
Elisa Barney Smith.
21 April 2023
Reading between ancient lines with machine learning
Elisa Barney Smith brings historical documents to life through her research on document analysis and natural language processing. – I love the interdisciplinary part of my work, says Elisa Barney Smith, Professor of Machine Learning at Luleå University of Technology.
New techniques can help us extract historical information from documents. Archives around the world, both public and private, are full of documents containing information that is hard to access. Machine Learning (ML) and Natural Language Processing (NLP) can help increase that access. The fusion of Machine Learning, Natural Language Processing, and history is an amazing example of what an interdisciplinary approach can look like.
Elisa Barney is working on several projects with historical documents from 5 decades to 2 millennia old.
– Right now for example, a history professor and I are investigating a reprint of a book originally written in 1125 that was a primary book used in universities and similar study during that time. By comparing the 1125 version with one from 1450, we learn a lot about how society changed over time. I like to say, that our job is to read between the lines.
Machine Learning
Machine Learning uses data to infer information about previously unseen items. This can be recognizing images: cars on roads, identification of faces, or the letters that printed text represents. Natural Language Processing focuses on the interaction between computers and human language. It involves developing algorithms and models that enable computers to understand, interpret, and generate human language. NLP has many applications, for example language translation, speech recognition, text summarization, and chatbots.
Elisa Barney Smith has backgrounds in both Computer Science and Electrical Engineering. She has focused on the image aspect of documents. She uses programming of algorithms to solve not only conventional, but unconventional problems. When she started her academic career as a PhD student, she was trying to improve the quality and ability of machines to read images of documents produced by printers, photocopy and fax machines.
– As time passed, printers and scanners got better and better. So, instead I started to look for other ways of applying my research. That’s where the historical documents came in.
Recently, she left Boise State University, where she’s still a Professor Emerita, to move to Luleå and join Machine Learning research group.
With what will you contribute to the research group in Machine Learning?
– I bring many years of experience. I have worked in numerous different fields, from electronics to medical imaging. I appreciate the interdisciplinary nature of the work being done at Luleå University of Technology. Many groups on campus are doing interesting work. I also look forward to becoming more involved in research carried out in collaboration with industries. I’ve been working in the USA and to work closely with industries is not as common over there.
Distinguished Visitor
Since arriving at Luleå University of Technology, Elisa Barney Smith is trying to build up a new network, in Luleå, in Sweden and in Europe. She has already interacted with several other groups and started to share information about her research. In January, she was appointed 2023 Distinguished Visitor by IEEE Computer Society, the largest global community of computer scientists and engineers. She is one out of 16 selected this year. As a Distinguished Visitor, the global IEEE will financially support local units so she can travel to give lectures about her research to professors, students and working professionals all over the world.
– It is a recognition of my knowledge and my expertise in the field. I have been selected by my peers – they think that I have knowledge worth sharing – and it is a great honor for me.
The research field of Machine Learning is constantly developing. When asked what future ML achievements possibly could look like, Elisa Barney Smith turns to The Jetsons, the animated tv family that lived in 2062.
– The future had features and technology that made the Jetsons’ life easier. If we chose to use AI and NPL to do good, to solve problems, and to educate ourselves, we can develop things for humanity. If we focus on technology that makes our lives easier, we could redefine what a good life is. The Jetsons wasn’t that far off after all.
Contact
Elisa Barney Smith
- Professor
- 0920-492194
- elisa.barney@ltu.se
- Elisa Barney Smith
Published:
Updated: