National Library of Scotland

The Archive of Tomorrow project (2022–2023), funded by Wellcome, focused on archiving health-related discourse on the internet. This collaborative effort across multiple institutions contributed to the UK Web Archive and explored ways to make the collection more accessible to both digital researchers and the wider public. The project also focused on the concept of treating web archives collections as data. This paper examines the documentation required to enable such collections to be utilised as data, particularly through the creation of the Datasheet for Datasets and the Data Foundry project, aiming to help make web archives machine-readable. The paper discusses the type of data and support archivists provided to researchers while navigating legal restrictions. It also highlights challenges in processing the data to ensure it is accessible to a non-technical audience, addressing the difficulties in scaling and handling sensitive health-related content. Finally, the paper outlines work on the processing pipeline required to make the material accessible to a broader audience, emphasising that providing datasets and documentation alone is insufficient. It also raises concerns about the paradox of turning initially unstructured web content into structured datasets for both archival and user interaction purposes. This project contributes to understanding how web archives can be transformed for greater accessibility and research usability.

We present a new end-to-end methodology for extracting symbols from historical maps and demonstrate an application of the method to extract details of the urban forests of Leeds and Edinburgh in the UK using Ordnance Survey maps from the 1890s. The methods presented allow tree symbols on 1:500 scale maps to be efficiently extracted, with our object detection model achieving an F1-score of 0.945. The results for each city are presented on the National Library of Scotland website and have been used to generate an estimate of 37 ± 1 tree symbols per hectare for Leeds in 1888–90 and 40 ± 1 tree symbols per hectare for Edinburgh in 1893–94. This is the first time that quantitative data has been obtained for historical urban tree counts in these two cities. The method presented can be expanded to other UK towns and cities and is a valuable tool for learning about the past, and changes to both the natural and built environment over time, aiding decisions on future tree planting. We discuss the process used to automate the generation of training data and to train a machine learning model to extract the symbols, comparing it with other possible models. This discussion provides context on how best to tackle similar problems of symbol extraction from historical maps and the issues that may arise in such automated analysis, alongside factors that must be considered when using historical maps as a data source.

The World Health Organization has a goal of universal drug susceptibility testing for patients with tuberculosis. However, molecular diagnostics to date have focused largely on first-line drugs and predicting susceptibilities in a binary manner (classifying strains as either susceptible or resistant). Here, we used a multivariable linear mixed model alongside whole genome sequencing and a quantitative microtiter plate assay to relate genomic mutations to minimum inhibitory concentration (MIC) in 15,211 Mycobacterium tuberculosis clinical isolates from 23 countries across five continents. We identified 492 unique MIC-elevating variants across 13 drugs, as well as 91 mutations likely linked to hypersensitivity. Our results advance genetics-based diagnostics for tuberculosis and serve as a curated training/testing dataset for development of drug resistance prediction algorithms.

PurposeThe purpose of this study is to offer a checklist that can be used for both creating and evaluating digital collections, which are also sometimes referred to as data sets as part of the collections as data movement, suitable for computational use.Design/methodology/approachThe checklist was built by synthesising and analysing the results of relevant research literature, articles and studies and the issues and needs obtained in an observational study. The checklist was tested and applied both as a tool for assessing a selection of digital collections made available by galleries, libraries, archives and museums (GLAM) institutions as proof of concept and as a supporting tool for creating collections as data.FindingsOver the past few years, there has been a growing interest in making available digital collections published by GLAM organisations for computational use. Based on previous work, the authors defined a methodology to build a checklist for the publication of Collections as data. The authors’ evaluation showed several examples of applications that can be useful to encourage other institutions to publish their digital collections for computational use.Originality/valueWhile some work on making available digital collections suitable for computational use exists, giving particular attention to data quality, planning and experimentation, to the best of the authors’ knowledge, none of the work to date provides an easy-to-follow and robust checklist to publish collection data sets in GLAM institutions. This checklist intends to encourage small- and medium-sized institutions to adopt the collection as data principles in daily workflows following best practices and guidelines.

GLAM organisations have been exploring the benefits of publishing their digital collections in a wide variety of forms since the 2000s. Many institutions, and in particular libraries, have applied the Semantic Web and Linked Data to their main catalogues. Recent advances in technology and innovative approaches concerning the reuse of the digital collections by means of computational access have paved the way for the creation of laboratories within GLAM organisations. In this work, we present a framework that works in several steps to transform into linked open data (LOD), assess and enrich with external repositories the data sets made available by GLAM organisations under open licences. The framework has been applied to three metadata data sets made available by the Data Foundry at the National Library of Scotland obtaining as a result a collection of LOD repositories, a data quality assessment and examples of exploration. The results of this work are publicly available and can be applied to other domains such as digital humanities and data science.

This paper discusses best practices for sharing and reusing Ground Truth in Handwritten Text Recognition infrastructures, as well as ways to reference and acknowledge contributions to the creation and enrichment of data within these systems. We discuss how one can place Ground Truth data in a repository and, subsequently, inform others through HTR-United. Furthermore, we want to we want to suggest appropriate citation methods for HTR data, models, and contributions made by volunteers. Moreover, when using digitised sources (digital facsimiles), it becomes increasingly important to distinguish between the physical object and the digital collection. These topics all relate to the proper acknowledgement of labour put into digitising, transcribing, and sharing Ground Truth HTR data. This also points to broader issues surrounding the use of machine learning in archival and library contexts, and how the community should begin to acknowledge and record both contributions and data provenance.

Funded by UCL Qatar (Mmember of Qatar Foundation) and Qatar University Library. There has also been support from the British Library Labs, and the Library of Congress Labs.