ml4arc – Machine Learning, Deep Learning, and Natural Language Processing Applications in Archives

by Emily Higgs


On Friday, July 26, 2019, academics and practitioners met at Wilson Library at UNC Chapel Hill for “ml4arc – Machine Learning, Deep Learning, and Natural Language Processing Applications in Archives.” This meeting featured expert panels and participant-driven discussions about how we can use natural language processing – using software to understand text and its meaning – and machine learning – a branch of artificial intelligence that learns to infer patterns from data – in the archives.

The meeting was hosted by the RATOM Project (Review, Appraisal, and Triage of Mail).  The RATOM project is a partnership between the State Archives of North Carolina and the School of Information and Library Science at UNC Chapel Hill. RATOM will extend the email processing capabilities currently present in the TOMES software and BitCurator environment, developing additional modules for identifying and extracting the contents of email-containing formats, NLP tasks, and machine learning approaches. RATOM and the ml4arc meeting are generously supported by the Andrew W. Mellon Foundation.

Presentations at ml4arc were split between successful applications of machine learning and problems that could potentially be addressed by machine learning in the future. In his talk, Mike Shallcross from Indiana University identified archival workflow pain points that provide opportunities for machine learning. In particular, he sees the potential for machine learning to address issues of authenticity and integrity in digital archives, PII and risk mitigation, aggregate description, and how all these processes are (or are not) scalable and sustainable. Many of the presentations addressed these key areas and how natural language processing and machine learning can lend aid to archivists and records managers. Additionally, attendees got to see presentations and demonstrations from tools for email such as RATOM, TOMES, and ePADD. Euan Cochrane also gave a talk about the EaaSI sandbox and discussed potential relationships between software preservation and machine learning.

The meeting agenda had a strong focus on using machine learning in email archives; collecting and processing emails is a large encumbrance in many archives that can stand to benefit greatly from machine learning tools. For example, Joanne Kaczmarek from the University of Illinois presented a project processing capstone email accounts using an e-discovery and predictive coding software called Ringtail. In partnership with the Illinois State Archives, Kaczmarek used Ringtail to identify groups of “archival” and “non-archival” emails from 62 capstone accounts, and to further break down the “archival” category into “restricted” and “public.” After 3-4 weeks of tagging training data with this software, the team was able to reduce the volume of emails by 45% by excluding “non-archival” messages, and identify 1.8 million emails that met the criteria to be made available to the public. Manually, this tagging process could have easily taken over 13 years of staff time.

After the ml4arc meeting, I am excited to see the evolution of these projects and how natural language processing and machine learning can help us with our responsibilities as archivists and records managers. From entity extraction to PII identification, there are myriad possibilities for these technologies to help speed up our processes and overcome challenges.


Emily Higgs is the Digital Archivist for the Swarthmore College Peace Collection and Friends Historical Library. Before moving to Swarthmore, she was a North Carolina State University Libraries Fellow. She is also the Assistant Team Leader for the SAA ERS section blog.


Diving into Computational Archival Science

by Jane Kelly

In December 2017, the IEEE Big Data conference came to Boston, and with it came the second annual computational archival science workshop! Workshop participants were generous enough to come share their work with the local library and archives community during a one-day public unconference held at the Harvard Law School. After some sessions from Harvard librarians that touched on how they use computational methods to explore archival collections, the unconference continued with lightning talks from CAS workshop participants and discussions about what participants need to learn to engage with computational archival science in the future.

So, what is computational archival science? It is defined by CAS scholars as:

“An interdisciplinary field concerned with the application of computational methods and resources to large-scale records/archives processing, analysis, storage, long-term preservation, and access, with aim of improving efficiency, productivity and precision in support of appraisal, arrangement and description, preservation and access decisions, and engaging and undertaking research with archival material.”

Lightning round (and they really did strike like a dozen 90-second bolts of lightning, I promise!) talks from CAS workshop participants ranged from computational curation of digitized records to blockchain to topic modeling for born-digital collections. Following a voting session, participants broke into two rounds of large group discussions to dig deeper into lightning round topics. These discussions considered natural language processing, computational curation of cultural heritage archives, blockchain, and computational finding aids. Slides from lightning round presenters and community notes can be found on the CAS Unconference website.

Lightning round talks. (Image credit)

 

What did we learn? (What questions do we have now?)

Beyond learning a bit about specific projects that leverage computational methods to explore archival material, we discussed some of the challenges that archivists may bump up against when they want to engage with this work. More questions were raised than answered, but the questions can help us build a solid foundation for future study.

First, and for some of us in attendance perhaps the most important point, is the need to familiarize ourselves with computational methods. Do we have the specific technical knowledge to understand what it really means to say we want to use topic modeling to describe digital records? If not, how can we build our skills with community support? Are our electronic records suitable for computational processes? How might these issues change the way we need to conceptualize or approach appraisal, processing, and access to electronic records?

Many conversations repeatedly turned to issues of bias, privacy, and ethical issues. How do our biases shape the tools we build and use? What skills do we need to develop in order to recognize and dismantle biases in technology?

Word cloud from the unconference created by event co-organizer Ceilyn Boyd.

 

What do we need?

The unconference was intended to provide a space to bring more voices into conversations about computational methods in archives and, more specifically, to connect those currently engaged in CAS with other library and archives practitioners. At the end of the day, we worked together to compile a list of things that we felt many of us would need to learn in order to engage with CAS.

These needs include lists of methodologies and existing tools, canonical data and/or open datasets to use in testing such tools, a robust community of practice, postmortem analysis of current/existing projects, and much more. Building a community of practice and skill development for folks without strong programming skills was identified as both particularly important and especially challenging.

Be sure to check out some of the lightning round slides and community notes to learn more about CAS as a field as well as specific projects!

Interested in connecting with the CAS community? Join the CAS Google Group at: computational-archival-science@googlegroups.com!

The Harvard CAS unconference was planned and administered by Ceilyn Boyd, Jane Kelly, and Jessica Farrell of Harvard Library, with help from Richard Marciano and Bill Underwood from the Digital Curation Innovation Center (DCIC) at the University of Maryland’s iSchool. Many thanks to all the organizers, presenters, and participants!


Jane Kelly is the Historical & Special Collections Assistant at the Harvard Law School Library. She will complete her MSLIS from the iSchool at the University of Illinois, Urbana-Champaign in December 2018.