We also make a quick overview of the existing computer-assisted text analysis (and, specifically, network text analysis), and text visualization methods in order to position our research in relation to the other available approaches. We then discuss several practical applications of our method ranging from automatic recovery of hidden agendas within a text and intertextual navigation graph-interfaces, to enhancing reading and writing, quick text summarization, as well as group sentiment profiling and text diagramming. The resulting data and graph representation are then used to detect the key concepts, which function as junctions for meaning circulation within a text, contextual clusters comprised of word communities (themes), as well as the most often used pathways for meaning circulation. This is done by visualizing normalized textual data as a graph and deriving the key metrics for the concepts and for the text as a whole using network analysis. In this work we propose a method and algorithm for identifying the pathways for meaning circulation within a text. The paper below was written by Dmitry Paranyushkin, Nodus Labs. A few years later we created an online open-source tool InfraNodus, which implements this approach, so if you would like to try the method on your own texts, feel free to use it on and also check out our latest paper “ Generating Insight Using Network Analysis” (Paranyushkin, 2019, The Web Conference), which presents an updated version of the approach below. 2019 Update: In this paper on text network analysis written in 2011 we proposed a method for visual graph-based topic modelling.
0 Comments
Leave a Reply. |