NATIONAL PHD IN ARTIFICIAL INTELLIGENCE

Equitable and inclusive education is a human right. Every student, irrespective of their origin or abilities, should be able to achieve their educational goals. Yet, significant differences and challenges persist, especially when it comes to children with special education needs and disabilities (SEND) diagnosed with a reading comprehension disorder. As the years go by, these children and young adults can improve their comprehension of single words, but, in most cases, the overall process of text comprehension remains often slow, painful and tiring. This has repercussions on attention capability and execution times and can lead them to drop out of school earlier than their peers. This is particularly problematic especially if we think that reading comprehension not only enables learners to attain primary and secondary education but is also a crucial life skill, serving as the foundation for many facets of daily life in the information age. Traditional clinical approaches propose education paths performed with speech therapists. One of the most efficient strategies to help children with reading comprehension disorders is the creation of a ``concept map'', a structured summary of the written text in a graph structure. Some AI models offering the possibility of automatically extracting concept maps from the text have been implemented over the years. However, different children with divergent levels of the same disorder or different disabilities approach a collection of documents with dissimilar needs and most existing methods automatically produce a canonical and fixed representation given an input text. Furthermore, an interface specifically designed for these clinical profiles is still lacking. Finally, the interactive process, fundamental to engaging in effective active learning, is completely missing. The expected output of this research project is therefore threefold. Firstly, we want to tackle the gap in the literature by implementing a new model for concept-map-based document summarization tailored to children with Reading Comprehension disorders. Secondly, we will embed it in a multi-modal and open-access Artificial-Intelligence powered interface that could help these children to make sense of the written text by enabling them to interactively create concept maps. Finally, we would evaluate it in real-world settings through a set of user studies performed by speech therapists.