The
National Archives of Finland’s READ project group is a mix of people from
various fields of study, e.g. history, ethnology, archive studies, information
science etc. To be able to join the project team from quite different
backgrounds and areas of interest is an asset, as it contributes to finding a
greater variety of new perspectives and utilizing a broader body of expertise.
In
addition to being multidisciplinary, projects like READ and Making a Modern
Archive exist as a result of internationality and other institutional
co-operation. This includes the close collaboration between archives and
universities. I am happy to say that, besides working closely with scholars, the
National Archives of Finland focuses on including university students to the
world of HTR as well. I myself am a major in Finnish literature, but found my
way to the READ project via my digital humanities studies and a multidisciplinary
course organized in collaboration between the University of Helsinki and NAF in
autumn 2018.
The 5-credit course called “Hand Written Materials and Automated Text Recognition” was directed mainly to students from both the Digital Humanities Minor Study Block and the Cultural Heritage Master’s Program. It provided the chance to practice working with the Transkribus program by producing manual transcriptions (i.e. GT) from old digitized Finnish war journals, and thus teaching the HTR model to automatically recognize hand-written text. This was intriguing to me, as it was the first time I got to see my literature studies and the digital humanities methods come so seamlessly together. We were able to see the results of the HTR model that our GT had trained, as well as reflect on how these kinds of AI technologies and research materials could be utilized in further research.
The
journals that we transcribed were from the Continuation War period and dated 1943-44,
with entries from officers and soldiers from various military units. The
research material in itself was fascinating, as it gave a closer insight on a
significant part of Finnish history. The wars are amazingly carefully documented – the
journals contained explicit details of the Finnish companies’ routes, contact
with the enemy, daily routines and activities, as well as the names of all
fallen, wounded and missing soldiers. To have all these materials transferred
to computer-readable form would certainly benefit the public, but also provide
new opportunities for researchers to work on. However, we barely scratched the
surface. It became clear that there are enormous amounts of valuable and
interesting hand-written cultural heritage hidden in the archives, and that there
surely is a vast need for projects like READ and MaMA that bring them out to
the open.
As I got a taste of it, I became more and more excited about this idea of modern archives and their AI-based contribution to open science. The next step was to look for a traineeship where I could continue working around these topics. That is when I heard that there would be an opening for a new trainee at NAF’s READ project. I applied immediately, and luckily, here I am today.
Laura
Kukkohovi
The 5-credit course called “Hand Written Materials and Automated Text Recognition” was directed mainly to students from both the Digital Humanities Minor Study Block and the Cultural Heritage Master’s Program. It provided the chance to practice working with the Transkribus program by producing manual transcriptions (i.e. GT) from old digitized Finnish war journals, and thus teaching the HTR model to automatically recognize hand-written text. This was intriguing to me, as it was the first time I got to see my literature studies and the digital humanities methods come so seamlessly together. We were able to see the results of the HTR model that our GT had trained, as well as reflect on how these kinds of AI technologies and research materials could be utilized in further research.
A selection of Finnish war journals from 1943. |
As I got a taste of it, I became more and more excited about this idea of modern archives and their AI-based contribution to open science. The next step was to look for a traineeship where I could continue working around these topics. That is when I heard that there would be an opening for a new trainee at NAF’s READ project. I applied immediately, and luckily, here I am today.
Comments
Post a Comment