Learning Various Meta-Paths in Dyslexia a Heterogeneous Information Network Approach
Main Article Content
Abstract
Heterogeneous information networks (HINs) represent complex systems where nodes and edges can belong to different types, capturing the multifaceted relationships within the data. This versatility makes HINs particularly suited for integrating diverse data sources, ranging from genetic information and neuro-imaging data to educational records and linguistic patterns. In the context of identifying dyslexia, a neuro-developmental disorder affecting reading and writing abilities, HINs offer a robust framework for uncovering the intricate interplay of biological, cognitive, and environmental factors. By leveraging advanced machine learning algorithms, HINs can assimilate and analyse this diverse information to identify potential biomarkers, predict dyslexia risk, and personalize intervention strategies. This approach not only enhances early detection and diagnostic accuracy but also contributes to a deeper understanding of the underlying mechanisms of dyslexia, paving the way for more effective and targeted therapies. This research paper focuses on identifying the meta-paths of developmental dyslexia and its sub-types. Identification of these meta-paths may pave a way for better diagnosis.