I’m Mike Jones.
I work with digital archives, historical data and the web, with a particular interest in making cultural heritage collections more accessible and useful.
My work includes machine learning, IIIF, Wikidata, newspaper archives and the interpretation of historical datasets.
Papers and talks
Utilising IIIF and Knowledge Bases for Searching People within Historical Newspaper and Journal Articles
Historical newspapers and journals are an important source of information for researchers studying the past. However, searching for specific people within these newspapers and journals can be a difficult task due to the large volume of articles and the lack of structured data. Traditional search methods, such as keyword search, are often not sufficient for finding information about specific individuals, especially when the information is spread across multiple articles. The proposed paper will cover the process and results of a research project to create an end-to-end Natural Language Processing (NLP) workflow which allows annotating of names, training of a machine learning model and inference of a corpus of historical data in order to disambiguate names found within a International Image Interoperability Framework (IIIF) document by linking each name to an entity within a knowledge base (Wikidata), through Named Entity Recognition (NER) and Named Entity Linking (NEL). The workflow, which was able to identify over 226 million names, allows researchers to query a database for a specific person or group of people based on a linked-data query, which will generate a collection level IIIF manifest of all articles within the National Library of Wales online Newspaper and Journals which mention that person. The paper will also demonstrate how additional IIIF content can be linked through Linked open data. This process allows researchers to easily collate articles that mentions a person or group of people and have the articles be viewable within a single interface (Universal Viewer or Mirador) and to annotate those articles. The paper will discuss the process of collecting and annotating data, training a machine learning model and limitations of named entity linking. While exploring other research possibilities such as identifying new people and creating new Wikidata articles for those people.
The People’s Collection Wales: Sustaining a National, Collaborative, Bilingual Digital Programme in Its First Decade
Casgliad y Werin Cymru - People’s Collection Wales is a collaborative and federated program developed by three of Wales’ national institutions (National Library of Wales, National Museum Wales, and the Royal Commission on the Ancient and Historical Monuments of Wales) to tell the story of a nation from the perspective of its people, and to deliver that story bilingually. This paper covers our journey over the past ten years; from the snatched conversations and seeds of ideas at MW2008 in Montreal, where we first formulated the concept of what a People’s Collection could be, through the first public iteration of the platform that we showcased at MW2011 in Philadelphia, to how we have sustained and strengthened the long-running partnership of our three national institutions. Over the past decade we have worked with various contractors, revised our system architecture, migrated platforms, and developed a program of activities that support community groups and individuals throughout Wales in order to contribute to, and benefit from, the People’s Collection Wales. Throughout this paper we will highlight the lessons we have learned, and offer insights we think might be applicable to other digital projects facing similar issues. We will also provide practical examples of how we have negotiated and overcome the challenges common to many in the sector—the expectation that we will deliver more in spite of decreasing budgets, and ensuring we remain relevant, both technologically and politically, in a rapidly evolving sector.
Capturing the story of Wales through the People's Collection
People's Collection Wales is a bilingual, collaborative and federated programme developed by national institutions of Wales to tell the (digital) story of a nation from the perspective of its people. Hear about APIs, widgets, microsites and visualisation of the data behind the website.