My research focusses on the automatic extraction of canonical references from publications in Classics. Such references are the standard way of citing classical texts and are found in great numbers throughout monographs, journal articles and commentaries. In chapters 1 and 2 I argue for the importance of canonical citations and for the need to capture them automatically. Their importance and function is to signal text passages that are studied and discussed, often in relation to one another as can be seen in parallel passages found in modern commentaries. Scholars in the field have long been exploiting this kind of information by manually creating indexes of cited passages, the so-called indices locorum. However, the challenge we now face is find new ways of indexing and retrieving information contained in the growing volume of digital archives and libraries. Chapters 3 and 4 look at how this problem can be tackled by translating the extraction of canonical citations into a computationally solvable problem. The approach I developed consists of treating the extraction of such citations as a problem of named entity extraction. This problem can be solved with some degree of accuracy by applying and adapting methods of Natural Language Processing. In this part of the dissertation I discuss the implementation of this approach as a working prototype and an evaluation of its performance. Once canonical references have been extracted from texts, the web of relations between documents that they create can be represented as a network. This network can then be searched, manipulated, visualised and analysed in various ways. In chapter 5 I focus specifically on how this network can be leveraged to search through bodies of secondary literature. Finally in chapter 6 I discuss how my work opens up new research perspectives in terms of visualisation, analysis and the application of such automatically extracted citation networks.