The National Librarian's Research Fellowship in Digital Scholarship

Dr Rosa Filgueira is an Assistant Professor with the School of Mathematical and Computer Sciences at Heriot-Watt University.

Her research background includes high-performance computing, data streaming, data-intensive computing, and large-scale distributed systems.  

Rosa has experience handling and mining large digital textual collections. She has worked on several digital humanities and data science projects, co-developing and co-designing text-mining applications at scale.

AI toolbox project

This project will explore new ways to unlock the full value of the National Library of Scotland's Data Foundry collections by building a new AI toolbox called 'frances' and a web user interface that allows researchers to interact with it.

It would help historians (and other users) to extract complex knowledge from digital collections in a fast and transparent manner without having to be an expert data scientist. The toolbox is named after Frances Wright (September 6, 1795 – December 13, 1852), a Scottish-born lecturer, writer, feminist, abolitionist and social reformer.

'frances' will include a set of complex text-mining queries (ready to use), and run them automatically and at scale in the background. These queries will employ the best-in-class machine learning / natural language processing algorithms, such as topic mining, sentiment analysis or text summarisation, along with the most suitable visualisation graphs for those results.

Through a visual, guided process, users will be able to:

• Select the appropriate digital Data Foundry collection(s),
• Select the analysis to be run (via text mining queries) and its configuration,
• Submit the analysis to be performed to the backend system,
• Visualise results automatically (text and graphics),
• Export and download results for further exploration.

frances will use Encyclopaedia Britannica as the core dataset and generate a practical guide for running analyses, and how to work with this, and / or any other, Data Foundry digital collection.

Dr Giles Bergel is a digital humanist, based in the Department of Engineering Science at the University of Oxford.

He has a long-standing interest in the digitisation of cheap print, and is the editor of an edition of broadside and chapbook versions of the English ballad 'The Wandering Jew's Chronicle'.

His other interests include bibliography, typography and book design; and the histories of copyright and the British book trades.

Chapbook dataset project

Giles's project involves computational analysis of the Library's 'Chapbooks printed in Scotland' dataset.

This dataset contains images, text and catalogue records of more than 3000 of the Library's collection of chapbooks printed between around 1700 and 1899. It was produced and made available through the Library's Data Foundry.

The fellowship project intends to study the illustrations within the dataset assisted by visual artificial intelligence.

Using a technique enabling features of similar images to be extracted and matched, the project plans to use the illustrations to learn more about chapbooks' origins.

It aims to uncover the type and range of imagery available to chapbook readers and to explore relationships between chapbook producers, distributors and audiences.

For more information about Giles's project, visit the Data Foundry Fellowship page.