Presenting results at the Turku Book Fair

The READ project has been chosen to be the main focus of the National Archive of Finland’s (NAF) section at Turku book fair in the beginning of October. The themes for this year’s book fair are theatre, cultural impacts of climate change, and literacy. READ project’s program will focus on literacy, we are after all developing a AI program that can read handwritten text and make documents more accessible for users. The book fair program will focus on showcasing the Transkribus platform and the results of the project. We also hope to have the material i.e. a few examples of the Renovated District Court Records on display and ready to be read and browsed by visitors.

One of the main results we hope to present at the book fair is a user interface for the text recognized material. The user interface will use the HTR model to perform searches and make it possible to make full text searches in the text recognized material. This means that that you can perform searches for e.g. professions, proper nouns or legislative vocabulary in the over 800,000 digitized and HTR processed pictures of the Renovated District Court Records. Right now we are testing an indexed search engine developed by Universitat Politècnica de València with ca 130,000 pictures and the search results are good, especially for the mid and late 19th century material.

A search example with the very common surname Andersson in the Keski-Pohjanmaa-alinen region from the time period 1830-1831. In this case it gives 152 page hits with the word Andersson and four matches on the current page.

One of the challenges we are facing now is how to create a user friendly and accessible interface. With such a vast material the search results can be overwhelming. Therefore, different kind of filtering options and a visually perspicuous outline of the search results are crucial for the user. So a lot of developing work lays ahead before we can present a user friendly interface, but hopefully you can test it at the Turku book fair in October!

Hanna Strandberg