Narralyzer offers a user-friendly tool for researchers interested in finding characters in texts, as well as the relationships between them.


    If you have ever wanted to automatically detect actantial roles and relations in Dutch literary texts, then Narralyzer is a tool right up your alley. Inspired by literary-linguistic and computational theory and method, Narralyzer will help you perform (semi-)automatic role analysis.


    Since the rise of Russian Formalism in the early 19th century, literary theorists have been interested in finding ways to detect actants in narratives. The recent rise of computational methods within the humanities offers new ways of tackling this issue. Within the Narralyzer project, we have aimed to develop a computation-based model for analyzing character roles in large datasets in Dutch.

    Using Narralyzer

    Narralyzer will help you quickly analyze the occurrence of and relations between entities in narrative texts, such as novels, short stories and newspaper articles in four languages (the online demo is limited to Dutch). The following aspects of narrative texts are included in the analysis:

    • NER (Named Entity Recognition)
    • Relations between Entities
    • Entity-text extraction (aura of n = 5)

    Users can either upload a .TEI-file for analysis, or copy-paste/type in the text using the Graphic User Interface. It is possible to automatically analyze an entire text, but also to compare character-occurrence and –relations between, for example chapters/parts in a novel. Narralyzer also provides graphical output of the characters and relations, based on Graphviz.

    The Narralyzer logo uses a CC0 image, available on Wikipedia.


    When using this tool we request you cite it as follows:

    Wildschut, P., Faber, W.J. (2017) Narralyzer. KB Lab: The Hague.

    Visualisation of Blauwe Maandagen by Arnon Grunberg done with Narralyzer

    Visualisation of Blauwe Maandagen by Arnon Grunberg done with Narralyzer