Automatic detection of dyslexia based on eye movements during reading in Russian

ACL ARR 2024 December Submission46 Authors

Abstract:

Dyslexia, a common learning disability, requires an early diagnosis. However, current screening tests are very time- and resource-consuming. We present an LSTM model that aims to automatically classify dyslexia based on eye movements recorded during natural reading combined with basic demographic information and linguistic features of the fixated words. The proposed model outperforms the
state-of-the-art model and reaches the AUC of 0.93. We additionally discuss the outcomes of several ablation studies assessing which features are critical for model performance.

Paper Type: Short

Research Area: NLP Applications

Research Area Keywords: educational applications, healthcare applications, clinical NLP

Contribution Types: NLP engineering experiment

Languages Studied: Russian

Submission Number: 46