The research focuses on innovative analysis of historical text corpora to extract and visualize geographically related information in different periods. By exploiting the capabilities of machine learning, speech and language technologies, and computer vision, we aim to reveal the spatial dimensions that are part of digitized historical collections. With the newly developed approaches and tools in corpus linguistics, we want to analyze and visualize the geo-referenced data and, with the domain researchers, enable tracking the development of geographical references over time and offer new insights into the movement, spread, and transformation of cultural, linguistic and historical phenomena. We want the expected results to make key contributions in the fields of digital humanities, visual analytics, and data visualization and provide a deeper understanding of the geographic contexts that shape historical narratives.