Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Division
Browse By Session Type
Search Tips
Personal Schedule
Sign In
Integrating the concept of knowledge representation and mental models of comprehension as well as relying on machine learning and social network analysis methodologies, this paper proposes to model political information as a complex, two-mode network. Underlying latent topics (topic modeling methodology) and political entities (named entity recognition) are extracted from unstructured text and are then connected according to their affiliation in the text. A two-mode semantic network analysis is applied to model political information as a network of interconnected political topics and actors.
This model of text as a network gives an advantage of representing textual political information as knowledge structure, providing a model of information that likely resembles how it is processed by humans. Having a structural, comprehensive model of information is crucial for better understanding of people’s psychological engagement with political processes such as political learning, acquisition and processing of political information, information retention, etc.
Petro Tolochko
Hyunjin Song, Department of Communication, U of Vienna
Hajo G. Boomgaarden, U of Vienna