Linguistic Prediction in Early Childhood: A Machine Learning Analysis of Developmental Indicators
Year: 2026
Type: article
Abstract: The study explores linguistic prediction in early childhood language acquisition, examining how children anticipate upcoming words and structures to support syntactic and semantic development. Theoretical foundations draw from predictive-processing and expectation-based models that explain how prediction errors refine learning. Empirical analysis uses the Autism Prediction Dataset (Kaggle), applying machine-learning models to link predictive behavior with developmental outcomes. Logistic Regression and Random Forest achieved accuracies above 90%, confirming strong predictive indicators in early cognition. Key features influencing prediction include AQ-10 scores, age, and family history of autism. Results highlight that efficient prediction correlates with better linguistic growth and comprehension. Ethically, data remain anonymized and group-level. Overall, the research establishes prediction as a measurable, foundational process in child language development.
Author: Hamsa Ahmed
Institution: University of Anbar
Language: English
Subfield: Developmental and Educational Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W7212431828