing the challenges associated with data collection and feature ion. The results suggest that there is substantial promise for future enhancements.
University Communications