Navigating Stories in Times of Transition

Kevin Pijpers, Stefan Bastholm Andrade, Anneke M. Sools, Erik Fajoen Tjong-Kim-Sang, Gerben J. Westerhof, Malte Lüken

Research output: Contribution to conferencePosterAcademic

19 Downloads (Pure)


Introduction: In this paper, we present the project Navigating Stories in Times of Transition, a collaboration between the University of Twente and the Netherlands eScience Center. The project aims to make state-of-the-art tools for natural language processing available to researchers in the social sciences and humanities (SSH). The tools we develop advance multidisciplinary approaches to analyzing stories across different media and time. We are particularly interested in further developing digital story grammar, a computational method for narrative analysis (Andrade & Andersen, 2020). We want to show how an analysis of personal narratives collected in the times of COVID-19 pandemic with our computerized narrative tools will help researchers to chart how people make sense of the pandemic and respond to its socio-political framings in uncertain times (Murray & Sools, 2014). We will embed our tools in relevant infrastructures to make them sustainable for future use (such as CLARIAH or the SSH Open Marketplace). As a platform for integrating the tools, we use Orange, a modular data mining toolkit (Demšar et al., 2013).

Current practices: Narrative researchers already use several software programs, such as Atlas.ti and NVivo for qualitative data analysis, LIWC for automatic text analysis, and Excel, R, SPSS, and Stata for statistical analysis. In the past decade, automated natural language analysis tools have become available that could be useful for narrative analysis. Whereas several methods for natural language analysis (e.g., named entity recognition and sentiment analysis) have already been integrated into various tools used for narrative research studying textual data in English, the situation is direr for other languages. In addition, the application of more advanced approaches such as semantic role labelling and digital story grammar requires programming ability, which prevents broad application.

Goals: We aim at making digital story grammar available for other languages than English. In our initial work, we have developed crude versions of digital story grammar based on semantic role labelling for Dutch, Danish and German. Our next work has two objectives. First, inspired by narrative methodology, we want to extend our tools to advance the analysis from the level of sentences to the story level. Second, to register changes in narratives in response to societal events, we intend to enable comparative analyses across time and space with computational methods. Initially, we will focus on analyzing the dynamic relationship between narratives and societal conditions during the COVID-19 pandemic.

Concluding remarks: Our project aims at making state-of-the-art tools for natural language processing and data visualization available to SSH researchers. In our initial work, we have developed a new version of digital story grammar for the languages of Dutch, Danish and German. Our project will extend the digital toolbox for narrative analysis and thus support researchers in studying larger volumes of digital texts. All software produced by the project will be open source and we strive to balance usability and complexity when developing our tools for narrative research.
Original languageEnglish
Number of pages1
Publication statusPublished - 31 May 2022
EventDARIAH Annual Event 2022: Storytelling - Athens / Hybrid, Athens, Greece
Duration: 31 May 20223 Jun 2022


ConferenceDARIAH Annual Event 2022
Internet address


Dive into the research topics of 'Navigating Stories in Times of Transition'. Together they form a unique fingerprint.

Cite this