Projektet Mining textual data for simplified reading har tilldelats 5 miljoner kronor från Markus och Amalia Wallenbergs minnesfond. Projektet är treårigt och leds av professor Arne Jönsson vid Linköpings universitet, med projektdeltagare vid Språkbanken.
Projektsammanfattning på engelska:
If students can find texts that are adapted to their reading abilities, the information becomes more accessible, and easier for them to transform to knowledge. Students with reading problems get easier texts whereas students with no reading problems get more advanced texts. Individually adapted simplifications of the texts will further support the students' learning, and stimulate their interest in reading, increase their actual reading efforts and facilitate source critique, as they understand the texts better.
In the project we will develop sophisticated measures for assessing a student’s reading ability and a tool for the student and teacher to create a profile of this ability. We will also investigate how these measures can be transformed to values on known criteria like vocabulary, grammatical fluency and so forth, and how these can be used to analyse texts. Such text criteria, sensitive to content, readability and genre in combination with the profile of a student’s reading ability will form the base to individually adapted texts. Techniques and tools will be developed for selecting suitable texts, automatic summarisation of texts to a length that is adapted to the individual student and for automatic transformation to easy-to-read Swedish.