The “Collections as ML Data” checklist for machine learning and cultural heritage

Abstract

Within cultural heritage, there has been a growing and concerted effort to consider a critical sociotechnical lens when applying machine learning techniques to digital collections. Though the cultural heritage community has collectively developed an emerging body of work detailing responsible operations for machine learning in galleries, museums, archives, and libraries at the organizational level, there remains a paucity of guidelines created for researchers embarking on machine learning projects with digital collections. The manifold stakes and sensitivities involved in applying machine learning to cultural heritage underscore the importance of developing such guidelines. This article contributes to this need by formulating a detailed checklist with guiding questions and practices that can be employed while developing a machine learning project that utilizes cultural heritage data. I call the resulting checklist the “Collections as ML Data” checklist, which, when completed, can be published with the deliverables of the project. By surveying existing projects, including my own project, Newspaper Navigator, I justify the “Collections as ML Data” checklist and demonstrate how the formulated guiding questions can be employed by researchers.

Value co‐creation in cultural heritage information practices: Literature review and future agenda: An Annual Review of Information Science and Technology (ARIST) paper

“…In addition, existing studies are less likely to use machine learning methods for causal inference examination of VCC in the cultural heritage domain (Zhang et al, 2022). Some scholars believe that there is great potential for exploring digital collections in the cultural heritage domain by applying machine learning methods (Candela, 2023;Fiorucci et al, 2020), and even proposed the "Collections as ML Data" checklist for machine learning and cultural heritage (Lee, 2023). Given this, massive datasets (e.g., open data, open Linked Data, etc.…”

An automatic data quality approach to assess semantic data from cultural heritage institutions

“…For example, historical newspapers made available to the public in Chronicling America have been used to train a deep learning model to extract several classes of visual content, including headlines, photographs, illustrations, maps, comics, editorial cartoons, and advertisements (Lee et al, 2020). A ML checklist has been recently published that provides guidelines and best practices to develop a ML project based on CH data (Lee, 2022). An entity linking approach addressed to multilingual newspapers has also been recently proposed that can be applied to a broad range of text categories (Labusch & Neudecker, 2020).…”

A checklist to publish collections as data in GLAM institutions

“…Here, a checklist publication workflow was proposed including aspects such as source data management, reproducible data transformation, version control, data documentation and publication (Reyserhove et al 2020). Other initiatives include a checklist for developing a machine learning project based on cultural heritage data (Lee 2022) or a checklist for a Data Management Plan (Digital Curation Centre 2013). Regarding Collections as data at GLAM institutions, previous work has proposed a methodology to select datasets for computationally driven research applied to Spanish text corpora in order to encourage Spanish and Latin American institutions to publish machine-actionable collections (Candela et al 2021).…”