Hybrid approach in digital humanities research. A global comparative opinion mining media study
Thesis event information
Date and time of the thesis defence
Place of the thesis defence
Linnanmaa L2, Zoom link: https://oulu.zoom.us/j/64637448847?pwd=K0JLekM1d21QMUNGNjJkTHpiRUxBdz09
Topic of the dissertation
Hybrid approach in digital humanities research. A global comparative opinion mining media study
Doctoral candidate
PhD(tech)/Master of Arts Kalle Nuortimo
Faculty and unit
University of Oulu Graduate School, Faculty of Humanities, Science communication
Subject of study
Science communications
Opponent
Professor Pekka Abrahamsson, University of Jyväskylä
Custos
Professor Erkki Karvonen, University of Oulu
Hybrid approach in digital humanities research. A global comparative opinion mining media study
Digital humanities is a strongly emerging research field including various algorithm and big-data based methods, suitable for various research topics.
In this dissertation, an opinion mining approach was taken to discover the global media sentiment toward renewable and nuclear power, in order to test and compare the algorithm based approach utilizing a large dataset to traditional ones.
This thesis analyzes the media sentiment toward different near-zero emission power production technologies in order to describe differences in different technologies. The method used is opinion mining: big-data-based machine learning enhanced media analysis made by using M-Adaptive tool for media monitoring.
As a summary, sentiment towards different global power production forms has been identified in this study, complemented by more qualitative human analysis in the case of nuclear power. The differences are highlighted and also compared against literature from communication, digitalization, social media influence and opinion mining, an area belonging to the multidisciplinary field of computational linguistics.
A principal finding is that an opinion mining approach can be used to reveal the global media sentiment of different energy technologies, both in editorial publications and social media. The media sentiment in social media provides a more unfiltered point of comparison to support that of edited media, and represents one possibility to discover opposition groups communicating via SoMe.
Editorial media analysis can provide information from content created mostly in editorial style, with news frames. The media analysis reveals solar and wind power being the best-known technologies with positive media sentiment. Biomass power appears less known, yet with an inclination towards positive sentiment. Hydropower is distinctly present in editorial media with a bit lower inclination towards positive sentiment, and nuclear has clearly negative global SoMe sentiment with differences on the local level, especially in editorial sentiment. This indicates a possible presence of favourable attitude from Finnish press and corporate/PR -communication activities, further studied via qualitative manual rhetoric analysis. Ultimately, a hybrid research approach for studying this type of topics is presented.
In this dissertation, an opinion mining approach was taken to discover the global media sentiment toward renewable and nuclear power, in order to test and compare the algorithm based approach utilizing a large dataset to traditional ones.
This thesis analyzes the media sentiment toward different near-zero emission power production technologies in order to describe differences in different technologies. The method used is opinion mining: big-data-based machine learning enhanced media analysis made by using M-Adaptive tool for media monitoring.
As a summary, sentiment towards different global power production forms has been identified in this study, complemented by more qualitative human analysis in the case of nuclear power. The differences are highlighted and also compared against literature from communication, digitalization, social media influence and opinion mining, an area belonging to the multidisciplinary field of computational linguistics.
A principal finding is that an opinion mining approach can be used to reveal the global media sentiment of different energy technologies, both in editorial publications and social media. The media sentiment in social media provides a more unfiltered point of comparison to support that of edited media, and represents one possibility to discover opposition groups communicating via SoMe.
Editorial media analysis can provide information from content created mostly in editorial style, with news frames. The media analysis reveals solar and wind power being the best-known technologies with positive media sentiment. Biomass power appears less known, yet with an inclination towards positive sentiment. Hydropower is distinctly present in editorial media with a bit lower inclination towards positive sentiment, and nuclear has clearly negative global SoMe sentiment with differences on the local level, especially in editorial sentiment. This indicates a possible presence of favourable attitude from Finnish press and corporate/PR -communication activities, further studied via qualitative manual rhetoric analysis. Ultimately, a hybrid research approach for studying this type of topics is presented.
Last updated: 1.3.2023