This article explores how big data – or very large collections of data that are too extensive to be evaluated in any meaningful way by conventional literary analysis – and machine learning can help us to recover the cultural impact of Irish-authored texts published by London’s Minerva Press, the most prolific – and critically decried – publisher of popular fiction in Romantic-era Britain. In particular, it outlines how Named-Entity Recognition and Natural Language Processing can facilitate an analysis of intertextual references to Minerva’s Irish-authored works in the British Library’s open access Nineteenth-Century Literature Dataset, which comprises approximately 68,000 digitized volumes of text originally published between 1789 and 1900. Identifying and exploring these allusions helps to reveal the ongoing influence of Minerva texts in the long nineteenth century despite critical condemnation both then and now. Quantitative data invites qualitative exploration of authorial engagement with these publications and with the Minerva Press more generally in nineteenth-century Anglophone literature. In this, the article argues, machine learning provides a useful tool in recovering the network of textual relations fundamentally linked to Minerva’s Irish writers and indicative of their long-lasting impact – both negatively and positively construed – on literary production of the period.