Transparent generosity. Introducing the impresso interface for the exploration of semantically enriched historical newspapers

Abstract

Semantically enriched historical newspapers offer a multitude of opportunities for data-driven exploration and analysis. In this paper we introduce the impresso interface which integrates several types of semantic enrichments and data visualization and thereby supports new exploratory workflows and the critical assessment of large-scale digitized source collections. The interface targets historians and integrates search, filtering, comparison, and recommendation based on automatically detected topics, linked named entities, text reuse, n-grams, image similarity, language, and OCR quality. We introduce the theoretical principles which guided interface development and reflect on the user requirements gathering process together with a case-study driven exemplification of novel workflows facilitated by the interface. We conclude with an overview of accompanying educational materials and discuss results from a user evaluation.