The Data Foundry in Action: Research Projects Equipped with NLS Data Collections

Digitised images featuring the Encyclopaedia Britannica and historic industrial infrastructure
Original images courtesy of the Data Foundry, the National Library of Scotland’s open-access data collections; graphic mashup created using Canva.
Data Foundry website

It has been six months since the launch of the Data Foundry – a publicly accessible hub for the National Library of Scotland’s data collections. As of this week, the NLS is showcasing projects that have been undertaken using the data that has been made available online. The Data Foundry allows users to access the Library’s collections in machine-readable form, including digitised collections (text and images), metadata collections, map data, and organisational data. 

Researchers based at the University of Edinburgh are involved with many of the Data Foundry’s featured projects. Professor Melissa Terras (Director of the Centre for Data, Culture & Society) and Rosa Filgueira (EPCC) have been working with several Data Foundry datasets, developing text mining approaches for use with the collections to form the basis of a new text and data mining platform. One of the collections they have been working with is Encyclopaedia Britannica - early work began with n-grams which provide interesting at-a-glance visualisations. Initial topics searched for included sports, Scottish philosophers, Scottish cities and animals, showing the changing focus of the Encyclopaedia as more editions were produced. We are pleased to share that we will be supporting more text-mining projects using this data set and others over the coming months.  

Melissa Terras is also working with Dr Beatrice Alex and Dr Claire Grover, who are based in the School of Informatics, on devising automatic methods to geoparse the Gazetteers of Scotland. The Ordnance Gazetteers of Scotland is a collection of twenty volumes of the most popular descriptive historical gazetteers of 19th century Scotland, providing ‘dictionaries’ of towns, cities, castles and antiquities. The project aims to enable better quality geoparsing performance for mapping historical Scottish texts.