Interactive Text Mining Suite: Data Visualization for Literary Studies

In recent years, there has been growing interest in visualization methods for literary text analysis. While text mining and visualization tools have evolved into mainstream research methods in many fields (e.g. social sciences, machine learning), their application to literary studies still remains infrequent. In addition to technological challenges, the use of these tools requires a methodological shift from traditional close reading to distant reading approaches. This transition also aligns digital humanities with corpus linguistics, which still “remains obscure” and not fully embraced by digital humanists [16]. To address some of these challenges, we introduce Interactive Text Mining Suite, a user-friendly toolkit developed both for digital humanists and corpus linguists. We further demonstrate that the integration of visual analytics and corpus linguistics methods helps unveil language patterns, otherwise hidden from a human eye. Making use of both linguistically annotated data and natural language processing techniques, we are able to discern patterns of part-of-speech uses in Medieval Occitan manuscript Romance de Flamenca and its English translation. Furthermore, visual analysis not only detects stylistic differences at a word level, but also at sentential and document levels. While preserving traditional close reading techniques, this toolkit makes it possible to apply an interactive control over documents, thus allowing for a “synthesis of computational and humanistic modes of inquiry” [18].