No description available.

Information is power in the 21st century. The on-going “information revolution” has forced states to adapt to the new world arena and to the following demand of being able to use information strategically. This study aimed to contribute to the lack of research regarding the Swedish case, and thus give a greater understanding of how Sweden tackles information warfare campaigns. The study examined two cases: The host-nation-agreement between Sweden and NATO (2016) and the Swedish military exercise, Aurora 17 (2017). The study applied a qualitative analytical method to: firstly, identify the Swedish strategic narrative by using governmental policy documents; secondly, identify the Russian intrusive narrative by looking at two cases using editorial articles of RT and Sputnik International, and finally analyze the Swedish way of handling the intrusive narrative, using an ideal type analysis method.  The study confirmed that Russia, through state-owned media, intervened and tried to undermine the Swedish strategic narrative in both cases. The study concluded that Sweden’s strategy to counter these actions is moving from a previously passive approach – to a more antagonistic approach towards the sender of the intruding narrative. However, the study also concluded that Sweden is lacking a coherent strategy to handling these kinds of threats.

This thesis explores to what extent Multinomial Naive Bayes (MNB) and Support Vector Machines (SVM) classifiers can be used to determine the polarity of news, specifically the news coverage of Sweden by the Russian state-funded news outlets RT and Sputnik. Three experiments are conducted.  In the first experiment, an MNB and an SVM classifier are trained with the Large Movie Review Dataset (Maas et al., 2011) with a varying number of samples to determine how training data size affects classifier performance.  In the second experiment, the classifiers are trained with 300 positive, negative, and neutral news articles (Agarwal et al., 2019) and tested on 95 RT and Sputnik news articles about Sweden (Bengtsson, 2019) to determine if the domain specificity of the training data outweighs its limited size.  In the third experiment, the movie-trained classifiers are put up against the domain-specific classifiers to determine if well-trained classifiers from another domain perform better than relatively untrained, domain-specific classifiers.  Four different types of feature sets (unigrams, unigrams without stop words removal, bigrams, trigrams) were used in the experiments. Some of the model parameters (TF-IDF vs. feature count and SVM’s C parameter) were optimized with 10-fold cross-validation.  Other than the superior performance of SVM, the results highlight the need for comprehensive and domain-specific training data when conducting machine learning tasks, as well as the benefits of feature engineering, and to a limited extent, the removal of stop words. Interestingly, the classifiers performed the best on the negative news articles, which made up most of the test set (and possibly of Russian news coverage of Sweden in general).